gstreamer/gst/gstpad.c

3612 lines
93 KiB
C
Raw Normal View History

/* GStreamer
* Copyright (C) 1999,2000 Erik Walthinsen <omega@cse.ogi.edu>
* 2000 Wim Taymans <wtay@chello.be>
*
* gstpad.c: Pads for linking elements together
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Library General Public
* License as published by the Free Software Foundation; either
* version 2 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Library General Public License for more details.
*
* You should have received a copy of the GNU Library General Public
* License along with this library; if not, write to the
* Free Software Foundation, Inc., 59 Temple Place - Suite 330,
* Boston, MA 02111-1307, USA.
*/
#include "gst_private.h"
#include "gstpad.h"
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
#include "gstenumtypes.h"
#include "gstmarshal.h"
#include "gstutils.h"
#include "gstinfo.h"
#include "gsterror.h"
#include "gstvalue.h"
GST_DEBUG_CATEGORY_STATIC (debug_dataflow);
#define DEBUG_DATA(obj,data,notice) G_STMT_START{\
if (!data) { \
GST_CAT_DEBUG_OBJECT (debug_dataflow, obj, "NULL data value"); \
} else if (GST_IS_EVENT (data)) { \
GST_CAT_DEBUG_OBJECT (debug_dataflow, obj, "%s event %p (type %d, refcount %d)", notice, data, \
GST_EVENT_TYPE (data), GST_DATA_REFCOUNT_VALUE (data)); \
} else { \
GST_CAT_LOG_OBJECT (debug_dataflow, obj, "%s buffer %p (size %u, refcount %d)", notice, data, \
GST_BUFFER_SIZE (data), GST_BUFFER_REFCOUNT_VALUE (data)); \
} \
}G_STMT_END
#define GST_CAT_DEFAULT GST_CAT_PADS
enum
{
TEMPL_PAD_CREATED,
/* FILL ME */
TEMPL_LAST_SIGNAL
};
static GstObject *padtemplate_parent_class = NULL;
static guint gst_pad_template_signals[TEMPL_LAST_SIGNAL] = { 0 };
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
/* Pad signals and args */
enum
{
PAD_LINKED,
PAD_UNLINKED,
PAD_REQUEST_LINK,
PAD_HAVE_DATA,
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
/* FILL ME */
PAD_LAST_SIGNAL
};
enum
{
PAD_PROP_0,
PAD_PROP_CAPS,
PAD_PROP_DIRECTION,
PAD_PROP_TEMPLATE,
/* FILL ME */
};
GType _gst_pad_type = 0;
static void gst_pad_class_init (GstPadClass * klass);
static void gst_pad_init (GstPad * pad);
static void gst_pad_dispose (GObject * object);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
static void gst_pad_finalize (GObject * object);
static void gst_pad_set_property (GObject * object, guint prop_id,
const GValue * value, GParamSpec * pspec);
static void gst_pad_get_property (GObject * object, guint prop_id,
GValue * value, GParamSpec * pspec);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
static GstCaps *gst_pad_get_caps_unlocked (GstPad * pad);
static void gst_pad_set_pad_template (GstPad * pad, GstPadTemplate * templ);
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
static gboolean gst_pad_activate_default (GstPad * pad);
#ifndef GST_DISABLE_LOADSAVE
static xmlNodePtr gst_pad_save_thyself (GstObject * object, xmlNodePtr parent);
#endif
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
static GstObjectClass *pad_parent_class = NULL;
static guint gst_pad_signals[PAD_LAST_SIGNAL] = { 0 };
GType
gst_pad_get_type (void)
{
if (!_gst_pad_type) {
static const GTypeInfo pad_info = {
sizeof (GstPadClass), NULL, NULL,
(GClassInitFunc) gst_pad_class_init, NULL, NULL,
sizeof (GstPad),
0,
(GInstanceInitFunc) gst_pad_init, NULL
};
_gst_pad_type = g_type_register_static (GST_TYPE_OBJECT, "GstPad",
&pad_info, 0);
GST_DEBUG_CATEGORY_INIT (debug_dataflow, "GST_DATAFLOW",
GST_DEBUG_BOLD | GST_DEBUG_FG_GREEN, "dataflow inside pads");
}
return _gst_pad_type;
}
static gboolean
_gst_do_pass_data_accumulator (GSignalInvocationHint * ihint,
GValue * return_accu, const GValue * handler_return, gpointer dummy)
{
gboolean ret = g_value_get_boolean (handler_return);
GST_DEBUG ("accumulated %d", ret);
g_value_set_boolean (return_accu, ret);
return ret;
}
static gboolean
default_have_data (GstPad * pad, GstMiniObject * o)
{
return TRUE;
}
static void
gst_pad_class_init (GstPadClass * klass)
{
GObjectClass *gobject_class;
GstObjectClass *gstobject_class;
gobject_class = (GObjectClass *) klass;
gstobject_class = (GstObjectClass *) klass;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
pad_parent_class = g_type_class_ref (GST_TYPE_OBJECT);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
gobject_class->dispose = GST_DEBUG_FUNCPTR (gst_pad_dispose);
gobject_class->finalize = GST_DEBUG_FUNCPTR (gst_pad_finalize);
gobject_class->set_property = GST_DEBUG_FUNCPTR (gst_pad_set_property);
gobject_class->get_property = GST_DEBUG_FUNCPTR (gst_pad_get_property);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
gst_pad_signals[PAD_LINKED] =
g_signal_new ("linked", G_TYPE_FROM_CLASS (klass), G_SIGNAL_RUN_LAST,
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
G_STRUCT_OFFSET (GstPadClass, linked), NULL, NULL,
gst_marshal_VOID__OBJECT, G_TYPE_NONE, 1, GST_TYPE_PAD);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
gst_pad_signals[PAD_UNLINKED] =
g_signal_new ("unlinked", G_TYPE_FROM_CLASS (klass), G_SIGNAL_RUN_LAST,
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
G_STRUCT_OFFSET (GstPadClass, unlinked), NULL, NULL,
gst_marshal_VOID__OBJECT, G_TYPE_NONE, 1, GST_TYPE_PAD);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
gst_pad_signals[PAD_REQUEST_LINK] =
g_signal_new ("request_link", G_TYPE_FROM_CLASS (klass),
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
G_SIGNAL_RUN_LAST, G_STRUCT_OFFSET (GstPadClass, request_link), NULL,
NULL, gst_marshal_VOID__OBJECT, G_TYPE_NONE, 0);
gst_pad_signals[PAD_HAVE_DATA] =
g_signal_new ("have-data", G_TYPE_FROM_CLASS (klass),
G_SIGNAL_RUN_LAST, G_STRUCT_OFFSET (GstPadClass, have_data),
_gst_do_pass_data_accumulator,
NULL, gst_marshal_BOOLEAN__POINTER, G_TYPE_BOOLEAN, 1,
GST_TYPE_MINI_OBJECT);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_object_class_install_property (G_OBJECT_CLASS (klass), PAD_PROP_CAPS,
g_param_spec_boxed ("caps", "Caps", "The capabilities of the pad",
GST_TYPE_CAPS, G_PARAM_READABLE));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_object_class_install_property (G_OBJECT_CLASS (klass), PAD_PROP_DIRECTION,
g_param_spec_enum ("direction", "Direction", "The direction of the pad",
GST_TYPE_PAD_DIRECTION, GST_PAD_UNKNOWN,
G_PARAM_READWRITE | G_PARAM_CONSTRUCT_ONLY));
g_object_class_install_property (G_OBJECT_CLASS (klass), PAD_PROP_TEMPLATE,
g_param_spec_object ("template", "Template",
"The GstPadTemplate of this pad", GST_TYPE_PAD_TEMPLATE,
G_PARAM_READWRITE | G_PARAM_CONSTRUCT_ONLY));
#ifndef GST_DISABLE_LOADSAVE
gstobject_class->save_thyself = GST_DEBUG_FUNCPTR (gst_pad_save_thyself);
#endif
gstobject_class->path_string_separator = ".";
klass->have_data = default_have_data;
}
static void
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
gst_pad_init (GstPad * pad)
{
pad->direction = GST_PAD_UNKNOWN;
pad->peer = NULL;
WARNING: Don't grab this updated unless you're really, REALLY sure. Original commit message from CVS: WARNING: Don't grab this updated unless you're really, REALLY sure. WARNING: Wait for the next one. Whole lotta changes here, including a few random bits: examples/*/Makefile: updated to use `libtool gcc`, not just `gcc` gst/ gstbuffer.h: updated to new flag style gst.c, gstdebug.h: added new debugging for function ptrs gstpipeline.c: set type of parent_class to the class, not the object gstthread.c: ditto plugins/ cdparanoia/cdparanoia.c: added an argument type, updated some defaults cobin/spindentity.c: updated to new do/while loopfunction style mp3encode/lame/gstlame.c: argument types, whole lotta lame options tests/: various changes Now, for the big changes: Once again, the scheduling system has changed. And once again, it broke a whole bunch of things. The gist of the change is that there is now a function pointer for gst_pad_push and gst_pad_pull, instead of a hard-wired function. Well, currently they are functions, but that's for debugging purposes only, they just call the function pointer after spewing lots of DEBUG(). This changed the GstPad structure a bit, and the GstPad API as well. Where elements used to provide chain() and pull() functions, they provide chain() and get() functions. gst_pad_set_pull[region]_function has been changed to get_pad_set_get[region]_function. This means all the elements out there that used to have pull functions need to be updated. The calls to that function have been changed in the normal elements, but the names of the functions passed is still _pull[region](), which is an aesthetic issue more than anything. As for what doesn't work yet, just about anything dealing with Connections is hosed, meaning threaded stuff won't work. This will be fixed about 12 hours from now, after I've slept, etc. The simplefake.c test works in both cothreaded and chained cases, but not much else will work due to the Connection problem. Needless to say, don't grab this unless you *need* these features *now*, else wait to update this stuff until tomorrow. I'm going to sleep now.
2000-12-16 10:18:09 +00:00
pad->chainfunc = NULL;
WARNING: Don't grab this updated unless you're really, REALLY sure. Original commit message from CVS: WARNING: Don't grab this updated unless you're really, REALLY sure. WARNING: Wait for the next one. Whole lotta changes here, including a few random bits: examples/*/Makefile: updated to use `libtool gcc`, not just `gcc` gst/ gstbuffer.h: updated to new flag style gst.c, gstdebug.h: added new debugging for function ptrs gstpipeline.c: set type of parent_class to the class, not the object gstthread.c: ditto plugins/ cdparanoia/cdparanoia.c: added an argument type, updated some defaults cobin/spindentity.c: updated to new do/while loopfunction style mp3encode/lame/gstlame.c: argument types, whole lotta lame options tests/: various changes Now, for the big changes: Once again, the scheduling system has changed. And once again, it broke a whole bunch of things. The gist of the change is that there is now a function pointer for gst_pad_push and gst_pad_pull, instead of a hard-wired function. Well, currently they are functions, but that's for debugging purposes only, they just call the function pointer after spewing lots of DEBUG(). This changed the GstPad structure a bit, and the GstPad API as well. Where elements used to provide chain() and pull() functions, they provide chain() and get() functions. gst_pad_set_pull[region]_function has been changed to get_pad_set_get[region]_function. This means all the elements out there that used to have pull functions need to be updated. The calls to that function have been changed in the normal elements, but the names of the functions passed is still _pull[region](), which is an aesthetic issue more than anything. As for what doesn't work yet, just about anything dealing with Connections is hosed, meaning threaded stuff won't work. This will be fixed about 12 hours from now, after I've slept, etc. The simplefake.c test works in both cothreaded and chained cases, but not much else will work due to the Connection problem. Needless to say, don't grab this unless you *need* these features *now*, else wait to update this stuff until tomorrow. I'm going to sleep now.
2000-12-16 10:18:09 +00:00
pad->caps = NULL;
pad->linkfunc = NULL;
pad->getcapsfunc = NULL;
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
pad->activatefunc = gst_pad_activate_default;
pad->eventfunc = gst_pad_event_default;
pad->querytypefunc = gst_pad_get_query_types_default;
pad->queryfunc = gst_pad_query_default;
pad->intlinkfunc = gst_pad_get_internal_links_default;
pad->do_buffer_signals = 0;
pad->do_event_signals = 0;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_UNSET_FLUSHING (pad);
pad->preroll_lock = g_mutex_new ();
pad->preroll_cond = g_cond_new ();
pad->stream_rec_lock = g_new (GStaticRecMutex, 1);
g_static_rec_mutex_init (pad->stream_rec_lock);
pad->block_cond = g_cond_new ();
}
static void
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
gst_pad_dispose (GObject * object)
{
GstPad *pad = GST_PAD (object);
gst_pad_set_pad_template (pad, NULL);
/* FIXME, we have links to many other things like caps
* and the peer pad... */
/* No linked pad can ever be disposed.
* It has to have a parent to be linked
* and a parent would hold a reference */
/* FIXME: what about if g_object_dispose is explicitly called on the pad? Is
that legal? otherwise we could assert GST_OBJECT_PARENT (pad) == NULL as
well... */
g_assert (GST_PAD_PEER (pad) == NULL);
GST_CAT_DEBUG (GST_CAT_REFCOUNTING, "dispose %s:%s",
GST_DEBUG_PAD_NAME (pad));
/* clear the caps */
gst_caps_replace (&GST_PAD_CAPS (pad), NULL);
if (GST_IS_ELEMENT (GST_OBJECT_PARENT (pad))) {
GST_CAT_DEBUG (GST_CAT_REFCOUNTING, "removing pad from element '%s'",
GST_OBJECT_NAME (GST_OBJECT (GST_ELEMENT (GST_OBJECT_PARENT (pad)))));
gst_element_remove_pad (GST_ELEMENT (GST_OBJECT_PARENT (pad)), pad);
}
G_OBJECT_CLASS (pad_parent_class)->dispose (object);
}
static void
gst_pad_finalize (GObject * object)
{
GstPad *pad = GST_PAD (object);
if (pad->stream_rec_lock) {
g_static_rec_mutex_free (pad->stream_rec_lock);
g_free (pad->stream_rec_lock);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
pad->stream_rec_lock = NULL;
}
if (pad->preroll_lock) {
g_mutex_free (pad->preroll_lock);
g_cond_free (pad->preroll_cond);
pad->preroll_lock = NULL;
pad->preroll_cond = NULL;
}
if (pad->block_cond) {
g_cond_free (pad->block_cond);
pad->block_cond = NULL;
}
G_OBJECT_CLASS (pad_parent_class)->finalize (object);
}
static void
gst_pad_set_property (GObject * object, guint prop_id,
const GValue * value, GParamSpec * pspec)
{
g_return_if_fail (GST_IS_PAD (object));
switch (prop_id) {
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
case PAD_PROP_DIRECTION:
GST_PAD_DIRECTION (object) = g_value_get_enum (value);
break;
case PAD_PROP_TEMPLATE:
gst_pad_set_pad_template (GST_PAD_CAST (object),
(GstPadTemplate *) g_value_dup_object (value));
break;
default:
G_OBJECT_WARN_INVALID_PROPERTY_ID (object, prop_id, pspec);
break;
}
}
static void
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
gst_pad_get_property (GObject * object, guint prop_id,
GValue * value, GParamSpec * pspec)
{
g_return_if_fail (GST_IS_PAD (object));
switch (prop_id) {
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
case PAD_PROP_CAPS:
g_value_set_boxed (value, GST_PAD_CAPS (object));
break;
case PAD_PROP_DIRECTION:
g_value_set_enum (value, GST_PAD_DIRECTION (object));
break;
case PAD_PROP_TEMPLATE:
g_value_set_object (value, GST_PAD_TEMPLATE (object));
break;
default:
G_OBJECT_WARN_INVALID_PROPERTY_ID (object, prop_id, pspec);
break;
}
}
/**
* gst_pad_new:
* @name: the name of the new pad.
* @direction: the #GstPadDirection of the pad.
*
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* Creates a new pad with the given name in the given direction.
* If name is NULL, a guaranteed unique name (across all pads)
* will be assigned.
* This function makes a copy of the name so you can safely free the name.
*
* Returns: a new #GstPad, or NULL in case of an error.
*
* MT safe.
*/
GstPad *
gst_pad_new (const gchar * name, GstPadDirection direction)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
return g_object_new (GST_TYPE_PAD,
"name", name, "direction", direction, NULL);
}
/**
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* gst_pad_new_from_template:
* @templ: the pad template to use
* @name: the name of the element
*
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* Creates a new pad with the given name from the given template.
* If name is NULL, a guaranteed unique name (across all pads)
* will be assigned.
* This function makes a copy of the name so you can safely free the name.
*
* Returns: a new #GstPad, or NULL in case of an error.
*/
GstPad *
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
gst_pad_new_from_template (GstPadTemplate * templ, const gchar * name)
{
gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. Original commit message from CVS: 2004-02-24 Andy Wingo <wingo@pobox.com> * gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. * gst/schedulers/gstoptimalscheduler.c I added a mess of prototypes at the top of the file by way of documentation. Some of the operations on chains and groups were re-organized. (create_group): Added a type argument so if the group is enabled, the setup_group_scheduler knows what to do. (group_elements): Added a type argument here, too, to be passed on to create_group. (group_element_set_enabled): If an unlinked PLAYING element is added to a bin, we have to create a new group to hold the element, and this function will be called before the group is added to the chain. Thus we have a valid case for group->chain==NULL. Instead of calling chain_group_set_enabled, just set the flag on the group (the chain's status will be set when the group is added to it). (gst_opt_scheduler_state_transition, chain_group_set_enabled): Setup the group scheduler when the group is enabled, not specifically when an element goes PAUSED->PLAYING. This means PLAYING elements can be added, linked, and scheduled into a PLAYING pipeline, as was intended. (add_to_group): Don't ref the group twice. I don't know when this double-ref got in here. Removing it has the potential to cause segfaults if other parts of the scheduler are buggy. If you find that the scheduler is segfaulting for you, put in an extra ref here and see if that hacks over the underlying issue. Of course, then find out what code is unreffing a group it doesn't own... (create_group): Make the extra refcount floating, and remove it after adding the element. This means that... (unref_group): Destroy when the refcount reaches 0, not 1, like every other refcounted object in the known universe. (remove_from_group): When a group becomes empty, set it to be not active, and remove it from its chain. Don't unref it again, there's no floating reference any more. (destroy_group): We have to remove the group from the chain in remove_from_group (rather than here) to break refcounting cycles (the chain always has a ref on the group). So assert that group->chain==NULL. (ref_group_by_count): Removed, it was commented out anyway. (merge_chains): Use the remove_from_chain and add_to_chain primitives to do the reparenting, instead of rolling our own implementation. (add_to_chain): The first non-disabled group in the chain's group list will be the entry point for the chain. Because buffers can accumulate in loop elements' peer bufpens, we preferentially schedule loop groups before get groups to avoid unnecessary execution of get-based groups when the bufpens are already full. (gst_opt_scheduler_schedule_run_queue): Debug fixes. (get_group_schedule_function): Ditto. (loop_group_schedule_function): Ditto. (gst_opt_scheduler_loop_wrapper): Ditto. (gst_opt_scheduler_iterate): Ditto. I understand the opt scheduler now, yippee! * gst/gstpad.c: All throughout, added FIXMEs to look at for 0.9. (gst_pad_get_name, gst_pad_set_chain_function) (gst_pad_set_get_function, gst_pad_set_event_function) (gst_pad_set_event_mask_function, gst_pad_get_event_masks) (gst_pad_get_event_masks_default, gst_pad_set_convert_function) (gst_pad_set_query_function, gst_pad_get_query_types) (gst_pad_get_query_types_default) (gst_pad_set_internal_link_function) (gst_pad_set_formats_function, gst_pad_set_link_function) (gst_pad_set_fixate_function, gst_pad_set_getcaps_function) (gst_pad_set_bufferalloc_function, gst_pad_unlink) (gst_pad_renegotiate, gst_pad_set_parent, gst_pad_get_parent) (gst_pad_add_ghost_pad, gst_pad_proxy_getcaps) (gst_pad_proxy_pad_link, gst_pad_proxy_fixate) (gst_pad_get_pad_template_caps, gst_pad_check_compatibility) (gst_pad_get_peer, gst_pad_get_allowed_caps) (gst_pad_alloc_buffer, gst_pad_push, gst_pad_pull) (gst_pad_selectv, gst_pad_select, gst_pad_template_get_caps) (gst_pad_event_default_dispatch, gst_pad_event_default) (gst_pad_dispatcher, gst_pad_send_event, gst_pad_convert_default) (gst_pad_convert, gst_pad_query_default, gst_pad_query) (gst_pad_get_formats_default, gst_pad_get_formats): Better argument checks, and some doc fixes. (gst_pad_custom_new_from_template): Um, does anyone use these functions? Actually make a custom pad instead of a normal one. (gst_pad_try_set_caps): Transpose some checks. (gst_pad_try_set_caps_nonfixed): Same, and use a macro to check if the pad is in negotiation. (gst_pad_try_relink_filtered): Use pad_link_prepare. * gst/gstelement.c: Remove prototypes also defined in gstclock.h. * gst/gstelement.h: * gst/gstclock.h: Un-deprecate the old clocking API, as discussed on the list.
2004-02-25 13:16:12 +00:00
g_return_val_if_fail (GST_IS_PAD_TEMPLATE (templ), NULL);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
return g_object_new (GST_TYPE_PAD,
"name", name, "direction", templ->direction, "template", templ, NULL);
}
/**
* gst_pad_get_direction:
* @pad: a #GstPad to get the direction of.
*
* Gets the direction of the pad. The direction of the pad is
* decided at construction time so this function does not take
* the LOCK.
*
* Returns: the #GstPadDirection of the pad.
*
* MT safe.
*/
GstPadDirection
gst_pad_get_direction (GstPad * pad)
{
GstPadDirection result;
/* PAD_UNKNOWN is a little silly but we need some sort of
* error return value */
g_return_val_if_fail (GST_IS_PAD (pad), GST_PAD_UNKNOWN);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (pad);
result = GST_PAD_DIRECTION (pad);
GST_UNLOCK (pad);
return result;
}
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
static gboolean
gst_pad_activate_default (GstPad * pad)
{
return gst_pad_activate_push (pad, TRUE);
}
static void
pre_activate_switch (GstPad * pad, gboolean new_active)
{
if (new_active) {
return;
} else {
GST_LOCK (pad);
GST_PAD_SET_FLUSHING (pad);
/* unlock blocked pads so element can resume and stop */
GST_PAD_BLOCK_SIGNAL (pad);
GST_UNLOCK (pad);
}
}
static void
post_activate_switch (GstPad * pad, gboolean new_active)
{
if (new_active) {
GST_LOCK (pad);
GST_PAD_UNSET_FLUSHING (pad);
GST_UNLOCK (pad);
} else {
/* make streaming stop */
GST_STREAM_LOCK (pad);
GST_STREAM_UNLOCK (pad);
}
}
/**
* gst_pad_set_active:
* @pad: the #GstPad to activate or deactivate.
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
* @active: whether or not the pad should be active.
*
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
* Activates or deactivates the given pad. Must be called with the STATE_LOCK.
* Normally called from within core state change functions.
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
*
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
* If @active, makes sure the pad is active. If it is already active, either in
* push or pull mode, just return. Otherwise dispatches to the pad's activate
* function to perform the actual activation.
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
*
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
* If not @active, checks the pad's current mode and calls
* gst_pad_activate_push() or gst_pad_activate_pull(), as appropriate, with a
* FALSE argument.
*
* Returns: TRUE if the operation was successfull.
*
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
* MT safe. Must be called with STATE_LOCK.
*/
gboolean
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
gst_pad_set_active (GstPad * pad, gboolean active)
{
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
GstActivateMode old;
gboolean ret = FALSE;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (pad);
old = GST_PAD_ACTIVATE_MODE (pad);
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
GST_UNLOCK (pad);
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
if (active) {
switch (old) {
case GST_ACTIVATE_PUSH:
case GST_ACTIVATE_PULL:
ret = TRUE;
break;
case GST_ACTIVATE_NONE:
ret = (GST_PAD_ACTIVATEFUNC (pad)) (pad);
break;
}
} else {
switch (old) {
case GST_ACTIVATE_PUSH:
ret = gst_pad_activate_push (pad, FALSE);
break;
case GST_ACTIVATE_PULL:
ret = gst_pad_activate_pull (pad, FALSE);
break;
case GST_ACTIVATE_NONE:
ret = TRUE;
break;
}
}
return ret;
}
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
/**
* gst_pad_activate_pull:
* @pad: the #GstPad to activate or deactivate.
* @active: whether or not the pad should be active.
*
* Activates or deactivates the given pad in pull mode via dispatching to the
* pad's activatepullfunc. For use from within pad activation functions only.
* When called on sink pads, will first proxy the call to the peer pad, which is
* expected to activate its internally linked pads from within its activate_pull
* function.
*
* If you don't know what this is, you probably don't want to call it.
*
* Returns: TRUE if the operation was successfull.
*
* MT safe.
*/
gboolean
gst_pad_activate_pull (GstPad * pad, gboolean active)
{
GstActivateMode old;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
GST_LOCK (pad);
old = GST_PAD_ACTIVATE_MODE (pad);
GST_UNLOCK (pad);
if ((active && old == GST_ACTIVATE_PULL)
|| (!active && old == GST_ACTIVATE_NONE))
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
goto was_ok;
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
if (active) {
g_return_val_if_fail (old == GST_ACTIVATE_NONE, FALSE);
} else {
g_return_val_if_fail (old == GST_ACTIVATE_PULL, FALSE);
}
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
if (gst_pad_get_direction (pad) == GST_PAD_SINK) {
GstPad *peer = gst_pad_get_peer (pad);
if (peer) {
if (!gst_pad_activate_pull (peer, active)) {
GST_LOCK (peer);
GST_CAT_DEBUG_OBJECT (GST_CAT_PADS, pad,
"activate_pull on peer (%s:%s) failed", GST_DEBUG_PAD_NAME (peer));
GST_UNLOCK (peer);
gst_object_unref (peer);
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
goto failure;
}
}
}
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
pre_activate_switch (pad, active);
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
if (GST_PAD_ACTIVATEPULLFUNC (pad)) {
if (GST_PAD_ACTIVATEPULLFUNC (pad) (pad, active)) {
goto success;
} else {
goto failure;
}
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
} else {
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
/* can happen for sinks of passthrough elements */
goto success;
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
}
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
was_ok:
{
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
GST_CAT_DEBUG_OBJECT (GST_CAT_PADS, pad, "already %s in pull mode",
active ? "activated" : "deactivated");
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
return TRUE;
}
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
success:
{
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
GST_LOCK (pad);
GST_PAD_ACTIVATE_MODE (pad) =
active ? GST_ACTIVATE_PULL : GST_ACTIVATE_NONE;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (pad);
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
post_activate_switch (pad, active);
GST_CAT_DEBUG_OBJECT (GST_CAT_PADS, pad, "%s in pull mode",
active ? "activated" : "deactivated");
return TRUE;
}
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
failure:
{
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
GST_CAT_INFO_OBJECT (GST_CAT_PADS, pad, "failed to %s in pull mode",
active ? "activate" : "deactivate");
return FALSE;
}
}
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
/**
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
* gst_pad_activate_push:
* @pad: the #GstPad to activate or deactivate.
* @active: whether or not the pad should be active.
*
* Activates or deactivates the given pad in push mode via dispatching to the
* pad's activatepushfunc. For use from within pad activation functions only.
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
*
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
* If you don't know what this is, you probably don't want to call it.
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
*
* Returns: TRUE if the operation was successfull.
*
* MT safe.
*/
gboolean
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
gst_pad_activate_push (GstPad * pad, gboolean active)
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
{
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
GstActivateMode old;
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
GST_LOCK (pad);
old = GST_PAD_ACTIVATE_MODE (pad);
GST_UNLOCK (pad);
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
if ((active && old == GST_ACTIVATE_PUSH)
|| (!active && old == GST_ACTIVATE_NONE))
goto was_ok;
if (active) {
g_return_val_if_fail (old == GST_ACTIVATE_NONE, FALSE);
} else {
g_return_val_if_fail (old == GST_ACTIVATE_PUSH, FALSE);
}
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
pre_activate_switch (pad, active);
if (GST_PAD_ACTIVATEPUSHFUNC (pad)) {
if (GST_PAD_ACTIVATEPUSHFUNC (pad) (pad, active)) {
goto success;
} else {
goto failure;
}
} else {
/* quite ok, element relies on state change func to prepare itself */
goto success;
}
was_ok:
{
GST_CAT_DEBUG_OBJECT (GST_CAT_PADS, pad, "already %s in push mode",
active ? "activated" : "deactivated");
return TRUE;
}
success:
{
GST_LOCK (pad);
GST_PAD_ACTIVATE_MODE (pad) =
active ? GST_ACTIVATE_PUSH : GST_ACTIVATE_NONE;
GST_UNLOCK (pad);
post_activate_switch (pad, active);
GST_CAT_DEBUG_OBJECT (GST_CAT_PADS, pad, "%s in push mode",
active ? "activated" : "deactivated");
return TRUE;
}
failure:
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
{
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
GST_CAT_INFO_OBJECT (GST_CAT_PADS, pad, "failed to %s in push mode",
active ? "activate" : "deactivate");
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
return FALSE;
}
}
/**
* gst_pad_is_active:
* @pad: the #GstPad to query
*
* Query if a pad is active
*
* Returns: TRUE if the pad is active.
*
* MT safe.
*/
gboolean
gst_pad_is_active (GstPad * pad)
{
gboolean result = FALSE;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (pad);
result = GST_PAD_MODE_ACTIVATE (GST_PAD_ACTIVATE_MODE (pad));
GST_UNLOCK (pad);
return result;
}
/**
* gst_pad_set_blocked_async:
* @pad: the #GstPad to block or unblock
* @blocked: boolean indicating we should block or unblock
* @callback: #GstPadBlockCallback that will be called when the
* operation succeeds.
* @user_data: user data passed to the callback
*
* Blocks or unblocks the dataflow on a pad. The provided callback
* is called when the operation succeeds. This can take a while as
* the pad can only become blocked when real dataflow is happening.
* When the pipeline is stalled, for example in PAUSED, this can
* take an indeterminate amount of time.
* You can pass NULL as the callback to make this call block. Be
* carefull with this blocking call as it might not return for
* reasons stated above.
*
* Returns: TRUE if the pad could be blocked. This function can fail
* if wrong parameters were passed or the pad was already in the
* requested state.
*
* MT safe.
*/
gboolean
gst_pad_set_blocked_async (GstPad * pad, gboolean blocked,
GstPadBlockCallback callback, gpointer user_data)
{
gboolean was_blocked;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
was_blocked = GST_PAD_IS_BLOCKED (pad);
if (G_UNLIKELY (was_blocked == blocked))
goto had_right_state;
if (blocked) {
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad, "blocking pad %s:%s",
GST_DEBUG_PAD_NAME (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_FLAG_SET (pad, GST_PAD_BLOCKED);
pad->block_callback = callback;
pad->block_data = user_data;
if (!callback) {
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad, "waiting for block");
GST_PAD_BLOCK_WAIT (pad);
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad, "blocked");
}
} else {
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad, "unblocking pad %s:%s",
GST_DEBUG_PAD_NAME (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_FLAG_UNSET (pad, GST_PAD_BLOCKED);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
pad->block_callback = callback;
pad->block_data = user_data;
if (callback) {
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_BLOCK_SIGNAL (pad);
} else {
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_BLOCK_SIGNAL (pad);
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad, "waiting for unblock");
GST_PAD_BLOCK_WAIT (pad);
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad, "unblocked");
}
}
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (pad);
return TRUE;
had_right_state:
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad,
"pad %s:%s was in right state", GST_DEBUG_PAD_NAME (pad));
GST_UNLOCK (pad);
return FALSE;
}
}
/**
* gst_pad_set_blocked:
* @pad: the #GstPad to block or unblock
* @blocked: boolean indicating we should block or unblock
*
* Blocks or unblocks the dataflow on a pad. This function is
* a shortcut for @gst_pad_set_blocked_async() with a NULL
* callback.
*
* Returns: TRUE if the pad could be blocked. This function can fail
* wrong parameters were passed or the pad was already in the
* requested state.
*
* MT safe.
*/
gboolean
gst_pad_set_blocked (GstPad * pad, gboolean blocked)
{
return gst_pad_set_blocked_async (pad, blocked, NULL, NULL);
}
/**
* gst_pad_is_blocked:
* @pad: the #GstPad to query
*
* Checks if the pad is blocked or not. This function returns the
* last requested state of the pad. It is not certain that the pad
* is actually blocked at this point.
*
* Returns: TRUE if the pad is blocked.
*
* MT safe.
*/
gboolean
gst_pad_is_blocked (GstPad * pad)
{
gboolean result = FALSE;
g_return_val_if_fail (GST_IS_PAD (pad), result);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (pad);
result = GST_FLAG_IS_SET (pad, GST_PAD_BLOCKED);
GST_UNLOCK (pad);
return result;
}
/**
* gst_pad_set_activate_function:
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* @pad: a sink #GstPad.
* @chain: the #GstPadActivateFunction to set.
*
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
* Sets the given activate function for the pad. The activate function will
* dispatch to activate_push or activate_pull to perform the actual activation.
* Only makes sense to set on sink pads.
*
* Call this function if your sink pad can start a pull-based task.
*/
void
gst_pad_set_activate_function (GstPad * pad, GstPadActivateFunction activate)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_if_fail (GST_IS_PAD (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_ACTIVATEFUNC (pad) = activate;
GST_CAT_DEBUG (GST_CAT_PADS, "activatefunc for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (activate));
}
gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. Original commit message from CVS: 2005-06-27 Andy Wingo <wingo@pobox.com> * gst/base/gsttypefindhelper.c (gst_type_find_helper): Unref any remaining buffer. * gst/gsttrace.c (gst_alloc_trace_list_sorted): New helper, returns a sorted copy of the trace list. (gst_alloc_trace_print_live): New API, only prints traces with live objects. Sort the list. (gst_alloc_trace_print_all): Sort the list. (gst_alloc_trace_print): Align columns. * gst/elements/gstttypefindelement.c: * gst/elements/gsttee.c: * gst/base/gstbasesrc.c: * gst/base/gstbasesink.c: * gst/base/gstbasetransform.c: * gst/gstqueue.c: Adapt for pad activation changes. * gst/gstpipeline.c (gst_pipeline_init): Unref after parenting sched. (gst_pipeline_dispose): Drop ref on sched. * gst/gstpad.c (gst_pad_init): Set the default activate func. (gst_pad_activate_default): Push mode by default. (pre_activate_switch, post_activate_switch): New stubs, things to do before and after switching activation modes on pads. (gst_pad_set_active): Take a boolean and not a mode, dispatch to the pad's activate function to choose which mode to activate. Shortcut on deactivation and call the right function directly. (gst_pad_activate_pull): New API, (de)activates a pad in pull mode. (gst_pad_activate_push): New API, same for push mode. (gst_pad_set_activate_function) (gst_pad_set_activatepull_function) (gst_pad_set_activatepush_function): Setters for new API. * gst/gstminiobject.c (gst_mini_object_new, gst_mini_object_free): Trace all miniobjects. (gst_mini_object_make_writable): Unref the arg if we copy, like gst_caps_make_writable. * gst/gstmessage.c (_gst_message_initialize): No trace init. * gst/gstghostpad.c (gst_proxy_pad_do_activate) (gst_proxy_pad_do_activatepull, gst_proxy_pad_do_activatepush): Adapt for new pad API. * gst/gstevent.c (_gst_event_initialize): Don't initialize trace. * gst/gstelement.h: * gst/gstelement.c (gst_element_iterate_src_pads) (gst_element_iterate_sink_pads): New API functions. * gst/gstelement.c (iterator_fold_with_resync): New utility, should fold into gstiterator.c in some form. (gst_element_pads_activate): Simplified via use of fold and delegation of decisions to gstpad->activate. * gst/gstbus.c (gst_bus_source_finalize): Set the bus to NULL, help in debugging. * gst/gstbuffer.c (_gst_buffer_initialize): Ref the buffer type class once in init, like gstmessage. Didn't run into this issue but it seems correct. Don't initialize a trace, gstminiobject does that. * check/pipelines/simple_launch_lines.c (test_stop_from_app): New test, runs fakesrc ! fakesink, stopping on ::handoff via a message to the bus. (assert_live_count): New util function, uses alloc traces to check cleanup. * check/gst/gstghostpad.c (test_ghost_pads): More refcount checks. To be modified when unlink drops the internal pad.
2005-06-27 18:35:05 +00:00
/**
* gst_pad_set_activatepull_function:
* @pad: a sink #GstPad.
* @chain: the #GstPadActivateModeFunction to set.
*
* Sets the given activate_pull function for the pad. An activate_pull function
* prepares the element and any upstream connections for pulling. See XXX
* part-activation.txt for details.
*/
void
gst_pad_set_activatepull_function (GstPad * pad,
GstPadActivateModeFunction activatepull)
{
g_return_if_fail (GST_IS_PAD (pad));
GST_PAD_ACTIVATEPULLFUNC (pad) = activatepull;
GST_CAT_DEBUG (GST_CAT_PADS, "activatepullfunc for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (activatepull));
}
/**
* gst_pad_set_activatepush_function:
* @pad: a sink #GstPad.
* @chain: the #GstPadActivateModeFunction to set.
*
* Sets the given activate_push function for the pad. An activate_push function
* prepares the element for pushing. See XXX part-activation.txt for details.
*/
void
gst_pad_set_activatepush_function (GstPad * pad,
GstPadActivateModeFunction activatepush)
{
g_return_if_fail (GST_IS_PAD (pad));
GST_PAD_ACTIVATEPUSHFUNC (pad) = activatepush;
GST_CAT_DEBUG (GST_CAT_PADS, "activatepushfunc for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (activatepush));
}
/**
WARNING: Don't grab this updated unless you're really, REALLY sure. Original commit message from CVS: WARNING: Don't grab this updated unless you're really, REALLY sure. WARNING: Wait for the next one. Whole lotta changes here, including a few random bits: examples/*/Makefile: updated to use `libtool gcc`, not just `gcc` gst/ gstbuffer.h: updated to new flag style gst.c, gstdebug.h: added new debugging for function ptrs gstpipeline.c: set type of parent_class to the class, not the object gstthread.c: ditto plugins/ cdparanoia/cdparanoia.c: added an argument type, updated some defaults cobin/spindentity.c: updated to new do/while loopfunction style mp3encode/lame/gstlame.c: argument types, whole lotta lame options tests/: various changes Now, for the big changes: Once again, the scheduling system has changed. And once again, it broke a whole bunch of things. The gist of the change is that there is now a function pointer for gst_pad_push and gst_pad_pull, instead of a hard-wired function. Well, currently they are functions, but that's for debugging purposes only, they just call the function pointer after spewing lots of DEBUG(). This changed the GstPad structure a bit, and the GstPad API as well. Where elements used to provide chain() and pull() functions, they provide chain() and get() functions. gst_pad_set_pull[region]_function has been changed to get_pad_set_get[region]_function. This means all the elements out there that used to have pull functions need to be updated. The calls to that function have been changed in the normal elements, but the names of the functions passed is still _pull[region](), which is an aesthetic issue more than anything. As for what doesn't work yet, just about anything dealing with Connections is hosed, meaning threaded stuff won't work. This will be fixed about 12 hours from now, after I've slept, etc. The simplefake.c test works in both cothreaded and chained cases, but not much else will work due to the Connection problem. Needless to say, don't grab this unless you *need* these features *now*, else wait to update this stuff until tomorrow. I'm going to sleep now.
2000-12-16 10:18:09 +00:00
* gst_pad_set_chain_function:
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* @pad: a sink #GstPad.
* @chain: the #GstPadChainFunction to set.
*
gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. Original commit message from CVS: 2004-02-24 Andy Wingo <wingo@pobox.com> * gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. * gst/schedulers/gstoptimalscheduler.c I added a mess of prototypes at the top of the file by way of documentation. Some of the operations on chains and groups were re-organized. (create_group): Added a type argument so if the group is enabled, the setup_group_scheduler knows what to do. (group_elements): Added a type argument here, too, to be passed on to create_group. (group_element_set_enabled): If an unlinked PLAYING element is added to a bin, we have to create a new group to hold the element, and this function will be called before the group is added to the chain. Thus we have a valid case for group->chain==NULL. Instead of calling chain_group_set_enabled, just set the flag on the group (the chain's status will be set when the group is added to it). (gst_opt_scheduler_state_transition, chain_group_set_enabled): Setup the group scheduler when the group is enabled, not specifically when an element goes PAUSED->PLAYING. This means PLAYING elements can be added, linked, and scheduled into a PLAYING pipeline, as was intended. (add_to_group): Don't ref the group twice. I don't know when this double-ref got in here. Removing it has the potential to cause segfaults if other parts of the scheduler are buggy. If you find that the scheduler is segfaulting for you, put in an extra ref here and see if that hacks over the underlying issue. Of course, then find out what code is unreffing a group it doesn't own... (create_group): Make the extra refcount floating, and remove it after adding the element. This means that... (unref_group): Destroy when the refcount reaches 0, not 1, like every other refcounted object in the known universe. (remove_from_group): When a group becomes empty, set it to be not active, and remove it from its chain. Don't unref it again, there's no floating reference any more. (destroy_group): We have to remove the group from the chain in remove_from_group (rather than here) to break refcounting cycles (the chain always has a ref on the group). So assert that group->chain==NULL. (ref_group_by_count): Removed, it was commented out anyway. (merge_chains): Use the remove_from_chain and add_to_chain primitives to do the reparenting, instead of rolling our own implementation. (add_to_chain): The first non-disabled group in the chain's group list will be the entry point for the chain. Because buffers can accumulate in loop elements' peer bufpens, we preferentially schedule loop groups before get groups to avoid unnecessary execution of get-based groups when the bufpens are already full. (gst_opt_scheduler_schedule_run_queue): Debug fixes. (get_group_schedule_function): Ditto. (loop_group_schedule_function): Ditto. (gst_opt_scheduler_loop_wrapper): Ditto. (gst_opt_scheduler_iterate): Ditto. I understand the opt scheduler now, yippee! * gst/gstpad.c: All throughout, added FIXMEs to look at for 0.9. (gst_pad_get_name, gst_pad_set_chain_function) (gst_pad_set_get_function, gst_pad_set_event_function) (gst_pad_set_event_mask_function, gst_pad_get_event_masks) (gst_pad_get_event_masks_default, gst_pad_set_convert_function) (gst_pad_set_query_function, gst_pad_get_query_types) (gst_pad_get_query_types_default) (gst_pad_set_internal_link_function) (gst_pad_set_formats_function, gst_pad_set_link_function) (gst_pad_set_fixate_function, gst_pad_set_getcaps_function) (gst_pad_set_bufferalloc_function, gst_pad_unlink) (gst_pad_renegotiate, gst_pad_set_parent, gst_pad_get_parent) (gst_pad_add_ghost_pad, gst_pad_proxy_getcaps) (gst_pad_proxy_pad_link, gst_pad_proxy_fixate) (gst_pad_get_pad_template_caps, gst_pad_check_compatibility) (gst_pad_get_peer, gst_pad_get_allowed_caps) (gst_pad_alloc_buffer, gst_pad_push, gst_pad_pull) (gst_pad_selectv, gst_pad_select, gst_pad_template_get_caps) (gst_pad_event_default_dispatch, gst_pad_event_default) (gst_pad_dispatcher, gst_pad_send_event, gst_pad_convert_default) (gst_pad_convert, gst_pad_query_default, gst_pad_query) (gst_pad_get_formats_default, gst_pad_get_formats): Better argument checks, and some doc fixes. (gst_pad_custom_new_from_template): Um, does anyone use these functions? Actually make a custom pad instead of a normal one. (gst_pad_try_set_caps): Transpose some checks. (gst_pad_try_set_caps_nonfixed): Same, and use a macro to check if the pad is in negotiation. (gst_pad_try_relink_filtered): Use pad_link_prepare. * gst/gstelement.c: Remove prototypes also defined in gstclock.h. * gst/gstelement.h: * gst/gstclock.h: Un-deprecate the old clocking API, as discussed on the list.
2004-02-25 13:16:12 +00:00
* Sets the given chain function for the pad. The chain function is called to
* process a #GstBuffer input buffer.
*/
void
gst_pad_set_chain_function (GstPad * pad, GstPadChainFunction chain)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_if_fail (GST_IS_PAD (pad));
g_return_if_fail (GST_PAD_DIRECTION (pad) == GST_PAD_SINK);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_CHAINFUNC (pad) = chain;
GST_CAT_DEBUG (GST_CAT_PADS, "chainfunc for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (chain));
}
/**
* gst_pad_set_getrange_function:
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* @pad: a source #GstPad.
* @get: the #GstPadGetRangeFunction to set.
*
* Sets the given getrange function for the pad. The getrange function is called to
* produce a new #GstBuffer to start the processing pipeline. Getrange functions cannot
gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. Original commit message from CVS: 2004-02-24 Andy Wingo <wingo@pobox.com> * gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. * gst/schedulers/gstoptimalscheduler.c I added a mess of prototypes at the top of the file by way of documentation. Some of the operations on chains and groups were re-organized. (create_group): Added a type argument so if the group is enabled, the setup_group_scheduler knows what to do. (group_elements): Added a type argument here, too, to be passed on to create_group. (group_element_set_enabled): If an unlinked PLAYING element is added to a bin, we have to create a new group to hold the element, and this function will be called before the group is added to the chain. Thus we have a valid case for group->chain==NULL. Instead of calling chain_group_set_enabled, just set the flag on the group (the chain's status will be set when the group is added to it). (gst_opt_scheduler_state_transition, chain_group_set_enabled): Setup the group scheduler when the group is enabled, not specifically when an element goes PAUSED->PLAYING. This means PLAYING elements can be added, linked, and scheduled into a PLAYING pipeline, as was intended. (add_to_group): Don't ref the group twice. I don't know when this double-ref got in here. Removing it has the potential to cause segfaults if other parts of the scheduler are buggy. If you find that the scheduler is segfaulting for you, put in an extra ref here and see if that hacks over the underlying issue. Of course, then find out what code is unreffing a group it doesn't own... (create_group): Make the extra refcount floating, and remove it after adding the element. This means that... (unref_group): Destroy when the refcount reaches 0, not 1, like every other refcounted object in the known universe. (remove_from_group): When a group becomes empty, set it to be not active, and remove it from its chain. Don't unref it again, there's no floating reference any more. (destroy_group): We have to remove the group from the chain in remove_from_group (rather than here) to break refcounting cycles (the chain always has a ref on the group). So assert that group->chain==NULL. (ref_group_by_count): Removed, it was commented out anyway. (merge_chains): Use the remove_from_chain and add_to_chain primitives to do the reparenting, instead of rolling our own implementation. (add_to_chain): The first non-disabled group in the chain's group list will be the entry point for the chain. Because buffers can accumulate in loop elements' peer bufpens, we preferentially schedule loop groups before get groups to avoid unnecessary execution of get-based groups when the bufpens are already full. (gst_opt_scheduler_schedule_run_queue): Debug fixes. (get_group_schedule_function): Ditto. (loop_group_schedule_function): Ditto. (gst_opt_scheduler_loop_wrapper): Ditto. (gst_opt_scheduler_iterate): Ditto. I understand the opt scheduler now, yippee! * gst/gstpad.c: All throughout, added FIXMEs to look at for 0.9. (gst_pad_get_name, gst_pad_set_chain_function) (gst_pad_set_get_function, gst_pad_set_event_function) (gst_pad_set_event_mask_function, gst_pad_get_event_masks) (gst_pad_get_event_masks_default, gst_pad_set_convert_function) (gst_pad_set_query_function, gst_pad_get_query_types) (gst_pad_get_query_types_default) (gst_pad_set_internal_link_function) (gst_pad_set_formats_function, gst_pad_set_link_function) (gst_pad_set_fixate_function, gst_pad_set_getcaps_function) (gst_pad_set_bufferalloc_function, gst_pad_unlink) (gst_pad_renegotiate, gst_pad_set_parent, gst_pad_get_parent) (gst_pad_add_ghost_pad, gst_pad_proxy_getcaps) (gst_pad_proxy_pad_link, gst_pad_proxy_fixate) (gst_pad_get_pad_template_caps, gst_pad_check_compatibility) (gst_pad_get_peer, gst_pad_get_allowed_caps) (gst_pad_alloc_buffer, gst_pad_push, gst_pad_pull) (gst_pad_selectv, gst_pad_select, gst_pad_template_get_caps) (gst_pad_event_default_dispatch, gst_pad_event_default) (gst_pad_dispatcher, gst_pad_send_event, gst_pad_convert_default) (gst_pad_convert, gst_pad_query_default, gst_pad_query) (gst_pad_get_formats_default, gst_pad_get_formats): Better argument checks, and some doc fixes. (gst_pad_custom_new_from_template): Um, does anyone use these functions? Actually make a custom pad instead of a normal one. (gst_pad_try_set_caps): Transpose some checks. (gst_pad_try_set_caps_nonfixed): Same, and use a macro to check if the pad is in negotiation. (gst_pad_try_relink_filtered): Use pad_link_prepare. * gst/gstelement.c: Remove prototypes also defined in gstclock.h. * gst/gstelement.h: * gst/gstclock.h: Un-deprecate the old clocking API, as discussed on the list.
2004-02-25 13:16:12 +00:00
* return %NULL.
*/
void
gst_pad_set_getrange_function (GstPad * pad, GstPadGetRangeFunction get)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_if_fail (GST_IS_PAD (pad));
g_return_if_fail (GST_PAD_DIRECTION (pad) == GST_PAD_SRC);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_GETRANGEFUNC (pad) = get;
GST_CAT_DEBUG (GST_CAT_PADS, "getrangefunc for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (get));
}
gst/: More work on the generic source base class, implement seeking, query. Original commit message from CVS: * gst/base/README: * gst/base/gstbasesrc.c: (gst_basesrc_get_type), (gst_basesrc_init), (gst_basesrc_get_formats), (gst_basesrc_query), (gst_basesrc_get_event_mask), (gst_basesrc_do_seek), (gst_basesrc_event_handler), (gst_basesrc_get_range_unlocked), (gst_basesrc_check_get_range), (gst_basesrc_loop), (gst_basesrc_unlock), (gst_basesrc_get_size), (gst_basesrc_start), (gst_basesrc_stop), (gst_basesrc_activate), (gst_basesrc_change_state), (basesrc_find_peek), (basesrc_find_suggest), (gst_basesrc_type_find): * gst/base/gstbasesrc.h: * gst/elements/gstfilesrc.c: (gst_filesrc_base_init), (gst_filesrc_class_init), (gst_filesrc_init), (gst_filesrc_finalize), (gst_filesrc_set_location), (gst_filesrc_set_property), (gst_filesrc_get_property), (gst_filesrc_free_parent_mmap), (gst_filesrc_map_region), (gst_filesrc_map_small_region), (gst_filesrc_create_mmap), (gst_filesrc_create_read), (gst_filesrc_create), (gst_filesrc_get_size), (gst_filesrc_start), (gst_filesrc_stop): * gst/elements/gstfilesrc.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_set_checkgetrange_function), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: More work on the generic source base class, implement seeking, query. Make filesrc extend the base source class. Added gst_pad_set_checkgetrange_function to GstPad.
2005-04-06 17:30:48 +00:00
/**
* gst_pad_set_checkgetrange_function:
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* @pad: a source #GstPad.
gst/: More work on the generic source base class, implement seeking, query. Original commit message from CVS: * gst/base/README: * gst/base/gstbasesrc.c: (gst_basesrc_get_type), (gst_basesrc_init), (gst_basesrc_get_formats), (gst_basesrc_query), (gst_basesrc_get_event_mask), (gst_basesrc_do_seek), (gst_basesrc_event_handler), (gst_basesrc_get_range_unlocked), (gst_basesrc_check_get_range), (gst_basesrc_loop), (gst_basesrc_unlock), (gst_basesrc_get_size), (gst_basesrc_start), (gst_basesrc_stop), (gst_basesrc_activate), (gst_basesrc_change_state), (basesrc_find_peek), (basesrc_find_suggest), (gst_basesrc_type_find): * gst/base/gstbasesrc.h: * gst/elements/gstfilesrc.c: (gst_filesrc_base_init), (gst_filesrc_class_init), (gst_filesrc_init), (gst_filesrc_finalize), (gst_filesrc_set_location), (gst_filesrc_set_property), (gst_filesrc_get_property), (gst_filesrc_free_parent_mmap), (gst_filesrc_map_region), (gst_filesrc_map_small_region), (gst_filesrc_create_mmap), (gst_filesrc_create_read), (gst_filesrc_create), (gst_filesrc_get_size), (gst_filesrc_start), (gst_filesrc_stop): * gst/elements/gstfilesrc.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_set_checkgetrange_function), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: More work on the generic source base class, implement seeking, query. Make filesrc extend the base source class. Added gst_pad_set_checkgetrange_function to GstPad.
2005-04-06 17:30:48 +00:00
* @check: the #GstPadCheckGetRangeFunction to set.
*
* Sets the given checkgetrange function for the pad. Implement this function on
* a pad if you dynamically support getrange based scheduling on the pad.
*/
void
gst_pad_set_checkgetrange_function (GstPad * pad,
GstPadCheckGetRangeFunction check)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_if_fail (GST_IS_PAD (pad));
g_return_if_fail (GST_PAD_DIRECTION (pad) == GST_PAD_SRC);
gst/: More work on the generic source base class, implement seeking, query. Original commit message from CVS: * gst/base/README: * gst/base/gstbasesrc.c: (gst_basesrc_get_type), (gst_basesrc_init), (gst_basesrc_get_formats), (gst_basesrc_query), (gst_basesrc_get_event_mask), (gst_basesrc_do_seek), (gst_basesrc_event_handler), (gst_basesrc_get_range_unlocked), (gst_basesrc_check_get_range), (gst_basesrc_loop), (gst_basesrc_unlock), (gst_basesrc_get_size), (gst_basesrc_start), (gst_basesrc_stop), (gst_basesrc_activate), (gst_basesrc_change_state), (basesrc_find_peek), (basesrc_find_suggest), (gst_basesrc_type_find): * gst/base/gstbasesrc.h: * gst/elements/gstfilesrc.c: (gst_filesrc_base_init), (gst_filesrc_class_init), (gst_filesrc_init), (gst_filesrc_finalize), (gst_filesrc_set_location), (gst_filesrc_set_property), (gst_filesrc_get_property), (gst_filesrc_free_parent_mmap), (gst_filesrc_map_region), (gst_filesrc_map_small_region), (gst_filesrc_create_mmap), (gst_filesrc_create_read), (gst_filesrc_create), (gst_filesrc_get_size), (gst_filesrc_start), (gst_filesrc_stop): * gst/elements/gstfilesrc.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_set_checkgetrange_function), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: More work on the generic source base class, implement seeking, query. Make filesrc extend the base source class. Added gst_pad_set_checkgetrange_function to GstPad.
2005-04-06 17:30:48 +00:00
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_CHECKGETRANGEFUNC (pad) = check;
gst/: More work on the generic source base class, implement seeking, query. Original commit message from CVS: * gst/base/README: * gst/base/gstbasesrc.c: (gst_basesrc_get_type), (gst_basesrc_init), (gst_basesrc_get_formats), (gst_basesrc_query), (gst_basesrc_get_event_mask), (gst_basesrc_do_seek), (gst_basesrc_event_handler), (gst_basesrc_get_range_unlocked), (gst_basesrc_check_get_range), (gst_basesrc_loop), (gst_basesrc_unlock), (gst_basesrc_get_size), (gst_basesrc_start), (gst_basesrc_stop), (gst_basesrc_activate), (gst_basesrc_change_state), (basesrc_find_peek), (basesrc_find_suggest), (gst_basesrc_type_find): * gst/base/gstbasesrc.h: * gst/elements/gstfilesrc.c: (gst_filesrc_base_init), (gst_filesrc_class_init), (gst_filesrc_init), (gst_filesrc_finalize), (gst_filesrc_set_location), (gst_filesrc_set_property), (gst_filesrc_get_property), (gst_filesrc_free_parent_mmap), (gst_filesrc_map_region), (gst_filesrc_map_small_region), (gst_filesrc_create_mmap), (gst_filesrc_create_read), (gst_filesrc_create), (gst_filesrc_get_size), (gst_filesrc_start), (gst_filesrc_stop): * gst/elements/gstfilesrc.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_set_checkgetrange_function), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: More work on the generic source base class, implement seeking, query. Make filesrc extend the base source class. Added gst_pad_set_checkgetrange_function to GstPad.
2005-04-06 17:30:48 +00:00
GST_CAT_DEBUG (GST_CAT_PADS, "checkgetrangefunc for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (check));
}
/**
* gst_pad_set_event_function:
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* @pad: a source #GstPad.
* @event: the #GstPadEventFunction to set.
*
* Sets the given event handler for the pad.
*/
void
gst_pad_set_event_function (GstPad * pad, GstPadEventFunction event)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_if_fail (GST_IS_PAD (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_EVENTFUNC (pad) = event;
GST_CAT_DEBUG (GST_CAT_PADS, "eventfunc for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (event));
}
/**
* gst_pad_set_query_function:
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* @pad: a #GstPad of either direction.
* @query: the #GstPadQueryFunction to set.
*
* Set the given query function for the pad.
*/
void
gst_pad_set_query_function (GstPad * pad, GstPadQueryFunction query)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_if_fail (GST_IS_PAD (pad));
WARNING: Don't grab this updated unless you're really, REALLY sure. Original commit message from CVS: WARNING: Don't grab this updated unless you're really, REALLY sure. WARNING: Wait for the next one. Whole lotta changes here, including a few random bits: examples/*/Makefile: updated to use `libtool gcc`, not just `gcc` gst/ gstbuffer.h: updated to new flag style gst.c, gstdebug.h: added new debugging for function ptrs gstpipeline.c: set type of parent_class to the class, not the object gstthread.c: ditto plugins/ cdparanoia/cdparanoia.c: added an argument type, updated some defaults cobin/spindentity.c: updated to new do/while loopfunction style mp3encode/lame/gstlame.c: argument types, whole lotta lame options tests/: various changes Now, for the big changes: Once again, the scheduling system has changed. And once again, it broke a whole bunch of things. The gist of the change is that there is now a function pointer for gst_pad_push and gst_pad_pull, instead of a hard-wired function. Well, currently they are functions, but that's for debugging purposes only, they just call the function pointer after spewing lots of DEBUG(). This changed the GstPad structure a bit, and the GstPad API as well. Where elements used to provide chain() and pull() functions, they provide chain() and get() functions. gst_pad_set_pull[region]_function has been changed to get_pad_set_get[region]_function. This means all the elements out there that used to have pull functions need to be updated. The calls to that function have been changed in the normal elements, but the names of the functions passed is still _pull[region](), which is an aesthetic issue more than anything. As for what doesn't work yet, just about anything dealing with Connections is hosed, meaning threaded stuff won't work. This will be fixed about 12 hours from now, after I've slept, etc. The simplefake.c test works in both cothreaded and chained cases, but not much else will work due to the Connection problem. Needless to say, don't grab this unless you *need* these features *now*, else wait to update this stuff until tomorrow. I'm going to sleep now.
2000-12-16 10:18:09 +00:00
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_QUERYFUNC (pad) = query;
GST_CAT_DEBUG (GST_CAT_PADS, "queryfunc for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (query));
}
/**
* gst_pad_set_query_type_function:
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* @pad: a #GstPad of either direction.
* @type_func: the #GstPadQueryTypeFunction to set.
*
* Set the given query type function for the pad.
*/
void
gst_pad_set_query_type_function (GstPad * pad,
GstPadQueryTypeFunction type_func)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_if_fail (GST_IS_PAD (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_QUERYTYPEFUNC (pad) = type_func;
GST_CAT_DEBUG (GST_CAT_PADS, "querytypefunc for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (type_func));
}
/**
* gst_pad_get_query_types:
gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. Original commit message from CVS: 2004-02-24 Andy Wingo <wingo@pobox.com> * gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. * gst/schedulers/gstoptimalscheduler.c I added a mess of prototypes at the top of the file by way of documentation. Some of the operations on chains and groups were re-organized. (create_group): Added a type argument so if the group is enabled, the setup_group_scheduler knows what to do. (group_elements): Added a type argument here, too, to be passed on to create_group. (group_element_set_enabled): If an unlinked PLAYING element is added to a bin, we have to create a new group to hold the element, and this function will be called before the group is added to the chain. Thus we have a valid case for group->chain==NULL. Instead of calling chain_group_set_enabled, just set the flag on the group (the chain's status will be set when the group is added to it). (gst_opt_scheduler_state_transition, chain_group_set_enabled): Setup the group scheduler when the group is enabled, not specifically when an element goes PAUSED->PLAYING. This means PLAYING elements can be added, linked, and scheduled into a PLAYING pipeline, as was intended. (add_to_group): Don't ref the group twice. I don't know when this double-ref got in here. Removing it has the potential to cause segfaults if other parts of the scheduler are buggy. If you find that the scheduler is segfaulting for you, put in an extra ref here and see if that hacks over the underlying issue. Of course, then find out what code is unreffing a group it doesn't own... (create_group): Make the extra refcount floating, and remove it after adding the element. This means that... (unref_group): Destroy when the refcount reaches 0, not 1, like every other refcounted object in the known universe. (remove_from_group): When a group becomes empty, set it to be not active, and remove it from its chain. Don't unref it again, there's no floating reference any more. (destroy_group): We have to remove the group from the chain in remove_from_group (rather than here) to break refcounting cycles (the chain always has a ref on the group). So assert that group->chain==NULL. (ref_group_by_count): Removed, it was commented out anyway. (merge_chains): Use the remove_from_chain and add_to_chain primitives to do the reparenting, instead of rolling our own implementation. (add_to_chain): The first non-disabled group in the chain's group list will be the entry point for the chain. Because buffers can accumulate in loop elements' peer bufpens, we preferentially schedule loop groups before get groups to avoid unnecessary execution of get-based groups when the bufpens are already full. (gst_opt_scheduler_schedule_run_queue): Debug fixes. (get_group_schedule_function): Ditto. (loop_group_schedule_function): Ditto. (gst_opt_scheduler_loop_wrapper): Ditto. (gst_opt_scheduler_iterate): Ditto. I understand the opt scheduler now, yippee! * gst/gstpad.c: All throughout, added FIXMEs to look at for 0.9. (gst_pad_get_name, gst_pad_set_chain_function) (gst_pad_set_get_function, gst_pad_set_event_function) (gst_pad_set_event_mask_function, gst_pad_get_event_masks) (gst_pad_get_event_masks_default, gst_pad_set_convert_function) (gst_pad_set_query_function, gst_pad_get_query_types) (gst_pad_get_query_types_default) (gst_pad_set_internal_link_function) (gst_pad_set_formats_function, gst_pad_set_link_function) (gst_pad_set_fixate_function, gst_pad_set_getcaps_function) (gst_pad_set_bufferalloc_function, gst_pad_unlink) (gst_pad_renegotiate, gst_pad_set_parent, gst_pad_get_parent) (gst_pad_add_ghost_pad, gst_pad_proxy_getcaps) (gst_pad_proxy_pad_link, gst_pad_proxy_fixate) (gst_pad_get_pad_template_caps, gst_pad_check_compatibility) (gst_pad_get_peer, gst_pad_get_allowed_caps) (gst_pad_alloc_buffer, gst_pad_push, gst_pad_pull) (gst_pad_selectv, gst_pad_select, gst_pad_template_get_caps) (gst_pad_event_default_dispatch, gst_pad_event_default) (gst_pad_dispatcher, gst_pad_send_event, gst_pad_convert_default) (gst_pad_convert, gst_pad_query_default, gst_pad_query) (gst_pad_get_formats_default, gst_pad_get_formats): Better argument checks, and some doc fixes. (gst_pad_custom_new_from_template): Um, does anyone use these functions? Actually make a custom pad instead of a normal one. (gst_pad_try_set_caps): Transpose some checks. (gst_pad_try_set_caps_nonfixed): Same, and use a macro to check if the pad is in negotiation. (gst_pad_try_relink_filtered): Use pad_link_prepare. * gst/gstelement.c: Remove prototypes also defined in gstclock.h. * gst/gstelement.h: * gst/gstclock.h: Un-deprecate the old clocking API, as discussed on the list.
2004-02-25 13:16:12 +00:00
* @pad: a #GstPad.
*
* Get an array of supported queries that can be performed
* on this pad.
*
gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. Original commit message from CVS: 2004-02-10 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. (gst_pad_get_event_masks, gst_pad_get_event_masks_default) (gst_pad_get_query_types, gst_pad_get_query_types_default): Routine docs. (gst_pad_set_link_function, gst_pad_set_fixate_function) (gst_pad_set_getcaps_function): Doc from Dave's negotation random doc. (gst_pad_unlink, gst_pad_is_linked): Docs. (gst_pad_renegotiate): A brief description of capsnego. (gst_pad_try_set_caps): Document. (gst_pad_try_set_caps_nonfixed): Document. (gst_pad_can_link_filtered, gst_pad_link_filtered): Doc fixes. (gst_pad_set_parent): Deprecated (although not out of the API). (gst_pad_get_parent): Deprecated, although many plugins use this. (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): Doc that these are private and will go away in 0.9. (gst_pad_perform_negotiate): Doc. (gst_pad_link_unnegotiate): I think this is meant to be static. (gst_pad_get_negotiated_caps, gst_pad_get_pad_template_caps) (gst_pad_template_get_caps_by_name, gst_pad_check_compatibility) (gst_pad_get_peer): Doc updates. (gst_pad_caps_change_notify): Doc. (gst_pad_alloc_buffer, gst_pad_push, gst_static_pad_template_get) (gst_ghost_pad_new): Doc fixes. * gst/gstobject.c (gst_object_get_parent, gst_object_unparent) (gst_object_check_uniqueness): * gst/gstelement.c (gst_element_add_pad) (gst_element_add_ghost_pad, gst_element_remove_pad) (gst_element_remove_ghost_pad, gst_element_get_pad) (gst_element_get_static_pad, gst_element_get_pad_list) (gst_element_class_get_pad_template_list) (gst_element_class_get_pad_template): Work on the docs. (gst_element_get_pad_template_list): Uses the class method. (gst_element_get_compatible_pad_template): Docs, and consolidate some test conditions. (gst_element_get_pad_from_template): New static function. (gst_element_request_compatible_pad): Docs, and work with non-request compatible templates. (gst_element_get_compatible_pad_filtered): Docs and remove redundant checks. (gst_element_get_compatible_pad, gst_element_link_pads_filtered) (gst_element_link_filtered, gst_element_link_many) (gst_element_link, gst_element_link_pads) (gst_element_unlink_many): Docs. 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-11 13:26:04 +00:00
* Returns: a zero-terminated array of #GstQueryType.
*/
const GstQueryType *
gst_pad_get_query_types (GstPad * pad)
{
GstPadQueryTypeFunction func;
gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. Original commit message from CVS: 2004-02-24 Andy Wingo <wingo@pobox.com> * gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. * gst/schedulers/gstoptimalscheduler.c I added a mess of prototypes at the top of the file by way of documentation. Some of the operations on chains and groups were re-organized. (create_group): Added a type argument so if the group is enabled, the setup_group_scheduler knows what to do. (group_elements): Added a type argument here, too, to be passed on to create_group. (group_element_set_enabled): If an unlinked PLAYING element is added to a bin, we have to create a new group to hold the element, and this function will be called before the group is added to the chain. Thus we have a valid case for group->chain==NULL. Instead of calling chain_group_set_enabled, just set the flag on the group (the chain's status will be set when the group is added to it). (gst_opt_scheduler_state_transition, chain_group_set_enabled): Setup the group scheduler when the group is enabled, not specifically when an element goes PAUSED->PLAYING. This means PLAYING elements can be added, linked, and scheduled into a PLAYING pipeline, as was intended. (add_to_group): Don't ref the group twice. I don't know when this double-ref got in here. Removing it has the potential to cause segfaults if other parts of the scheduler are buggy. If you find that the scheduler is segfaulting for you, put in an extra ref here and see if that hacks over the underlying issue. Of course, then find out what code is unreffing a group it doesn't own... (create_group): Make the extra refcount floating, and remove it after adding the element. This means that... (unref_group): Destroy when the refcount reaches 0, not 1, like every other refcounted object in the known universe. (remove_from_group): When a group becomes empty, set it to be not active, and remove it from its chain. Don't unref it again, there's no floating reference any more. (destroy_group): We have to remove the group from the chain in remove_from_group (rather than here) to break refcounting cycles (the chain always has a ref on the group). So assert that group->chain==NULL. (ref_group_by_count): Removed, it was commented out anyway. (merge_chains): Use the remove_from_chain and add_to_chain primitives to do the reparenting, instead of rolling our own implementation. (add_to_chain): The first non-disabled group in the chain's group list will be the entry point for the chain. Because buffers can accumulate in loop elements' peer bufpens, we preferentially schedule loop groups before get groups to avoid unnecessary execution of get-based groups when the bufpens are already full. (gst_opt_scheduler_schedule_run_queue): Debug fixes. (get_group_schedule_function): Ditto. (loop_group_schedule_function): Ditto. (gst_opt_scheduler_loop_wrapper): Ditto. (gst_opt_scheduler_iterate): Ditto. I understand the opt scheduler now, yippee! * gst/gstpad.c: All throughout, added FIXMEs to look at for 0.9. (gst_pad_get_name, gst_pad_set_chain_function) (gst_pad_set_get_function, gst_pad_set_event_function) (gst_pad_set_event_mask_function, gst_pad_get_event_masks) (gst_pad_get_event_masks_default, gst_pad_set_convert_function) (gst_pad_set_query_function, gst_pad_get_query_types) (gst_pad_get_query_types_default) (gst_pad_set_internal_link_function) (gst_pad_set_formats_function, gst_pad_set_link_function) (gst_pad_set_fixate_function, gst_pad_set_getcaps_function) (gst_pad_set_bufferalloc_function, gst_pad_unlink) (gst_pad_renegotiate, gst_pad_set_parent, gst_pad_get_parent) (gst_pad_add_ghost_pad, gst_pad_proxy_getcaps) (gst_pad_proxy_pad_link, gst_pad_proxy_fixate) (gst_pad_get_pad_template_caps, gst_pad_check_compatibility) (gst_pad_get_peer, gst_pad_get_allowed_caps) (gst_pad_alloc_buffer, gst_pad_push, gst_pad_pull) (gst_pad_selectv, gst_pad_select, gst_pad_template_get_caps) (gst_pad_event_default_dispatch, gst_pad_event_default) (gst_pad_dispatcher, gst_pad_send_event, gst_pad_convert_default) (gst_pad_convert, gst_pad_query_default, gst_pad_query) (gst_pad_get_formats_default, gst_pad_get_formats): Better argument checks, and some doc fixes. (gst_pad_custom_new_from_template): Um, does anyone use these functions? Actually make a custom pad instead of a normal one. (gst_pad_try_set_caps): Transpose some checks. (gst_pad_try_set_caps_nonfixed): Same, and use a macro to check if the pad is in negotiation. (gst_pad_try_relink_filtered): Use pad_link_prepare. * gst/gstelement.c: Remove prototypes also defined in gstclock.h. * gst/gstelement.h: * gst/gstclock.h: Un-deprecate the old clocking API, as discussed on the list.
2004-02-25 13:16:12 +00:00
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY ((func = GST_PAD_QUERYTYPEFUNC (pad)) == NULL))
goto no_func;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
return func (pad);
no_func:
{
return NULL;
}
}
static gboolean
gst_pad_get_query_types_dispatcher (GstPad * pad, const GstQueryType ** data)
{
*data = gst_pad_get_query_types (pad);
return TRUE;
}
/**
* gst_pad_get_query_types_default:
gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. Original commit message from CVS: 2004-02-24 Andy Wingo <wingo@pobox.com> * gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. * gst/schedulers/gstoptimalscheduler.c I added a mess of prototypes at the top of the file by way of documentation. Some of the operations on chains and groups were re-organized. (create_group): Added a type argument so if the group is enabled, the setup_group_scheduler knows what to do. (group_elements): Added a type argument here, too, to be passed on to create_group. (group_element_set_enabled): If an unlinked PLAYING element is added to a bin, we have to create a new group to hold the element, and this function will be called before the group is added to the chain. Thus we have a valid case for group->chain==NULL. Instead of calling chain_group_set_enabled, just set the flag on the group (the chain's status will be set when the group is added to it). (gst_opt_scheduler_state_transition, chain_group_set_enabled): Setup the group scheduler when the group is enabled, not specifically when an element goes PAUSED->PLAYING. This means PLAYING elements can be added, linked, and scheduled into a PLAYING pipeline, as was intended. (add_to_group): Don't ref the group twice. I don't know when this double-ref got in here. Removing it has the potential to cause segfaults if other parts of the scheduler are buggy. If you find that the scheduler is segfaulting for you, put in an extra ref here and see if that hacks over the underlying issue. Of course, then find out what code is unreffing a group it doesn't own... (create_group): Make the extra refcount floating, and remove it after adding the element. This means that... (unref_group): Destroy when the refcount reaches 0, not 1, like every other refcounted object in the known universe. (remove_from_group): When a group becomes empty, set it to be not active, and remove it from its chain. Don't unref it again, there's no floating reference any more. (destroy_group): We have to remove the group from the chain in remove_from_group (rather than here) to break refcounting cycles (the chain always has a ref on the group). So assert that group->chain==NULL. (ref_group_by_count): Removed, it was commented out anyway. (merge_chains): Use the remove_from_chain and add_to_chain primitives to do the reparenting, instead of rolling our own implementation. (add_to_chain): The first non-disabled group in the chain's group list will be the entry point for the chain. Because buffers can accumulate in loop elements' peer bufpens, we preferentially schedule loop groups before get groups to avoid unnecessary execution of get-based groups when the bufpens are already full. (gst_opt_scheduler_schedule_run_queue): Debug fixes. (get_group_schedule_function): Ditto. (loop_group_schedule_function): Ditto. (gst_opt_scheduler_loop_wrapper): Ditto. (gst_opt_scheduler_iterate): Ditto. I understand the opt scheduler now, yippee! * gst/gstpad.c: All throughout, added FIXMEs to look at for 0.9. (gst_pad_get_name, gst_pad_set_chain_function) (gst_pad_set_get_function, gst_pad_set_event_function) (gst_pad_set_event_mask_function, gst_pad_get_event_masks) (gst_pad_get_event_masks_default, gst_pad_set_convert_function) (gst_pad_set_query_function, gst_pad_get_query_types) (gst_pad_get_query_types_default) (gst_pad_set_internal_link_function) (gst_pad_set_formats_function, gst_pad_set_link_function) (gst_pad_set_fixate_function, gst_pad_set_getcaps_function) (gst_pad_set_bufferalloc_function, gst_pad_unlink) (gst_pad_renegotiate, gst_pad_set_parent, gst_pad_get_parent) (gst_pad_add_ghost_pad, gst_pad_proxy_getcaps) (gst_pad_proxy_pad_link, gst_pad_proxy_fixate) (gst_pad_get_pad_template_caps, gst_pad_check_compatibility) (gst_pad_get_peer, gst_pad_get_allowed_caps) (gst_pad_alloc_buffer, gst_pad_push, gst_pad_pull) (gst_pad_selectv, gst_pad_select, gst_pad_template_get_caps) (gst_pad_event_default_dispatch, gst_pad_event_default) (gst_pad_dispatcher, gst_pad_send_event, gst_pad_convert_default) (gst_pad_convert, gst_pad_query_default, gst_pad_query) (gst_pad_get_formats_default, gst_pad_get_formats): Better argument checks, and some doc fixes. (gst_pad_custom_new_from_template): Um, does anyone use these functions? Actually make a custom pad instead of a normal one. (gst_pad_try_set_caps): Transpose some checks. (gst_pad_try_set_caps_nonfixed): Same, and use a macro to check if the pad is in negotiation. (gst_pad_try_relink_filtered): Use pad_link_prepare. * gst/gstelement.c: Remove prototypes also defined in gstclock.h. * gst/gstelement.h: * gst/gstclock.h: Un-deprecate the old clocking API, as discussed on the list.
2004-02-25 13:16:12 +00:00
* @pad: a #GstPad.
*
* Invoke the default dispatcher for the query types on
* the pad.
*
gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. Original commit message from CVS: 2004-02-10 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. (gst_pad_get_event_masks, gst_pad_get_event_masks_default) (gst_pad_get_query_types, gst_pad_get_query_types_default): Routine docs. (gst_pad_set_link_function, gst_pad_set_fixate_function) (gst_pad_set_getcaps_function): Doc from Dave's negotation random doc. (gst_pad_unlink, gst_pad_is_linked): Docs. (gst_pad_renegotiate): A brief description of capsnego. (gst_pad_try_set_caps): Document. (gst_pad_try_set_caps_nonfixed): Document. (gst_pad_can_link_filtered, gst_pad_link_filtered): Doc fixes. (gst_pad_set_parent): Deprecated (although not out of the API). (gst_pad_get_parent): Deprecated, although many plugins use this. (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): Doc that these are private and will go away in 0.9. (gst_pad_perform_negotiate): Doc. (gst_pad_link_unnegotiate): I think this is meant to be static. (gst_pad_get_negotiated_caps, gst_pad_get_pad_template_caps) (gst_pad_template_get_caps_by_name, gst_pad_check_compatibility) (gst_pad_get_peer): Doc updates. (gst_pad_caps_change_notify): Doc. (gst_pad_alloc_buffer, gst_pad_push, gst_static_pad_template_get) (gst_ghost_pad_new): Doc fixes. * gst/gstobject.c (gst_object_get_parent, gst_object_unparent) (gst_object_check_uniqueness): * gst/gstelement.c (gst_element_add_pad) (gst_element_add_ghost_pad, gst_element_remove_pad) (gst_element_remove_ghost_pad, gst_element_get_pad) (gst_element_get_static_pad, gst_element_get_pad_list) (gst_element_class_get_pad_template_list) (gst_element_class_get_pad_template): Work on the docs. (gst_element_get_pad_template_list): Uses the class method. (gst_element_get_compatible_pad_template): Docs, and consolidate some test conditions. (gst_element_get_pad_from_template): New static function. (gst_element_request_compatible_pad): Docs, and work with non-request compatible templates. (gst_element_get_compatible_pad_filtered): Docs and remove redundant checks. (gst_element_get_compatible_pad, gst_element_link_pads_filtered) (gst_element_link_filtered, gst_element_link_many) (gst_element_link, gst_element_link_pads) (gst_element_unlink_many): Docs. 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-11 13:26:04 +00:00
* Returns: an zero-terminated array of #GstQueryType, or NULL if none of the
* internally-linked pads has a query types function.
*/
const GstQueryType *
gst_pad_get_query_types_default (GstPad * pad)
{
GstQueryType *result = NULL;
gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. Original commit message from CVS: 2004-02-24 Andy Wingo <wingo@pobox.com> * gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. * gst/schedulers/gstoptimalscheduler.c I added a mess of prototypes at the top of the file by way of documentation. Some of the operations on chains and groups were re-organized. (create_group): Added a type argument so if the group is enabled, the setup_group_scheduler knows what to do. (group_elements): Added a type argument here, too, to be passed on to create_group. (group_element_set_enabled): If an unlinked PLAYING element is added to a bin, we have to create a new group to hold the element, and this function will be called before the group is added to the chain. Thus we have a valid case for group->chain==NULL. Instead of calling chain_group_set_enabled, just set the flag on the group (the chain's status will be set when the group is added to it). (gst_opt_scheduler_state_transition, chain_group_set_enabled): Setup the group scheduler when the group is enabled, not specifically when an element goes PAUSED->PLAYING. This means PLAYING elements can be added, linked, and scheduled into a PLAYING pipeline, as was intended. (add_to_group): Don't ref the group twice. I don't know when this double-ref got in here. Removing it has the potential to cause segfaults if other parts of the scheduler are buggy. If you find that the scheduler is segfaulting for you, put in an extra ref here and see if that hacks over the underlying issue. Of course, then find out what code is unreffing a group it doesn't own... (create_group): Make the extra refcount floating, and remove it after adding the element. This means that... (unref_group): Destroy when the refcount reaches 0, not 1, like every other refcounted object in the known universe. (remove_from_group): When a group becomes empty, set it to be not active, and remove it from its chain. Don't unref it again, there's no floating reference any more. (destroy_group): We have to remove the group from the chain in remove_from_group (rather than here) to break refcounting cycles (the chain always has a ref on the group). So assert that group->chain==NULL. (ref_group_by_count): Removed, it was commented out anyway. (merge_chains): Use the remove_from_chain and add_to_chain primitives to do the reparenting, instead of rolling our own implementation. (add_to_chain): The first non-disabled group in the chain's group list will be the entry point for the chain. Because buffers can accumulate in loop elements' peer bufpens, we preferentially schedule loop groups before get groups to avoid unnecessary execution of get-based groups when the bufpens are already full. (gst_opt_scheduler_schedule_run_queue): Debug fixes. (get_group_schedule_function): Ditto. (loop_group_schedule_function): Ditto. (gst_opt_scheduler_loop_wrapper): Ditto. (gst_opt_scheduler_iterate): Ditto. I understand the opt scheduler now, yippee! * gst/gstpad.c: All throughout, added FIXMEs to look at for 0.9. (gst_pad_get_name, gst_pad_set_chain_function) (gst_pad_set_get_function, gst_pad_set_event_function) (gst_pad_set_event_mask_function, gst_pad_get_event_masks) (gst_pad_get_event_masks_default, gst_pad_set_convert_function) (gst_pad_set_query_function, gst_pad_get_query_types) (gst_pad_get_query_types_default) (gst_pad_set_internal_link_function) (gst_pad_set_formats_function, gst_pad_set_link_function) (gst_pad_set_fixate_function, gst_pad_set_getcaps_function) (gst_pad_set_bufferalloc_function, gst_pad_unlink) (gst_pad_renegotiate, gst_pad_set_parent, gst_pad_get_parent) (gst_pad_add_ghost_pad, gst_pad_proxy_getcaps) (gst_pad_proxy_pad_link, gst_pad_proxy_fixate) (gst_pad_get_pad_template_caps, gst_pad_check_compatibility) (gst_pad_get_peer, gst_pad_get_allowed_caps) (gst_pad_alloc_buffer, gst_pad_push, gst_pad_pull) (gst_pad_selectv, gst_pad_select, gst_pad_template_get_caps) (gst_pad_event_default_dispatch, gst_pad_event_default) (gst_pad_dispatcher, gst_pad_send_event, gst_pad_convert_default) (gst_pad_convert, gst_pad_query_default, gst_pad_query) (gst_pad_get_formats_default, gst_pad_get_formats): Better argument checks, and some doc fixes. (gst_pad_custom_new_from_template): Um, does anyone use these functions? Actually make a custom pad instead of a normal one. (gst_pad_try_set_caps): Transpose some checks. (gst_pad_try_set_caps_nonfixed): Same, and use a macro to check if the pad is in negotiation. (gst_pad_try_relink_filtered): Use pad_link_prepare. * gst/gstelement.c: Remove prototypes also defined in gstclock.h. * gst/gstelement.h: * gst/gstclock.h: Un-deprecate the old clocking API, as discussed on the list.
2004-02-25 13:16:12 +00:00
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
gst_pad_dispatcher (pad, (GstPadDispatcherFunction)
gst_pad_get_query_types_dispatcher, &result);
return result;
}
/**
* gst_pad_set_internal_link_function:
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* @pad: a #GstPad of either direction.
* @intlink: the #GstPadIntLinkFunction to set.
*
* Sets the given internal link function for the pad.
*/
void
gst_pad_set_internal_link_function (GstPad * pad, GstPadIntLinkFunction intlink)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_if_fail (GST_IS_PAD (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_INTLINKFUNC (pad) = intlink;
GST_CAT_DEBUG (GST_CAT_PADS, "internal link for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (intlink));
}
/**
* gst_pad_set_link_function:
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* @pad: a #GstPad.
* @link: the #GstPadLinkFunction to set.
gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. Original commit message from CVS: 2004-02-10 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. (gst_pad_get_event_masks, gst_pad_get_event_masks_default) (gst_pad_get_query_types, gst_pad_get_query_types_default): Routine docs. (gst_pad_set_link_function, gst_pad_set_fixate_function) (gst_pad_set_getcaps_function): Doc from Dave's negotation random doc. (gst_pad_unlink, gst_pad_is_linked): Docs. (gst_pad_renegotiate): A brief description of capsnego. (gst_pad_try_set_caps): Document. (gst_pad_try_set_caps_nonfixed): Document. (gst_pad_can_link_filtered, gst_pad_link_filtered): Doc fixes. (gst_pad_set_parent): Deprecated (although not out of the API). (gst_pad_get_parent): Deprecated, although many plugins use this. (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): Doc that these are private and will go away in 0.9. (gst_pad_perform_negotiate): Doc. (gst_pad_link_unnegotiate): I think this is meant to be static. (gst_pad_get_negotiated_caps, gst_pad_get_pad_template_caps) (gst_pad_template_get_caps_by_name, gst_pad_check_compatibility) (gst_pad_get_peer): Doc updates. (gst_pad_caps_change_notify): Doc. (gst_pad_alloc_buffer, gst_pad_push, gst_static_pad_template_get) (gst_ghost_pad_new): Doc fixes. * gst/gstobject.c (gst_object_get_parent, gst_object_unparent) (gst_object_check_uniqueness): * gst/gstelement.c (gst_element_add_pad) (gst_element_add_ghost_pad, gst_element_remove_pad) (gst_element_remove_ghost_pad, gst_element_get_pad) (gst_element_get_static_pad, gst_element_get_pad_list) (gst_element_class_get_pad_template_list) (gst_element_class_get_pad_template): Work on the docs. (gst_element_get_pad_template_list): Uses the class method. (gst_element_get_compatible_pad_template): Docs, and consolidate some test conditions. (gst_element_get_pad_from_template): New static function. (gst_element_request_compatible_pad): Docs, and work with non-request compatible templates. (gst_element_get_compatible_pad_filtered): Docs and remove redundant checks. (gst_element_get_compatible_pad, gst_element_link_pads_filtered) (gst_element_link_filtered, gst_element_link_many) (gst_element_link, gst_element_link_pads) (gst_element_unlink_many): Docs. 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-11 13:26:04 +00:00
*
* Sets the given link function for the pad. It will be called when the pad is
* linked or relinked with caps. The caps passed to the link function is
* the caps for the connnection. It can contain a non fixed caps.
gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. Original commit message from CVS: 2004-02-10 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. (gst_pad_get_event_masks, gst_pad_get_event_masks_default) (gst_pad_get_query_types, gst_pad_get_query_types_default): Routine docs. (gst_pad_set_link_function, gst_pad_set_fixate_function) (gst_pad_set_getcaps_function): Doc from Dave's negotation random doc. (gst_pad_unlink, gst_pad_is_linked): Docs. (gst_pad_renegotiate): A brief description of capsnego. (gst_pad_try_set_caps): Document. (gst_pad_try_set_caps_nonfixed): Document. (gst_pad_can_link_filtered, gst_pad_link_filtered): Doc fixes. (gst_pad_set_parent): Deprecated (although not out of the API). (gst_pad_get_parent): Deprecated, although many plugins use this. (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): Doc that these are private and will go away in 0.9. (gst_pad_perform_negotiate): Doc. (gst_pad_link_unnegotiate): I think this is meant to be static. (gst_pad_get_negotiated_caps, gst_pad_get_pad_template_caps) (gst_pad_template_get_caps_by_name, gst_pad_check_compatibility) (gst_pad_get_peer): Doc updates. (gst_pad_caps_change_notify): Doc. (gst_pad_alloc_buffer, gst_pad_push, gst_static_pad_template_get) (gst_ghost_pad_new): Doc fixes. * gst/gstobject.c (gst_object_get_parent, gst_object_unparent) (gst_object_check_uniqueness): * gst/gstelement.c (gst_element_add_pad) (gst_element_add_ghost_pad, gst_element_remove_pad) (gst_element_remove_ghost_pad, gst_element_get_pad) (gst_element_get_static_pad, gst_element_get_pad_list) (gst_element_class_get_pad_template_list) (gst_element_class_get_pad_template): Work on the docs. (gst_element_get_pad_template_list): Uses the class method. (gst_element_get_compatible_pad_template): Docs, and consolidate some test conditions. (gst_element_get_pad_from_template): New static function. (gst_element_request_compatible_pad): Docs, and work with non-request compatible templates. (gst_element_get_compatible_pad_filtered): Docs and remove redundant checks. (gst_element_get_compatible_pad, gst_element_link_pads_filtered) (gst_element_link_filtered, gst_element_link_many) (gst_element_link, gst_element_link_pads) (gst_element_unlink_many): Docs. 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-11 13:26:04 +00:00
*
* The return value GST_PAD_LINK_OK should be used when the connection can be
* made.
gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. Original commit message from CVS: 2004-02-10 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. (gst_pad_get_event_masks, gst_pad_get_event_masks_default) (gst_pad_get_query_types, gst_pad_get_query_types_default): Routine docs. (gst_pad_set_link_function, gst_pad_set_fixate_function) (gst_pad_set_getcaps_function): Doc from Dave's negotation random doc. (gst_pad_unlink, gst_pad_is_linked): Docs. (gst_pad_renegotiate): A brief description of capsnego. (gst_pad_try_set_caps): Document. (gst_pad_try_set_caps_nonfixed): Document. (gst_pad_can_link_filtered, gst_pad_link_filtered): Doc fixes. (gst_pad_set_parent): Deprecated (although not out of the API). (gst_pad_get_parent): Deprecated, although many plugins use this. (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): Doc that these are private and will go away in 0.9. (gst_pad_perform_negotiate): Doc. (gst_pad_link_unnegotiate): I think this is meant to be static. (gst_pad_get_negotiated_caps, gst_pad_get_pad_template_caps) (gst_pad_template_get_caps_by_name, gst_pad_check_compatibility) (gst_pad_get_peer): Doc updates. (gst_pad_caps_change_notify): Doc. (gst_pad_alloc_buffer, gst_pad_push, gst_static_pad_template_get) (gst_ghost_pad_new): Doc fixes. * gst/gstobject.c (gst_object_get_parent, gst_object_unparent) (gst_object_check_uniqueness): * gst/gstelement.c (gst_element_add_pad) (gst_element_add_ghost_pad, gst_element_remove_pad) (gst_element_remove_ghost_pad, gst_element_get_pad) (gst_element_get_static_pad, gst_element_get_pad_list) (gst_element_class_get_pad_template_list) (gst_element_class_get_pad_template): Work on the docs. (gst_element_get_pad_template_list): Uses the class method. (gst_element_get_compatible_pad_template): Docs, and consolidate some test conditions. (gst_element_get_pad_from_template): New static function. (gst_element_request_compatible_pad): Docs, and work with non-request compatible templates. (gst_element_get_compatible_pad_filtered): Docs and remove redundant checks. (gst_element_get_compatible_pad, gst_element_link_pads_filtered) (gst_element_link_filtered, gst_element_link_many) (gst_element_link, gst_element_link_pads) (gst_element_unlink_many): Docs. 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-11 13:26:04 +00:00
*
* The return value GST_PAD_LINK_REFUSED should be used when the connection
* cannot be made for some reason.
*/
void
gst_pad_set_link_function (GstPad * pad, GstPadLinkFunction link)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_if_fail (GST_IS_PAD (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_LINKFUNC (pad) = link;
GST_CAT_DEBUG (GST_CAT_PADS, "linkfunc for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (link));
}
/**
* gst_pad_set_unlink_function:
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* @pad: a #GstPad.
* @unlink: the #GstPadUnlinkFunction to set.
*
* Sets the given unlink function for the pad. It will be called
* when the pad is unlinked.
*/
void
gst_pad_set_unlink_function (GstPad * pad, GstPadUnlinkFunction unlink)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_if_fail (GST_IS_PAD (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_UNLINKFUNC (pad) = unlink;
GST_CAT_DEBUG (GST_CAT_PADS, "unlinkfunc for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (unlink));
}
/**
* gst_pad_set_getcaps_function:
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* @pad: a #GstPad.
* @getcaps: the #GstPadGetCapsFunction to set.
gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. Original commit message from CVS: 2004-02-10 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. (gst_pad_get_event_masks, gst_pad_get_event_masks_default) (gst_pad_get_query_types, gst_pad_get_query_types_default): Routine docs. (gst_pad_set_link_function, gst_pad_set_fixate_function) (gst_pad_set_getcaps_function): Doc from Dave's negotation random doc. (gst_pad_unlink, gst_pad_is_linked): Docs. (gst_pad_renegotiate): A brief description of capsnego. (gst_pad_try_set_caps): Document. (gst_pad_try_set_caps_nonfixed): Document. (gst_pad_can_link_filtered, gst_pad_link_filtered): Doc fixes. (gst_pad_set_parent): Deprecated (although not out of the API). (gst_pad_get_parent): Deprecated, although many plugins use this. (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): Doc that these are private and will go away in 0.9. (gst_pad_perform_negotiate): Doc. (gst_pad_link_unnegotiate): I think this is meant to be static. (gst_pad_get_negotiated_caps, gst_pad_get_pad_template_caps) (gst_pad_template_get_caps_by_name, gst_pad_check_compatibility) (gst_pad_get_peer): Doc updates. (gst_pad_caps_change_notify): Doc. (gst_pad_alloc_buffer, gst_pad_push, gst_static_pad_template_get) (gst_ghost_pad_new): Doc fixes. * gst/gstobject.c (gst_object_get_parent, gst_object_unparent) (gst_object_check_uniqueness): * gst/gstelement.c (gst_element_add_pad) (gst_element_add_ghost_pad, gst_element_remove_pad) (gst_element_remove_ghost_pad, gst_element_get_pad) (gst_element_get_static_pad, gst_element_get_pad_list) (gst_element_class_get_pad_template_list) (gst_element_class_get_pad_template): Work on the docs. (gst_element_get_pad_template_list): Uses the class method. (gst_element_get_compatible_pad_template): Docs, and consolidate some test conditions. (gst_element_get_pad_from_template): New static function. (gst_element_request_compatible_pad): Docs, and work with non-request compatible templates. (gst_element_get_compatible_pad_filtered): Docs and remove redundant checks. (gst_element_get_compatible_pad, gst_element_link_pads_filtered) (gst_element_link_filtered, gst_element_link_many) (gst_element_link, gst_element_link_pads) (gst_element_unlink_many): Docs. 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-11 13:26:04 +00:00
*
* Sets the given getcaps function for the pad. @getcaps should return the
* allowable caps for a pad in the context of the element's state, its link to
* other elements, and the devices or files it has opened. These caps must be a
* subset of the pad template caps. In the NULL state with no links, @getcaps
* should ideally return the same caps as the pad template. In rare
* circumstances, an object property can affect the caps returned by @getcaps,
* but this is discouraged.
*
* You do not need to call this function if @pad's allowed caps are always the
* same as the pad template caps. This can only be true if the padtemplate
* has fixed simple caps.
gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. Original commit message from CVS: 2004-02-10 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. (gst_pad_get_event_masks, gst_pad_get_event_masks_default) (gst_pad_get_query_types, gst_pad_get_query_types_default): Routine docs. (gst_pad_set_link_function, gst_pad_set_fixate_function) (gst_pad_set_getcaps_function): Doc from Dave's negotation random doc. (gst_pad_unlink, gst_pad_is_linked): Docs. (gst_pad_renegotiate): A brief description of capsnego. (gst_pad_try_set_caps): Document. (gst_pad_try_set_caps_nonfixed): Document. (gst_pad_can_link_filtered, gst_pad_link_filtered): Doc fixes. (gst_pad_set_parent): Deprecated (although not out of the API). (gst_pad_get_parent): Deprecated, although many plugins use this. (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): Doc that these are private and will go away in 0.9. (gst_pad_perform_negotiate): Doc. (gst_pad_link_unnegotiate): I think this is meant to be static. (gst_pad_get_negotiated_caps, gst_pad_get_pad_template_caps) (gst_pad_template_get_caps_by_name, gst_pad_check_compatibility) (gst_pad_get_peer): Doc updates. (gst_pad_caps_change_notify): Doc. (gst_pad_alloc_buffer, gst_pad_push, gst_static_pad_template_get) (gst_ghost_pad_new): Doc fixes. * gst/gstobject.c (gst_object_get_parent, gst_object_unparent) (gst_object_check_uniqueness): * gst/gstelement.c (gst_element_add_pad) (gst_element_add_ghost_pad, gst_element_remove_pad) (gst_element_remove_ghost_pad, gst_element_get_pad) (gst_element_get_static_pad, gst_element_get_pad_list) (gst_element_class_get_pad_template_list) (gst_element_class_get_pad_template): Work on the docs. (gst_element_get_pad_template_list): Uses the class method. (gst_element_get_compatible_pad_template): Docs, and consolidate some test conditions. (gst_element_get_pad_from_template): New static function. (gst_element_request_compatible_pad): Docs, and work with non-request compatible templates. (gst_element_get_compatible_pad_filtered): Docs and remove redundant checks. (gst_element_get_compatible_pad, gst_element_link_pads_filtered) (gst_element_link_filtered, gst_element_link_many) (gst_element_link, gst_element_link_pads) (gst_element_unlink_many): Docs. 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-11 13:26:04 +00:00
*
* For most filters, the caps returned by @getcaps is directly affected by the
* allowed caps on other pads. For demuxers and decoders, the caps returned by
* the srcpad's getcaps function is directly related to the stream data. Again,
* @getcaps should return the most specific caps it reasonably can, since this
* helps with autoplugging.
gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. Original commit message from CVS: 2004-02-10 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. (gst_pad_get_event_masks, gst_pad_get_event_masks_default) (gst_pad_get_query_types, gst_pad_get_query_types_default): Routine docs. (gst_pad_set_link_function, gst_pad_set_fixate_function) (gst_pad_set_getcaps_function): Doc from Dave's negotation random doc. (gst_pad_unlink, gst_pad_is_linked): Docs. (gst_pad_renegotiate): A brief description of capsnego. (gst_pad_try_set_caps): Document. (gst_pad_try_set_caps_nonfixed): Document. (gst_pad_can_link_filtered, gst_pad_link_filtered): Doc fixes. (gst_pad_set_parent): Deprecated (although not out of the API). (gst_pad_get_parent): Deprecated, although many plugins use this. (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): Doc that these are private and will go away in 0.9. (gst_pad_perform_negotiate): Doc. (gst_pad_link_unnegotiate): I think this is meant to be static. (gst_pad_get_negotiated_caps, gst_pad_get_pad_template_caps) (gst_pad_template_get_caps_by_name, gst_pad_check_compatibility) (gst_pad_get_peer): Doc updates. (gst_pad_caps_change_notify): Doc. (gst_pad_alloc_buffer, gst_pad_push, gst_static_pad_template_get) (gst_ghost_pad_new): Doc fixes. * gst/gstobject.c (gst_object_get_parent, gst_object_unparent) (gst_object_check_uniqueness): * gst/gstelement.c (gst_element_add_pad) (gst_element_add_ghost_pad, gst_element_remove_pad) (gst_element_remove_ghost_pad, gst_element_get_pad) (gst_element_get_static_pad, gst_element_get_pad_list) (gst_element_class_get_pad_template_list) (gst_element_class_get_pad_template): Work on the docs. (gst_element_get_pad_template_list): Uses the class method. (gst_element_get_compatible_pad_template): Docs, and consolidate some test conditions. (gst_element_get_pad_from_template): New static function. (gst_element_request_compatible_pad): Docs, and work with non-request compatible templates. (gst_element_get_compatible_pad_filtered): Docs and remove redundant checks. (gst_element_get_compatible_pad, gst_element_link_pads_filtered) (gst_element_link_filtered, gst_element_link_many) (gst_element_link, gst_element_link_pads) (gst_element_unlink_many): Docs. 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-11 13:26:04 +00:00
*
* Note that the return value from @getcaps is owned by the caller, so the caller
* should unref the caps after usage.
*/
void
gst_pad_set_getcaps_function (GstPad * pad, GstPadGetCapsFunction getcaps)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_if_fail (GST_IS_PAD (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_GETCAPSFUNC (pad) = getcaps;
GST_CAT_DEBUG (GST_CAT_PADS, "getcapsfunc for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (getcaps));
}
/**
* gst_pad_set_acceptcaps_function:
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* @pad: a #GstPad.
* @acceptcaps: the #GstPadAcceptCapsFunction to set.
*
* Sets the given acceptcaps function for the pad. The acceptcaps function
* will be called to check if the pad can accept the given caps.
*/
void
gst_pad_set_acceptcaps_function (GstPad * pad,
GstPadAcceptCapsFunction acceptcaps)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_if_fail (GST_IS_PAD (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_ACCEPTCAPSFUNC (pad) = acceptcaps;
GST_CAT_DEBUG (GST_CAT_PADS, "acceptcapsfunc for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (acceptcaps));
}
/**
* gst_pad_set_fixatecaps_function:
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* @pad: a #GstPad.
* @fixatecaps: the #GstPadFixateCapsFunction to set.
*
* Sets the given fixatecaps function for the pad. The fixatecaps function
* will be called whenever the default values for a GstCaps needs to be
* filled in.
*/
void
gst_pad_set_fixatecaps_function (GstPad * pad,
GstPadFixateCapsFunction fixatecaps)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_if_fail (GST_IS_PAD (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_FIXATECAPSFUNC (pad) = fixatecaps;
GST_CAT_DEBUG (GST_CAT_PADS, "fixatecapsfunc for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (fixatecaps));
}
/**
* gst_pad_set_setcaps_function:
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* @pad: a #GstPad.
* @setcaps: the #GstPadSetCapsFunction to set.
*
* Sets the given setcaps function for the pad. The setcaps function
* will be called whenever a buffer with a new media type is pushed or
* pulled from the pad. The pad/element needs to update it's internal
* structures to process the new media type. If this new type is not
* acceptable, the setcaps function should return FALSE.
*/
void
gst_pad_set_setcaps_function (GstPad * pad, GstPadSetCapsFunction setcaps)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_if_fail (GST_IS_PAD (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_SETCAPSFUNC (pad) = setcaps;
GST_CAT_DEBUG (GST_CAT_PADS, "setcapsfunc for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (setcaps));
}
/**
* gst_pad_set_bufferalloc_function:
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* @pad: a sink #GstPad.
* @bufalloc: the #GstPadBufferAllocFunction to set.
*
* Sets the given bufferalloc function for the pad. Note that the
* bufferalloc function can only be set on sinkpads.
*/
void
gst_pad_set_bufferalloc_function (GstPad * pad,
GstPadBufferAllocFunction bufalloc)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_if_fail (GST_IS_PAD (pad));
g_return_if_fail (GST_PAD_IS_SINK (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_BUFFERALLOCFUNC (pad) = bufalloc;
GST_CAT_DEBUG (GST_CAT_PADS, "bufferallocfunc for %s:%s set to %s",
GST_DEBUG_PAD_NAME (pad), GST_DEBUG_FUNCPTR_NAME (bufalloc));
}
/**
* gst_pad_unlink:
* @srcpad: the source #GstPad to unlink.
* @sinkpad: the sink #GstPad to unlink.
*
gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. Original commit message from CVS: 2004-02-10 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. (gst_pad_get_event_masks, gst_pad_get_event_masks_default) (gst_pad_get_query_types, gst_pad_get_query_types_default): Routine docs. (gst_pad_set_link_function, gst_pad_set_fixate_function) (gst_pad_set_getcaps_function): Doc from Dave's negotation random doc. (gst_pad_unlink, gst_pad_is_linked): Docs. (gst_pad_renegotiate): A brief description of capsnego. (gst_pad_try_set_caps): Document. (gst_pad_try_set_caps_nonfixed): Document. (gst_pad_can_link_filtered, gst_pad_link_filtered): Doc fixes. (gst_pad_set_parent): Deprecated (although not out of the API). (gst_pad_get_parent): Deprecated, although many plugins use this. (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): Doc that these are private and will go away in 0.9. (gst_pad_perform_negotiate): Doc. (gst_pad_link_unnegotiate): I think this is meant to be static. (gst_pad_get_negotiated_caps, gst_pad_get_pad_template_caps) (gst_pad_template_get_caps_by_name, gst_pad_check_compatibility) (gst_pad_get_peer): Doc updates. (gst_pad_caps_change_notify): Doc. (gst_pad_alloc_buffer, gst_pad_push, gst_static_pad_template_get) (gst_ghost_pad_new): Doc fixes. * gst/gstobject.c (gst_object_get_parent, gst_object_unparent) (gst_object_check_uniqueness): * gst/gstelement.c (gst_element_add_pad) (gst_element_add_ghost_pad, gst_element_remove_pad) (gst_element_remove_ghost_pad, gst_element_get_pad) (gst_element_get_static_pad, gst_element_get_pad_list) (gst_element_class_get_pad_template_list) (gst_element_class_get_pad_template): Work on the docs. (gst_element_get_pad_template_list): Uses the class method. (gst_element_get_compatible_pad_template): Docs, and consolidate some test conditions. (gst_element_get_pad_from_template): New static function. (gst_element_request_compatible_pad): Docs, and work with non-request compatible templates. (gst_element_get_compatible_pad_filtered): Docs and remove redundant checks. (gst_element_get_compatible_pad, gst_element_link_pads_filtered) (gst_element_link_filtered, gst_element_link_many) (gst_element_link, gst_element_link_pads) (gst_element_unlink_many): Docs. 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-11 13:26:04 +00:00
* Unlinks the source pad from the sink pad. Will emit the "unlinked" signal on
* both pads.
*
* Returns: TRUE if the pads were unlinked. This function returns FALSE if
* the pads were not linked together.
*
* MT safe.
*/
gboolean
gst_pad_unlink (GstPad * srcpad, GstPad * sinkpad)
{
g_return_val_if_fail (GST_IS_PAD (srcpad), FALSE);
g_return_val_if_fail (GST_IS_PAD (sinkpad), FALSE);
GST_CAT_INFO (GST_CAT_ELEMENT_PADS, "unlinking %s:%s(%p) and %s:%s(%p)",
GST_DEBUG_PAD_NAME (srcpad), srcpad,
GST_DEBUG_PAD_NAME (sinkpad), sinkpad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (srcpad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY (GST_PAD_DIRECTION (srcpad) != GST_PAD_SRC))
goto not_srcpad;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (sinkpad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY (GST_PAD_DIRECTION (sinkpad) != GST_PAD_SINK))
goto not_sinkpad;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY (GST_PAD_PEER (srcpad) != sinkpad))
goto not_linked_together;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (GST_PAD_UNLINKFUNC (srcpad)) {
GST_PAD_UNLINKFUNC (srcpad) (srcpad);
}
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (GST_PAD_UNLINKFUNC (sinkpad)) {
GST_PAD_UNLINKFUNC (sinkpad) (sinkpad);
}
/* first clear peers */
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_PEER (srcpad) = NULL;
GST_PAD_PEER (sinkpad) = NULL;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (sinkpad);
GST_UNLOCK (srcpad);
/* fire off a signal to each of the pads telling them
* that they've been unlinked */
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_signal_emit (G_OBJECT (srcpad), gst_pad_signals[PAD_UNLINKED], 0, sinkpad);
g_signal_emit (G_OBJECT (sinkpad), gst_pad_signals[PAD_UNLINKED], 0, srcpad);
GST_CAT_INFO (GST_CAT_ELEMENT_PADS, "unlinked %s:%s and %s:%s",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (srcpad), GST_DEBUG_PAD_NAME (sinkpad));
return TRUE;
not_srcpad:
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_critical ("pad %s is not a source pad", GST_PAD_NAME (srcpad));
GST_UNLOCK (srcpad);
return FALSE;
}
not_sinkpad:
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_critical ("pad %s is not a sink pad", GST_PAD_NAME (sinkpad));
GST_UNLOCK (sinkpad);
GST_UNLOCK (srcpad);
return FALSE;
}
not_linked_together:
{
/* we do not emit a warning in this case because unlinking cannot
* be made MT safe.*/
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (sinkpad);
GST_UNLOCK (srcpad);
return FALSE;
}
}
/**
* gst_pad_is_linked:
* @pad: pad to check
*
gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. Original commit message from CVS: 2004-02-10 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. (gst_pad_get_event_masks, gst_pad_get_event_masks_default) (gst_pad_get_query_types, gst_pad_get_query_types_default): Routine docs. (gst_pad_set_link_function, gst_pad_set_fixate_function) (gst_pad_set_getcaps_function): Doc from Dave's negotation random doc. (gst_pad_unlink, gst_pad_is_linked): Docs. (gst_pad_renegotiate): A brief description of capsnego. (gst_pad_try_set_caps): Document. (gst_pad_try_set_caps_nonfixed): Document. (gst_pad_can_link_filtered, gst_pad_link_filtered): Doc fixes. (gst_pad_set_parent): Deprecated (although not out of the API). (gst_pad_get_parent): Deprecated, although many plugins use this. (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): Doc that these are private and will go away in 0.9. (gst_pad_perform_negotiate): Doc. (gst_pad_link_unnegotiate): I think this is meant to be static. (gst_pad_get_negotiated_caps, gst_pad_get_pad_template_caps) (gst_pad_template_get_caps_by_name, gst_pad_check_compatibility) (gst_pad_get_peer): Doc updates. (gst_pad_caps_change_notify): Doc. (gst_pad_alloc_buffer, gst_pad_push, gst_static_pad_template_get) (gst_ghost_pad_new): Doc fixes. * gst/gstobject.c (gst_object_get_parent, gst_object_unparent) (gst_object_check_uniqueness): * gst/gstelement.c (gst_element_add_pad) (gst_element_add_ghost_pad, gst_element_remove_pad) (gst_element_remove_ghost_pad, gst_element_get_pad) (gst_element_get_static_pad, gst_element_get_pad_list) (gst_element_class_get_pad_template_list) (gst_element_class_get_pad_template): Work on the docs. (gst_element_get_pad_template_list): Uses the class method. (gst_element_get_compatible_pad_template): Docs, and consolidate some test conditions. (gst_element_get_pad_from_template): New static function. (gst_element_request_compatible_pad): Docs, and work with non-request compatible templates. (gst_element_get_compatible_pad_filtered): Docs and remove redundant checks. (gst_element_get_compatible_pad, gst_element_link_pads_filtered) (gst_element_link_filtered, gst_element_link_many) (gst_element_link, gst_element_link_pads) (gst_element_unlink_many): Docs. 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-11 13:26:04 +00:00
* Checks if a @pad is linked to another pad or not.
*
gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. Original commit message from CVS: 2004-02-10 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. (gst_pad_get_event_masks, gst_pad_get_event_masks_default) (gst_pad_get_query_types, gst_pad_get_query_types_default): Routine docs. (gst_pad_set_link_function, gst_pad_set_fixate_function) (gst_pad_set_getcaps_function): Doc from Dave's negotation random doc. (gst_pad_unlink, gst_pad_is_linked): Docs. (gst_pad_renegotiate): A brief description of capsnego. (gst_pad_try_set_caps): Document. (gst_pad_try_set_caps_nonfixed): Document. (gst_pad_can_link_filtered, gst_pad_link_filtered): Doc fixes. (gst_pad_set_parent): Deprecated (although not out of the API). (gst_pad_get_parent): Deprecated, although many plugins use this. (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): Doc that these are private and will go away in 0.9. (gst_pad_perform_negotiate): Doc. (gst_pad_link_unnegotiate): I think this is meant to be static. (gst_pad_get_negotiated_caps, gst_pad_get_pad_template_caps) (gst_pad_template_get_caps_by_name, gst_pad_check_compatibility) (gst_pad_get_peer): Doc updates. (gst_pad_caps_change_notify): Doc. (gst_pad_alloc_buffer, gst_pad_push, gst_static_pad_template_get) (gst_ghost_pad_new): Doc fixes. * gst/gstobject.c (gst_object_get_parent, gst_object_unparent) (gst_object_check_uniqueness): * gst/gstelement.c (gst_element_add_pad) (gst_element_add_ghost_pad, gst_element_remove_pad) (gst_element_remove_ghost_pad, gst_element_get_pad) (gst_element_get_static_pad, gst_element_get_pad_list) (gst_element_class_get_pad_template_list) (gst_element_class_get_pad_template): Work on the docs. (gst_element_get_pad_template_list): Uses the class method. (gst_element_get_compatible_pad_template): Docs, and consolidate some test conditions. (gst_element_get_pad_from_template): New static function. (gst_element_request_compatible_pad): Docs, and work with non-request compatible templates. (gst_element_get_compatible_pad_filtered): Docs and remove redundant checks. (gst_element_get_compatible_pad, gst_element_link_pads_filtered) (gst_element_link_filtered, gst_element_link_many) (gst_element_link, gst_element_link_pads) (gst_element_unlink_many): Docs. 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-11 13:26:04 +00:00
* Returns: TRUE if the pad is linked, FALSE otherwise.
*
* MT safe.
*/
gboolean
gst_pad_is_linked (GstPad * pad)
{
gboolean result;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (pad);
result = (GST_PAD_PEER (pad) != NULL);
GST_UNLOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
return result;
}
static gboolean
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
gst_pad_link_check_compatible_unlocked (GstPad * src, GstPad * sink)
{
GstCaps *srccaps;
GstCaps *sinkcaps;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
srccaps = gst_pad_get_caps_unlocked (src);
sinkcaps = gst_pad_get_caps_unlocked (sink);
GST_CAT_DEBUG (GST_CAT_CAPS, " src caps %" GST_PTR_FORMAT, srccaps);
GST_CAT_DEBUG (GST_CAT_CAPS, "sink caps %" GST_PTR_FORMAT, sinkcaps);
if (srccaps && sinkcaps) {
GstCaps *icaps;
icaps = gst_caps_intersect (srccaps, sinkcaps);
gst_caps_unref (srccaps);
gst_caps_unref (sinkcaps);
GST_CAT_DEBUG (GST_CAT_CAPS,
"intersection caps %p %" GST_PTR_FORMAT, icaps, icaps);
if (!icaps || gst_caps_is_empty (icaps)) {
gst_caps_unref (icaps);
return FALSE;
}
gst_caps_unref (icaps);
}
return TRUE;
}
/* FIXME leftover from an attempt at refactoring... */
/* call with the two pads unlocked */
static GstPadLinkReturn
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
gst_pad_link_prepare (GstPad * srcpad, GstPad * sinkpad)
{
/* generic checks */
g_return_val_if_fail (GST_IS_PAD (srcpad), GST_PAD_LINK_REFUSED);
g_return_val_if_fail (GST_IS_PAD (sinkpad), GST_PAD_LINK_REFUSED);
GST_CAT_INFO (GST_CAT_PADS, "trying to link %s:%s and %s:%s",
GST_DEBUG_PAD_NAME (srcpad), GST_DEBUG_PAD_NAME (sinkpad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (srcpad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY (GST_PAD_DIRECTION (srcpad) != GST_PAD_SRC))
goto not_srcpad;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY (GST_PAD_PEER (srcpad) != NULL))
goto src_was_linked;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (sinkpad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY (GST_PAD_DIRECTION (sinkpad) != GST_PAD_SINK))
goto not_sinkpad;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY (GST_PAD_PEER (sinkpad) != NULL))
goto sink_was_linked;
/* check pad caps for non-empty intersection */
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (!gst_pad_link_check_compatible_unlocked (srcpad, sinkpad)) {
goto no_format;
}
/* FIXME check pad scheduling for non-empty intersection */
return GST_PAD_LINK_OK;
not_srcpad:
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_critical ("pad %s is not a source pad", GST_PAD_NAME (srcpad));
GST_UNLOCK (srcpad);
return GST_PAD_LINK_WRONG_DIRECTION;
}
src_was_linked:
{
GST_CAT_INFO (GST_CAT_PADS, "src %s:%s was linked",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (srcpad));
/* we do not emit a warning in this case because unlinking cannot
* be made MT safe.*/
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (srcpad);
return GST_PAD_LINK_WAS_LINKED;
}
not_sinkpad:
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_critical ("pad %s is not a sink pad", GST_PAD_NAME (sinkpad));
GST_UNLOCK (sinkpad);
GST_UNLOCK (srcpad);
return GST_PAD_LINK_WRONG_DIRECTION;
}
sink_was_linked:
{
GST_CAT_INFO (GST_CAT_PADS, "sink %s:%s was linked",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (sinkpad));
/* we do not emit a warning in this case because unlinking cannot
* be made MT safe.*/
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (sinkpad);
GST_UNLOCK (srcpad);
return GST_PAD_LINK_WAS_LINKED;
}
no_format:
{
GST_CAT_INFO (GST_CAT_PADS, "caps are incompatible");
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (sinkpad);
GST_UNLOCK (srcpad);
return GST_PAD_LINK_NOFORMAT;
}
}
/**
* gst_pad_link:
* @srcpad: the source #GstPad to link.
* @sinkpad: the sink #GstPad to link.
*
* Links the source pad and the sink pad.
*
* Returns: A result code indicating if the connection worked or
* what went wrong.
*
* MT Safe.
*/
GstPadLinkReturn
gst_pad_link (GstPad * srcpad, GstPad * sinkpad)
{
GstPadLinkReturn result;
/* prepare will also lock the two pads */
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
result = gst_pad_link_prepare (srcpad, sinkpad);
if (result != GST_PAD_LINK_OK)
goto prepare_failed;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (sinkpad);
GST_UNLOCK (srcpad);
/* FIXME released the locks here, concurrent thread might link
* something else. */
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (GST_PAD_LINKFUNC (srcpad)) {
/* this one will call the peer link function */
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
result = GST_PAD_LINKFUNC (srcpad) (srcpad, sinkpad);
} else if (GST_PAD_LINKFUNC (sinkpad)) {
/* if no source link function, we need to call the sink link
* function ourselves. */
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
result = GST_PAD_LINKFUNC (sinkpad) (sinkpad, srcpad);
} else {
result = GST_PAD_LINK_OK;
}
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (srcpad);
GST_LOCK (sinkpad);
if (result == GST_PAD_LINK_OK) {
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_PEER (srcpad) = sinkpad;
GST_PAD_PEER (sinkpad) = srcpad;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (sinkpad);
GST_UNLOCK (srcpad);
/* fire off a signal to each of the pads telling them
* that they've been linked */
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_signal_emit (G_OBJECT (srcpad), gst_pad_signals[PAD_LINKED], 0, sinkpad);
g_signal_emit (G_OBJECT (sinkpad), gst_pad_signals[PAD_LINKED], 0, srcpad);
GST_CAT_INFO (GST_CAT_PADS, "linked %s:%s and %s:%s, successful",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (srcpad), GST_DEBUG_PAD_NAME (sinkpad));
} else {
GST_CAT_INFO (GST_CAT_PADS, "link between %s:%s and %s:%s failed",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (srcpad), GST_DEBUG_PAD_NAME (sinkpad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (sinkpad);
GST_UNLOCK (srcpad);
}
return result;
prepare_failed:
{
return result;
}
}
static void
gst_pad_set_pad_template (GstPad * pad, GstPadTemplate * templ)
{
/* this function would need checks if it weren't static */
GST_LOCK (pad);
gst_object_replace ((GstObject **) & pad->padtemplate, (GstObject *) templ);
GST_UNLOCK (pad);
if (templ) {
gst_object_sink (GST_OBJECT (templ));
g_signal_emit (G_OBJECT (templ),
gst_pad_template_signals[TEMPL_PAD_CREATED], 0, pad);
}
}
gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. Original commit message from CVS: 2004-02-24 Andy Wingo <wingo@pobox.com> * gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. * gst/schedulers/gstoptimalscheduler.c I added a mess of prototypes at the top of the file by way of documentation. Some of the operations on chains and groups were re-organized. (create_group): Added a type argument so if the group is enabled, the setup_group_scheduler knows what to do. (group_elements): Added a type argument here, too, to be passed on to create_group. (group_element_set_enabled): If an unlinked PLAYING element is added to a bin, we have to create a new group to hold the element, and this function will be called before the group is added to the chain. Thus we have a valid case for group->chain==NULL. Instead of calling chain_group_set_enabled, just set the flag on the group (the chain's status will be set when the group is added to it). (gst_opt_scheduler_state_transition, chain_group_set_enabled): Setup the group scheduler when the group is enabled, not specifically when an element goes PAUSED->PLAYING. This means PLAYING elements can be added, linked, and scheduled into a PLAYING pipeline, as was intended. (add_to_group): Don't ref the group twice. I don't know when this double-ref got in here. Removing it has the potential to cause segfaults if other parts of the scheduler are buggy. If you find that the scheduler is segfaulting for you, put in an extra ref here and see if that hacks over the underlying issue. Of course, then find out what code is unreffing a group it doesn't own... (create_group): Make the extra refcount floating, and remove it after adding the element. This means that... (unref_group): Destroy when the refcount reaches 0, not 1, like every other refcounted object in the known universe. (remove_from_group): When a group becomes empty, set it to be not active, and remove it from its chain. Don't unref it again, there's no floating reference any more. (destroy_group): We have to remove the group from the chain in remove_from_group (rather than here) to break refcounting cycles (the chain always has a ref on the group). So assert that group->chain==NULL. (ref_group_by_count): Removed, it was commented out anyway. (merge_chains): Use the remove_from_chain and add_to_chain primitives to do the reparenting, instead of rolling our own implementation. (add_to_chain): The first non-disabled group in the chain's group list will be the entry point for the chain. Because buffers can accumulate in loop elements' peer bufpens, we preferentially schedule loop groups before get groups to avoid unnecessary execution of get-based groups when the bufpens are already full. (gst_opt_scheduler_schedule_run_queue): Debug fixes. (get_group_schedule_function): Ditto. (loop_group_schedule_function): Ditto. (gst_opt_scheduler_loop_wrapper): Ditto. (gst_opt_scheduler_iterate): Ditto. I understand the opt scheduler now, yippee! * gst/gstpad.c: All throughout, added FIXMEs to look at for 0.9. (gst_pad_get_name, gst_pad_set_chain_function) (gst_pad_set_get_function, gst_pad_set_event_function) (gst_pad_set_event_mask_function, gst_pad_get_event_masks) (gst_pad_get_event_masks_default, gst_pad_set_convert_function) (gst_pad_set_query_function, gst_pad_get_query_types) (gst_pad_get_query_types_default) (gst_pad_set_internal_link_function) (gst_pad_set_formats_function, gst_pad_set_link_function) (gst_pad_set_fixate_function, gst_pad_set_getcaps_function) (gst_pad_set_bufferalloc_function, gst_pad_unlink) (gst_pad_renegotiate, gst_pad_set_parent, gst_pad_get_parent) (gst_pad_add_ghost_pad, gst_pad_proxy_getcaps) (gst_pad_proxy_pad_link, gst_pad_proxy_fixate) (gst_pad_get_pad_template_caps, gst_pad_check_compatibility) (gst_pad_get_peer, gst_pad_get_allowed_caps) (gst_pad_alloc_buffer, gst_pad_push, gst_pad_pull) (gst_pad_selectv, gst_pad_select, gst_pad_template_get_caps) (gst_pad_event_default_dispatch, gst_pad_event_default) (gst_pad_dispatcher, gst_pad_send_event, gst_pad_convert_default) (gst_pad_convert, gst_pad_query_default, gst_pad_query) (gst_pad_get_formats_default, gst_pad_get_formats): Better argument checks, and some doc fixes. (gst_pad_custom_new_from_template): Um, does anyone use these functions? Actually make a custom pad instead of a normal one. (gst_pad_try_set_caps): Transpose some checks. (gst_pad_try_set_caps_nonfixed): Same, and use a macro to check if the pad is in negotiation. (gst_pad_try_relink_filtered): Use pad_link_prepare. * gst/gstelement.c: Remove prototypes also defined in gstclock.h. * gst/gstelement.h: * gst/gstclock.h: Un-deprecate the old clocking API, as discussed on the list.
2004-02-25 13:16:12 +00:00
/**
* gst_pad_get_pad_template:
gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. Original commit message from CVS: 2004-02-24 Andy Wingo <wingo@pobox.com> * gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. * gst/schedulers/gstoptimalscheduler.c I added a mess of prototypes at the top of the file by way of documentation. Some of the operations on chains and groups were re-organized. (create_group): Added a type argument so if the group is enabled, the setup_group_scheduler knows what to do. (group_elements): Added a type argument here, too, to be passed on to create_group. (group_element_set_enabled): If an unlinked PLAYING element is added to a bin, we have to create a new group to hold the element, and this function will be called before the group is added to the chain. Thus we have a valid case for group->chain==NULL. Instead of calling chain_group_set_enabled, just set the flag on the group (the chain's status will be set when the group is added to it). (gst_opt_scheduler_state_transition, chain_group_set_enabled): Setup the group scheduler when the group is enabled, not specifically when an element goes PAUSED->PLAYING. This means PLAYING elements can be added, linked, and scheduled into a PLAYING pipeline, as was intended. (add_to_group): Don't ref the group twice. I don't know when this double-ref got in here. Removing it has the potential to cause segfaults if other parts of the scheduler are buggy. If you find that the scheduler is segfaulting for you, put in an extra ref here and see if that hacks over the underlying issue. Of course, then find out what code is unreffing a group it doesn't own... (create_group): Make the extra refcount floating, and remove it after adding the element. This means that... (unref_group): Destroy when the refcount reaches 0, not 1, like every other refcounted object in the known universe. (remove_from_group): When a group becomes empty, set it to be not active, and remove it from its chain. Don't unref it again, there's no floating reference any more. (destroy_group): We have to remove the group from the chain in remove_from_group (rather than here) to break refcounting cycles (the chain always has a ref on the group). So assert that group->chain==NULL. (ref_group_by_count): Removed, it was commented out anyway. (merge_chains): Use the remove_from_chain and add_to_chain primitives to do the reparenting, instead of rolling our own implementation. (add_to_chain): The first non-disabled group in the chain's group list will be the entry point for the chain. Because buffers can accumulate in loop elements' peer bufpens, we preferentially schedule loop groups before get groups to avoid unnecessary execution of get-based groups when the bufpens are already full. (gst_opt_scheduler_schedule_run_queue): Debug fixes. (get_group_schedule_function): Ditto. (loop_group_schedule_function): Ditto. (gst_opt_scheduler_loop_wrapper): Ditto. (gst_opt_scheduler_iterate): Ditto. I understand the opt scheduler now, yippee! * gst/gstpad.c: All throughout, added FIXMEs to look at for 0.9. (gst_pad_get_name, gst_pad_set_chain_function) (gst_pad_set_get_function, gst_pad_set_event_function) (gst_pad_set_event_mask_function, gst_pad_get_event_masks) (gst_pad_get_event_masks_default, gst_pad_set_convert_function) (gst_pad_set_query_function, gst_pad_get_query_types) (gst_pad_get_query_types_default) (gst_pad_set_internal_link_function) (gst_pad_set_formats_function, gst_pad_set_link_function) (gst_pad_set_fixate_function, gst_pad_set_getcaps_function) (gst_pad_set_bufferalloc_function, gst_pad_unlink) (gst_pad_renegotiate, gst_pad_set_parent, gst_pad_get_parent) (gst_pad_add_ghost_pad, gst_pad_proxy_getcaps) (gst_pad_proxy_pad_link, gst_pad_proxy_fixate) (gst_pad_get_pad_template_caps, gst_pad_check_compatibility) (gst_pad_get_peer, gst_pad_get_allowed_caps) (gst_pad_alloc_buffer, gst_pad_push, gst_pad_pull) (gst_pad_selectv, gst_pad_select, gst_pad_template_get_caps) (gst_pad_event_default_dispatch, gst_pad_event_default) (gst_pad_dispatcher, gst_pad_send_event, gst_pad_convert_default) (gst_pad_convert, gst_pad_query_default, gst_pad_query) (gst_pad_get_formats_default, gst_pad_get_formats): Better argument checks, and some doc fixes. (gst_pad_custom_new_from_template): Um, does anyone use these functions? Actually make a custom pad instead of a normal one. (gst_pad_try_set_caps): Transpose some checks. (gst_pad_try_set_caps_nonfixed): Same, and use a macro to check if the pad is in negotiation. (gst_pad_try_relink_filtered): Use pad_link_prepare. * gst/gstelement.c: Remove prototypes also defined in gstclock.h. * gst/gstelement.h: * gst/gstclock.h: Un-deprecate the old clocking API, as discussed on the list.
2004-02-25 13:16:12 +00:00
* @pad: a #GstPad.
*
gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. Original commit message from CVS: 2004-02-24 Andy Wingo <wingo@pobox.com> * gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. * gst/schedulers/gstoptimalscheduler.c I added a mess of prototypes at the top of the file by way of documentation. Some of the operations on chains and groups were re-organized. (create_group): Added a type argument so if the group is enabled, the setup_group_scheduler knows what to do. (group_elements): Added a type argument here, too, to be passed on to create_group. (group_element_set_enabled): If an unlinked PLAYING element is added to a bin, we have to create a new group to hold the element, and this function will be called before the group is added to the chain. Thus we have a valid case for group->chain==NULL. Instead of calling chain_group_set_enabled, just set the flag on the group (the chain's status will be set when the group is added to it). (gst_opt_scheduler_state_transition, chain_group_set_enabled): Setup the group scheduler when the group is enabled, not specifically when an element goes PAUSED->PLAYING. This means PLAYING elements can be added, linked, and scheduled into a PLAYING pipeline, as was intended. (add_to_group): Don't ref the group twice. I don't know when this double-ref got in here. Removing it has the potential to cause segfaults if other parts of the scheduler are buggy. If you find that the scheduler is segfaulting for you, put in an extra ref here and see if that hacks over the underlying issue. Of course, then find out what code is unreffing a group it doesn't own... (create_group): Make the extra refcount floating, and remove it after adding the element. This means that... (unref_group): Destroy when the refcount reaches 0, not 1, like every other refcounted object in the known universe. (remove_from_group): When a group becomes empty, set it to be not active, and remove it from its chain. Don't unref it again, there's no floating reference any more. (destroy_group): We have to remove the group from the chain in remove_from_group (rather than here) to break refcounting cycles (the chain always has a ref on the group). So assert that group->chain==NULL. (ref_group_by_count): Removed, it was commented out anyway. (merge_chains): Use the remove_from_chain and add_to_chain primitives to do the reparenting, instead of rolling our own implementation. (add_to_chain): The first non-disabled group in the chain's group list will be the entry point for the chain. Because buffers can accumulate in loop elements' peer bufpens, we preferentially schedule loop groups before get groups to avoid unnecessary execution of get-based groups when the bufpens are already full. (gst_opt_scheduler_schedule_run_queue): Debug fixes. (get_group_schedule_function): Ditto. (loop_group_schedule_function): Ditto. (gst_opt_scheduler_loop_wrapper): Ditto. (gst_opt_scheduler_iterate): Ditto. I understand the opt scheduler now, yippee! * gst/gstpad.c: All throughout, added FIXMEs to look at for 0.9. (gst_pad_get_name, gst_pad_set_chain_function) (gst_pad_set_get_function, gst_pad_set_event_function) (gst_pad_set_event_mask_function, gst_pad_get_event_masks) (gst_pad_get_event_masks_default, gst_pad_set_convert_function) (gst_pad_set_query_function, gst_pad_get_query_types) (gst_pad_get_query_types_default) (gst_pad_set_internal_link_function) (gst_pad_set_formats_function, gst_pad_set_link_function) (gst_pad_set_fixate_function, gst_pad_set_getcaps_function) (gst_pad_set_bufferalloc_function, gst_pad_unlink) (gst_pad_renegotiate, gst_pad_set_parent, gst_pad_get_parent) (gst_pad_add_ghost_pad, gst_pad_proxy_getcaps) (gst_pad_proxy_pad_link, gst_pad_proxy_fixate) (gst_pad_get_pad_template_caps, gst_pad_check_compatibility) (gst_pad_get_peer, gst_pad_get_allowed_caps) (gst_pad_alloc_buffer, gst_pad_push, gst_pad_pull) (gst_pad_selectv, gst_pad_select, gst_pad_template_get_caps) (gst_pad_event_default_dispatch, gst_pad_event_default) (gst_pad_dispatcher, gst_pad_send_event, gst_pad_convert_default) (gst_pad_convert, gst_pad_query_default, gst_pad_query) (gst_pad_get_formats_default, gst_pad_get_formats): Better argument checks, and some doc fixes. (gst_pad_custom_new_from_template): Um, does anyone use these functions? Actually make a custom pad instead of a normal one. (gst_pad_try_set_caps): Transpose some checks. (gst_pad_try_set_caps_nonfixed): Same, and use a macro to check if the pad is in negotiation. (gst_pad_try_relink_filtered): Use pad_link_prepare. * gst/gstelement.c: Remove prototypes also defined in gstclock.h. * gst/gstelement.h: * gst/gstclock.h: Un-deprecate the old clocking API, as discussed on the list.
2004-02-25 13:16:12 +00:00
* Gets the template for @pad.
*
gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. Original commit message from CVS: 2004-02-24 Andy Wingo <wingo@pobox.com> * gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. * gst/schedulers/gstoptimalscheduler.c I added a mess of prototypes at the top of the file by way of documentation. Some of the operations on chains and groups were re-organized. (create_group): Added a type argument so if the group is enabled, the setup_group_scheduler knows what to do. (group_elements): Added a type argument here, too, to be passed on to create_group. (group_element_set_enabled): If an unlinked PLAYING element is added to a bin, we have to create a new group to hold the element, and this function will be called before the group is added to the chain. Thus we have a valid case for group->chain==NULL. Instead of calling chain_group_set_enabled, just set the flag on the group (the chain's status will be set when the group is added to it). (gst_opt_scheduler_state_transition, chain_group_set_enabled): Setup the group scheduler when the group is enabled, not specifically when an element goes PAUSED->PLAYING. This means PLAYING elements can be added, linked, and scheduled into a PLAYING pipeline, as was intended. (add_to_group): Don't ref the group twice. I don't know when this double-ref got in here. Removing it has the potential to cause segfaults if other parts of the scheduler are buggy. If you find that the scheduler is segfaulting for you, put in an extra ref here and see if that hacks over the underlying issue. Of course, then find out what code is unreffing a group it doesn't own... (create_group): Make the extra refcount floating, and remove it after adding the element. This means that... (unref_group): Destroy when the refcount reaches 0, not 1, like every other refcounted object in the known universe. (remove_from_group): When a group becomes empty, set it to be not active, and remove it from its chain. Don't unref it again, there's no floating reference any more. (destroy_group): We have to remove the group from the chain in remove_from_group (rather than here) to break refcounting cycles (the chain always has a ref on the group). So assert that group->chain==NULL. (ref_group_by_count): Removed, it was commented out anyway. (merge_chains): Use the remove_from_chain and add_to_chain primitives to do the reparenting, instead of rolling our own implementation. (add_to_chain): The first non-disabled group in the chain's group list will be the entry point for the chain. Because buffers can accumulate in loop elements' peer bufpens, we preferentially schedule loop groups before get groups to avoid unnecessary execution of get-based groups when the bufpens are already full. (gst_opt_scheduler_schedule_run_queue): Debug fixes. (get_group_schedule_function): Ditto. (loop_group_schedule_function): Ditto. (gst_opt_scheduler_loop_wrapper): Ditto. (gst_opt_scheduler_iterate): Ditto. I understand the opt scheduler now, yippee! * gst/gstpad.c: All throughout, added FIXMEs to look at for 0.9. (gst_pad_get_name, gst_pad_set_chain_function) (gst_pad_set_get_function, gst_pad_set_event_function) (gst_pad_set_event_mask_function, gst_pad_get_event_masks) (gst_pad_get_event_masks_default, gst_pad_set_convert_function) (gst_pad_set_query_function, gst_pad_get_query_types) (gst_pad_get_query_types_default) (gst_pad_set_internal_link_function) (gst_pad_set_formats_function, gst_pad_set_link_function) (gst_pad_set_fixate_function, gst_pad_set_getcaps_function) (gst_pad_set_bufferalloc_function, gst_pad_unlink) (gst_pad_renegotiate, gst_pad_set_parent, gst_pad_get_parent) (gst_pad_add_ghost_pad, gst_pad_proxy_getcaps) (gst_pad_proxy_pad_link, gst_pad_proxy_fixate) (gst_pad_get_pad_template_caps, gst_pad_check_compatibility) (gst_pad_get_peer, gst_pad_get_allowed_caps) (gst_pad_alloc_buffer, gst_pad_push, gst_pad_pull) (gst_pad_selectv, gst_pad_select, gst_pad_template_get_caps) (gst_pad_event_default_dispatch, gst_pad_event_default) (gst_pad_dispatcher, gst_pad_send_event, gst_pad_convert_default) (gst_pad_convert, gst_pad_query_default, gst_pad_query) (gst_pad_get_formats_default, gst_pad_get_formats): Better argument checks, and some doc fixes. (gst_pad_custom_new_from_template): Um, does anyone use these functions? Actually make a custom pad instead of a normal one. (gst_pad_try_set_caps): Transpose some checks. (gst_pad_try_set_caps_nonfixed): Same, and use a macro to check if the pad is in negotiation. (gst_pad_try_relink_filtered): Use pad_link_prepare. * gst/gstelement.c: Remove prototypes also defined in gstclock.h. * gst/gstelement.h: * gst/gstclock.h: Un-deprecate the old clocking API, as discussed on the list.
2004-02-25 13:16:12 +00:00
* Returns: the #GstPadTemplate from which this pad was instantiated, or %NULL
* if this pad has no template.
*
* FIXME: currently returns an unrefcounted padtemplate.
*/
GstPadTemplate *
gst_pad_get_pad_template (GstPad * pad)
{
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
return GST_PAD_PAD_TEMPLATE (pad);
}
/* should be called with the pad LOCK held */
/* refs the caps, so caller is responsible for getting it unreffed */
static GstCaps *
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
gst_pad_get_caps_unlocked (GstPad * pad)
{
GstCaps *result = NULL;
GST_CAT_DEBUG (GST_CAT_CAPS, "get pad caps of %s:%s (%p)",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (pad), pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (GST_PAD_GETCAPSFUNC (pad)) {
GST_CAT_DEBUG (GST_CAT_CAPS, "dispatching to pad getcaps function");
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_FLAG_SET (pad, GST_PAD_IN_GETCAPS);
GST_UNLOCK (pad);
result = GST_PAD_GETCAPSFUNC (pad) (pad);
GST_LOCK (pad);
GST_FLAG_UNSET (pad, GST_PAD_IN_GETCAPS);
if (result == NULL) {
g_critical ("pad %s:%s returned NULL caps from getcaps function",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (pad));
} else {
GST_CAT_DEBUG (GST_CAT_CAPS,
"pad getcaps %s:%s returned %" GST_PTR_FORMAT,
GST_DEBUG_PAD_NAME (pad), result);
#ifndef G_DISABLE_ASSERT
/* check that the returned caps are a real subset of the template caps */
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (GST_PAD_PAD_TEMPLATE (pad)) {
const GstCaps *templ_caps =
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_TEMPLATE_CAPS (GST_PAD_PAD_TEMPLATE (pad));
if (!gst_caps_is_subset (result, templ_caps)) {
GstCaps *temp;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_CAT_ERROR_OBJECT (GST_CAT_CAPS, pad,
"pad returned caps %" GST_PTR_FORMAT
" which are not a real subset of its template caps %"
GST_PTR_FORMAT, result, templ_caps);
g_warning
("pad %s:%s returned caps that are not a real subset of its template caps",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (pad));
temp = gst_caps_intersect (templ_caps, result);
gst_caps_unref (result);
result = temp;
}
}
#endif
goto done;
}
}
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (GST_PAD_PAD_TEMPLATE (pad)) {
GstPadTemplate *templ = GST_PAD_PAD_TEMPLATE (pad);
result = GST_PAD_TEMPLATE_CAPS (templ);
GST_CAT_DEBUG (GST_CAT_CAPS,
"using pad template %p with caps %p %" GST_PTR_FORMAT, templ, result,
result);
result = gst_caps_ref (result);
goto done;
}
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (GST_PAD_CAPS (pad)) {
result = GST_PAD_CAPS (pad);
GST_CAT_DEBUG (GST_CAT_CAPS,
"using pad caps %p %" GST_PTR_FORMAT, result, result);
result = gst_caps_ref (result);
goto done;
}
GST_CAT_DEBUG (GST_CAT_CAPS, "pad has no caps");
result = gst_caps_new_empty ();
done:
return result;
}
/**
* gst_pad_get_caps:
* @pad: a #GstPad to get the capabilities of.
*
* Gets the capabilities of this pad.
*
* Returns: the #GstCaps of this pad. This function returns a new caps, so use
* gst_caps_unref to get rid of it.
*
* MT safe.
*/
GstCaps *
gst_pad_get_caps (GstPad * pad)
{
GstCaps *result = NULL;
gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. Original commit message from CVS: 2004-02-24 Andy Wingo <wingo@pobox.com> * gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. * gst/schedulers/gstoptimalscheduler.c I added a mess of prototypes at the top of the file by way of documentation. Some of the operations on chains and groups were re-organized. (create_group): Added a type argument so if the group is enabled, the setup_group_scheduler knows what to do. (group_elements): Added a type argument here, too, to be passed on to create_group. (group_element_set_enabled): If an unlinked PLAYING element is added to a bin, we have to create a new group to hold the element, and this function will be called before the group is added to the chain. Thus we have a valid case for group->chain==NULL. Instead of calling chain_group_set_enabled, just set the flag on the group (the chain's status will be set when the group is added to it). (gst_opt_scheduler_state_transition, chain_group_set_enabled): Setup the group scheduler when the group is enabled, not specifically when an element goes PAUSED->PLAYING. This means PLAYING elements can be added, linked, and scheduled into a PLAYING pipeline, as was intended. (add_to_group): Don't ref the group twice. I don't know when this double-ref got in here. Removing it has the potential to cause segfaults if other parts of the scheduler are buggy. If you find that the scheduler is segfaulting for you, put in an extra ref here and see if that hacks over the underlying issue. Of course, then find out what code is unreffing a group it doesn't own... (create_group): Make the extra refcount floating, and remove it after adding the element. This means that... (unref_group): Destroy when the refcount reaches 0, not 1, like every other refcounted object in the known universe. (remove_from_group): When a group becomes empty, set it to be not active, and remove it from its chain. Don't unref it again, there's no floating reference any more. (destroy_group): We have to remove the group from the chain in remove_from_group (rather than here) to break refcounting cycles (the chain always has a ref on the group). So assert that group->chain==NULL. (ref_group_by_count): Removed, it was commented out anyway. (merge_chains): Use the remove_from_chain and add_to_chain primitives to do the reparenting, instead of rolling our own implementation. (add_to_chain): The first non-disabled group in the chain's group list will be the entry point for the chain. Because buffers can accumulate in loop elements' peer bufpens, we preferentially schedule loop groups before get groups to avoid unnecessary execution of get-based groups when the bufpens are already full. (gst_opt_scheduler_schedule_run_queue): Debug fixes. (get_group_schedule_function): Ditto. (loop_group_schedule_function): Ditto. (gst_opt_scheduler_loop_wrapper): Ditto. (gst_opt_scheduler_iterate): Ditto. I understand the opt scheduler now, yippee! * gst/gstpad.c: All throughout, added FIXMEs to look at for 0.9. (gst_pad_get_name, gst_pad_set_chain_function) (gst_pad_set_get_function, gst_pad_set_event_function) (gst_pad_set_event_mask_function, gst_pad_get_event_masks) (gst_pad_get_event_masks_default, gst_pad_set_convert_function) (gst_pad_set_query_function, gst_pad_get_query_types) (gst_pad_get_query_types_default) (gst_pad_set_internal_link_function) (gst_pad_set_formats_function, gst_pad_set_link_function) (gst_pad_set_fixate_function, gst_pad_set_getcaps_function) (gst_pad_set_bufferalloc_function, gst_pad_unlink) (gst_pad_renegotiate, gst_pad_set_parent, gst_pad_get_parent) (gst_pad_add_ghost_pad, gst_pad_proxy_getcaps) (gst_pad_proxy_pad_link, gst_pad_proxy_fixate) (gst_pad_get_pad_template_caps, gst_pad_check_compatibility) (gst_pad_get_peer, gst_pad_get_allowed_caps) (gst_pad_alloc_buffer, gst_pad_push, gst_pad_pull) (gst_pad_selectv, gst_pad_select, gst_pad_template_get_caps) (gst_pad_event_default_dispatch, gst_pad_event_default) (gst_pad_dispatcher, gst_pad_send_event, gst_pad_convert_default) (gst_pad_convert, gst_pad_query_default, gst_pad_query) (gst_pad_get_formats_default, gst_pad_get_formats): Better argument checks, and some doc fixes. (gst_pad_custom_new_from_template): Um, does anyone use these functions? Actually make a custom pad instead of a normal one. (gst_pad_try_set_caps): Transpose some checks. (gst_pad_try_set_caps_nonfixed): Same, and use a macro to check if the pad is in negotiation. (gst_pad_try_relink_filtered): Use pad_link_prepare. * gst/gstelement.c: Remove prototypes also defined in gstclock.h. * gst/gstelement.h: * gst/gstclock.h: Un-deprecate the old clocking API, as discussed on the list.
2004-02-25 13:16:12 +00:00
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (pad);
GST_CAT_DEBUG (GST_CAT_CAPS, "get pad caps of %s:%s (%p)",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (pad), pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY (GST_PAD_IS_IN_GETCAPS (pad)))
goto was_dispatching;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
result = gst_pad_get_caps_unlocked (pad);
GST_UNLOCK (pad);
return result;
was_dispatching:
{
GST_CAT_DEBUG (GST_CAT_CAPS,
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
"pad %s:%s is already dispatching!", GST_DEBUG_PAD_NAME (pad));
g_warning ("pad %s:%s recursively called getcaps!",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (pad));
GST_UNLOCK (pad);
return NULL;
}
}
/**
* gst_pad_peer_get_caps:
* @pad: a #GstPad to get the peer capabilities of.
*
* Gets the capabilities of the peer connected to this pad.
*
* Returns: the #GstCaps of the peer pad. This function returns a new caps, so use
* gst_caps_unref to get rid of it. this function returns NULL if there is no
* peer pad or when this function is called recursively from a getcaps function.
*/
GstCaps *
gst_pad_peer_get_caps (GstPad * pad)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GstPad *peerpad;
GstCaps *result = NULL;
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (pad);
GST_CAT_DEBUG (GST_CAT_CAPS, "get peer caps of %s:%s (%p)",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (pad), pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
peerpad = GST_PAD_PEER (pad);
if (G_UNLIKELY (peerpad == NULL))
goto no_peer;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY (GST_PAD_IS_IN_GETCAPS (peerpad)))
goto was_dispatching;
gst_object_ref (peerpad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
result = gst_pad_get_caps (peerpad);
gst_object_unref (peerpad);
return result;
no_peer:
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (pad);
return NULL;
}
was_dispatching:
{
GST_CAT_DEBUG (GST_CAT_CAPS,
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
"pad %s:%s is already dispatching!", GST_DEBUG_PAD_NAME (pad));
g_warning ("pad %s:%s recursively called getcaps!",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (pad));
GST_UNLOCK (pad);
return NULL;
}
}
/**
* gst_pad_fixate_caps:
* @pad: a #GstPad to fixate
* @caps: the #GstCaps to fixate
*
tests/network-clock.scm: Commentary update. Original commit message from CVS: 2005-07-01 Andy Wingo <wingo@pobox.com> * tests/network-clock.scm: Commentary update. * gst/elements/gstidentity.c (PROP_DUPLICATE): Gone daddy gone. Didn't really make sense, not implementable with basetransform, etc. (gst_identity_transform): Unref inbuf via make_writable. Feeble attempt at implementing the sync property, needs an unlock method. * gst/base/gstbasetransform.c (gst_base_transform_transform_caps): New func, by default returns the same caps (the identity transformation). (gst_base_transform_getcaps): Uses transform_caps to return something sensible. (gst_base_transform_setcaps): Complicated logic to get caps on both pads, even if they are different, and to call set_caps once for every time both pads get their caps set. (gst_base_transform_handle_buffer): Give the ref to the transform function. Allows in-place modification of the buffer. * gst/base/gstbasetransform.h (transform_caps): New class method. Given caps on one side, what can I do on the other. (set_caps): Take two caps, one for each side of the element. * gst/gstpad.h: * gst/gstpad.c (gst_pad_fixate_caps): Change prototype to modify caps in place. This is safe because we can check the mutability of the caps, and a good idea because fixate functions are just called as a matter of last resort. (Not actually implemented.) (gst_pad_set_caps): If the caps we're setting is actually the same as the existing pad caps, just update the pointer without calling setcaps. Assert that caps is either NULL or fixed, as per the docs. * gst/gstghostpad.c: Update for fixate changes.
2005-07-01 16:46:59 +00:00
* Fixate a caps on the given pad. Modifies the caps in place, so you should be
* that the caps are actually writable (see gst_caps_make_writable()).
*/
tests/network-clock.scm: Commentary update. Original commit message from CVS: 2005-07-01 Andy Wingo <wingo@pobox.com> * tests/network-clock.scm: Commentary update. * gst/elements/gstidentity.c (PROP_DUPLICATE): Gone daddy gone. Didn't really make sense, not implementable with basetransform, etc. (gst_identity_transform): Unref inbuf via make_writable. Feeble attempt at implementing the sync property, needs an unlock method. * gst/base/gstbasetransform.c (gst_base_transform_transform_caps): New func, by default returns the same caps (the identity transformation). (gst_base_transform_getcaps): Uses transform_caps to return something sensible. (gst_base_transform_setcaps): Complicated logic to get caps on both pads, even if they are different, and to call set_caps once for every time both pads get their caps set. (gst_base_transform_handle_buffer): Give the ref to the transform function. Allows in-place modification of the buffer. * gst/base/gstbasetransform.h (transform_caps): New class method. Given caps on one side, what can I do on the other. (set_caps): Take two caps, one for each side of the element. * gst/gstpad.h: * gst/gstpad.c (gst_pad_fixate_caps): Change prototype to modify caps in place. This is safe because we can check the mutability of the caps, and a good idea because fixate functions are just called as a matter of last resort. (Not actually implemented.) (gst_pad_set_caps): If the caps we're setting is actually the same as the existing pad caps, just update the pointer without calling setcaps. Assert that caps is either NULL or fixed, as per the docs. * gst/gstghostpad.c: Update for fixate changes.
2005-07-01 16:46:59 +00:00
void
gst_pad_fixate_caps (GstPad * pad, GstCaps * caps)
{
GstPadFixateCapsFunction fixatefunc;
g_return_if_fail (GST_IS_PAD (pad));
g_return_if_fail (caps != NULL);
fixatefunc = GST_PAD_FIXATECAPSFUNC (pad);
if (fixatefunc) {
fixatefunc (pad, caps);
}
}
/**
* gst_pad_accept_caps:
* @pad: a #GstPad to check
*
* Check if the given pad accepts the caps.
*
* Returns: TRUE if the pad can accept the caps.
*/
gboolean
gst_pad_accept_caps (GstPad * pad, GstCaps * caps)
{
gboolean result;
GstPadAcceptCapsFunction acceptfunc;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
/* any pad can be unnegotiated */
if (caps == NULL)
return TRUE;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (pad);
acceptfunc = GST_PAD_ACCEPTCAPSFUNC (pad);
GST_CAT_DEBUG (GST_CAT_CAPS, "pad accept caps of %s:%s (%p)",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (pad), pad);
GST_UNLOCK (pad);
if (acceptfunc) {
/* we can call the function */
result = acceptfunc (pad, caps);
} else {
/* else see get the caps and see if it intersects to something
* not empty */
GstCaps *intersect;
GstCaps *allowed;
allowed = gst_pad_get_caps (pad);
if (allowed) {
intersect = gst_caps_intersect (allowed, caps);
result = !gst_caps_is_empty (intersect);
gst_caps_unref (allowed);
gst_caps_unref (intersect);
} else {
result = FALSE;
}
}
return result;
}
/**
* gst_pad_peer_accept_caps:
* @pad: a #GstPad to check
*
* Check if the given pad accepts the caps.
*
* Returns: TRUE if the pad can accept the caps.
*/
gboolean
gst_pad_peer_accept_caps (GstPad * pad, GstCaps * caps)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GstPad *peerpad;
gboolean result;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (pad);
GST_CAT_DEBUG (GST_CAT_CAPS, "peer accept caps of %s:%s (%p)",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (pad), pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
peerpad = GST_PAD_PEER (pad);
if (G_UNLIKELY (peerpad == NULL))
goto no_peer;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
result = gst_pad_accept_caps (peerpad, caps);
GST_UNLOCK (pad);
return result;
no_peer:
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (pad);
return TRUE;
}
}
/**
* gst_pad_set_caps:
* @pad: a #GstPad to set the capabilities of.
* @caps: a #GstCaps to set.
*
* Sets the capabilities of this pad. The caps must be fixed. Any previous
* caps on the pad will be unreffed. This function refs the caps so you should
* unref if as soon as you don't need it anymore.
* It is possible to set NULL caps, which will make the pad unnegotiated
* again.
*
* Returns: TRUE if the caps could be set. FALSE if the caps were not fixed
* or bad parameters were provided to this function.
*
* MT safe.
*/
gboolean
gst_pad_set_caps (GstPad * pad, GstCaps * caps)
{
GstPadSetCapsFunction setcaps;
tests/network-clock.scm: Commentary update. Original commit message from CVS: 2005-07-01 Andy Wingo <wingo@pobox.com> * tests/network-clock.scm: Commentary update. * gst/elements/gstidentity.c (PROP_DUPLICATE): Gone daddy gone. Didn't really make sense, not implementable with basetransform, etc. (gst_identity_transform): Unref inbuf via make_writable. Feeble attempt at implementing the sync property, needs an unlock method. * gst/base/gstbasetransform.c (gst_base_transform_transform_caps): New func, by default returns the same caps (the identity transformation). (gst_base_transform_getcaps): Uses transform_caps to return something sensible. (gst_base_transform_setcaps): Complicated logic to get caps on both pads, even if they are different, and to call set_caps once for every time both pads get their caps set. (gst_base_transform_handle_buffer): Give the ref to the transform function. Allows in-place modification of the buffer. * gst/base/gstbasetransform.h (transform_caps): New class method. Given caps on one side, what can I do on the other. (set_caps): Take two caps, one for each side of the element. * gst/gstpad.h: * gst/gstpad.c (gst_pad_fixate_caps): Change prototype to modify caps in place. This is safe because we can check the mutability of the caps, and a good idea because fixate functions are just called as a matter of last resort. (Not actually implemented.) (gst_pad_set_caps): If the caps we're setting is actually the same as the existing pad caps, just update the pointer without calling setcaps. Assert that caps is either NULL or fixed, as per the docs. * gst/gstghostpad.c: Update for fixate changes.
2005-07-01 16:46:59 +00:00
GstCaps *existing;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
tests/network-clock.scm: Commentary update. Original commit message from CVS: 2005-07-01 Andy Wingo <wingo@pobox.com> * tests/network-clock.scm: Commentary update. * gst/elements/gstidentity.c (PROP_DUPLICATE): Gone daddy gone. Didn't really make sense, not implementable with basetransform, etc. (gst_identity_transform): Unref inbuf via make_writable. Feeble attempt at implementing the sync property, needs an unlock method. * gst/base/gstbasetransform.c (gst_base_transform_transform_caps): New func, by default returns the same caps (the identity transformation). (gst_base_transform_getcaps): Uses transform_caps to return something sensible. (gst_base_transform_setcaps): Complicated logic to get caps on both pads, even if they are different, and to call set_caps once for every time both pads get their caps set. (gst_base_transform_handle_buffer): Give the ref to the transform function. Allows in-place modification of the buffer. * gst/base/gstbasetransform.h (transform_caps): New class method. Given caps on one side, what can I do on the other. (set_caps): Take two caps, one for each side of the element. * gst/gstpad.h: * gst/gstpad.c (gst_pad_fixate_caps): Change prototype to modify caps in place. This is safe because we can check the mutability of the caps, and a good idea because fixate functions are just called as a matter of last resort. (Not actually implemented.) (gst_pad_set_caps): If the caps we're setting is actually the same as the existing pad caps, just update the pointer without calling setcaps. Assert that caps is either NULL or fixed, as per the docs. * gst/gstghostpad.c: Update for fixate changes.
2005-07-01 16:46:59 +00:00
g_return_val_if_fail (caps == NULL || gst_caps_is_fixed (caps), FALSE);
GST_LOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
setcaps = GST_PAD_SETCAPSFUNC (pad);
tests/network-clock.scm: Commentary update. Original commit message from CVS: 2005-07-01 Andy Wingo <wingo@pobox.com> * tests/network-clock.scm: Commentary update. * gst/elements/gstidentity.c (PROP_DUPLICATE): Gone daddy gone. Didn't really make sense, not implementable with basetransform, etc. (gst_identity_transform): Unref inbuf via make_writable. Feeble attempt at implementing the sync property, needs an unlock method. * gst/base/gstbasetransform.c (gst_base_transform_transform_caps): New func, by default returns the same caps (the identity transformation). (gst_base_transform_getcaps): Uses transform_caps to return something sensible. (gst_base_transform_setcaps): Complicated logic to get caps on both pads, even if they are different, and to call set_caps once for every time both pads get their caps set. (gst_base_transform_handle_buffer): Give the ref to the transform function. Allows in-place modification of the buffer. * gst/base/gstbasetransform.h (transform_caps): New class method. Given caps on one side, what can I do on the other. (set_caps): Take two caps, one for each side of the element. * gst/gstpad.h: * gst/gstpad.c (gst_pad_fixate_caps): Change prototype to modify caps in place. This is safe because we can check the mutability of the caps, and a good idea because fixate functions are just called as a matter of last resort. (Not actually implemented.) (gst_pad_set_caps): If the caps we're setting is actually the same as the existing pad caps, just update the pointer without calling setcaps. Assert that caps is either NULL or fixed, as per the docs. * gst/gstghostpad.c: Update for fixate changes.
2005-07-01 16:46:59 +00:00
existing = GST_PAD_CAPS (pad);
if (caps == existing)
goto setting_same_caps;
else if (caps && existing && gst_caps_is_equal (caps, existing))
goto setting_same_caps;
/* call setcaps function to configure the pad */
if (setcaps != NULL && caps) {
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (!GST_PAD_IS_IN_SETCAPS (pad)) {
GST_FLAG_SET (pad, GST_PAD_IN_SETCAPS);
GST_UNLOCK (pad);
if (!setcaps (pad, caps))
goto could_not_set;
GST_LOCK (pad);
GST_FLAG_UNSET (pad, GST_PAD_IN_SETCAPS);
} else {
GST_CAT_DEBUG (GST_CAT_CAPS, "pad %s:%s was dispatching",
GST_DEBUG_PAD_NAME (pad));
}
}
gst_caps_replace (&GST_PAD_CAPS (pad), caps);
GST_CAT_DEBUG (GST_CAT_CAPS, "%s:%s caps %" GST_PTR_FORMAT,
GST_DEBUG_PAD_NAME (pad), caps);
GST_UNLOCK (pad);
g_object_notify (G_OBJECT (pad), "caps");
return TRUE;
tests/network-clock.scm: Commentary update. Original commit message from CVS: 2005-07-01 Andy Wingo <wingo@pobox.com> * tests/network-clock.scm: Commentary update. * gst/elements/gstidentity.c (PROP_DUPLICATE): Gone daddy gone. Didn't really make sense, not implementable with basetransform, etc. (gst_identity_transform): Unref inbuf via make_writable. Feeble attempt at implementing the sync property, needs an unlock method. * gst/base/gstbasetransform.c (gst_base_transform_transform_caps): New func, by default returns the same caps (the identity transformation). (gst_base_transform_getcaps): Uses transform_caps to return something sensible. (gst_base_transform_setcaps): Complicated logic to get caps on both pads, even if they are different, and to call set_caps once for every time both pads get their caps set. (gst_base_transform_handle_buffer): Give the ref to the transform function. Allows in-place modification of the buffer. * gst/base/gstbasetransform.h (transform_caps): New class method. Given caps on one side, what can I do on the other. (set_caps): Take two caps, one for each side of the element. * gst/gstpad.h: * gst/gstpad.c (gst_pad_fixate_caps): Change prototype to modify caps in place. This is safe because we can check the mutability of the caps, and a good idea because fixate functions are just called as a matter of last resort. (Not actually implemented.) (gst_pad_set_caps): If the caps we're setting is actually the same as the existing pad caps, just update the pointer without calling setcaps. Assert that caps is either NULL or fixed, as per the docs. * gst/gstghostpad.c: Update for fixate changes.
2005-07-01 16:46:59 +00:00
setting_same_caps:
{
GST_UNLOCK (pad);
gst_caps_replace (&GST_PAD_CAPS (pad), caps);
GST_CAT_DEBUG_OBJECT (GST_CAT_CAPS, pad,
"caps %" GST_PTR_FORMAT " same as existing, updating ptr only", caps);
return TRUE;
}
/* errors */
could_not_set:
{
GST_LOCK (pad);
GST_FLAG_UNSET (pad, GST_PAD_IN_SETCAPS);
GST_CAT_DEBUG (GST_CAT_CAPS,
"pad %s:%s, caps %" GST_PTR_FORMAT " could not be set",
GST_DEBUG_PAD_NAME (pad), caps);
GST_UNLOCK (pad);
return FALSE;
}
}
static gboolean
gst_pad_configure_sink (GstPad * pad, GstCaps * caps)
{
GstPadAcceptCapsFunction acceptcaps;
GstPadSetCapsFunction setcaps;
gboolean res;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
acceptcaps = GST_PAD_ACCEPTCAPSFUNC (pad);
setcaps = GST_PAD_SETCAPSFUNC (pad);
/* See if pad accepts the caps, by calling acceptcaps, only
* needed if no setcaps function */
if (setcaps == NULL && acceptcaps != NULL) {
if (!acceptcaps (pad, caps))
goto not_accepted;
}
/* set caps on pad if call succeeds */
res = gst_pad_set_caps (pad, caps);
/* no need to unref the caps here, set_caps takes a ref and
* our ref goes away when we leave this function. */
return res;
not_accepted:
{
GST_CAT_DEBUG (GST_CAT_CAPS, "caps %" GST_PTR_FORMAT " not accepted", caps);
return FALSE;
}
}
static gboolean
gst_pad_configure_src (GstPad * pad, GstCaps * caps)
{
GstPadAcceptCapsFunction acceptcaps;
GstPadSetCapsFunction setcaps;
gboolean res;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
acceptcaps = GST_PAD_ACCEPTCAPSFUNC (pad);
setcaps = GST_PAD_SETCAPSFUNC (pad);
/* See if pad accepts the caps, by calling acceptcaps, only
* needed if no setcaps function */
if (setcaps == NULL && acceptcaps != NULL) {
if (!acceptcaps (pad, caps))
goto not_accepted;
}
/* set caps on pad if call succeeds */
res = gst_pad_set_caps (pad, caps);
/* no need to unref the caps here, set_caps takes a ref and
* our ref goes away when we leave this function. */
return res;
not_accepted:
{
GST_CAT_DEBUG (GST_CAT_CAPS, "caps %" GST_PTR_FORMAT " not accepted", caps);
return FALSE;
}
}
/**
* gst_pad_get_pad_template_caps:
* @pad: a #GstPad to get the template capabilities from.
*
gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. Original commit message from CVS: 2004-02-10 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. (gst_pad_get_event_masks, gst_pad_get_event_masks_default) (gst_pad_get_query_types, gst_pad_get_query_types_default): Routine docs. (gst_pad_set_link_function, gst_pad_set_fixate_function) (gst_pad_set_getcaps_function): Doc from Dave's negotation random doc. (gst_pad_unlink, gst_pad_is_linked): Docs. (gst_pad_renegotiate): A brief description of capsnego. (gst_pad_try_set_caps): Document. (gst_pad_try_set_caps_nonfixed): Document. (gst_pad_can_link_filtered, gst_pad_link_filtered): Doc fixes. (gst_pad_set_parent): Deprecated (although not out of the API). (gst_pad_get_parent): Deprecated, although many plugins use this. (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): Doc that these are private and will go away in 0.9. (gst_pad_perform_negotiate): Doc. (gst_pad_link_unnegotiate): I think this is meant to be static. (gst_pad_get_negotiated_caps, gst_pad_get_pad_template_caps) (gst_pad_template_get_caps_by_name, gst_pad_check_compatibility) (gst_pad_get_peer): Doc updates. (gst_pad_caps_change_notify): Doc. (gst_pad_alloc_buffer, gst_pad_push, gst_static_pad_template_get) (gst_ghost_pad_new): Doc fixes. * gst/gstobject.c (gst_object_get_parent, gst_object_unparent) (gst_object_check_uniqueness): * gst/gstelement.c (gst_element_add_pad) (gst_element_add_ghost_pad, gst_element_remove_pad) (gst_element_remove_ghost_pad, gst_element_get_pad) (gst_element_get_static_pad, gst_element_get_pad_list) (gst_element_class_get_pad_template_list) (gst_element_class_get_pad_template): Work on the docs. (gst_element_get_pad_template_list): Uses the class method. (gst_element_get_compatible_pad_template): Docs, and consolidate some test conditions. (gst_element_get_pad_from_template): New static function. (gst_element_request_compatible_pad): Docs, and work with non-request compatible templates. (gst_element_get_compatible_pad_filtered): Docs and remove redundant checks. (gst_element_get_compatible_pad, gst_element_link_pads_filtered) (gst_element_link_filtered, gst_element_link_many) (gst_element_link, gst_element_link_pads) (gst_element_unlink_many): Docs. 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-11 13:26:04 +00:00
* Gets the capabilities for @pad's template.
*
gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. Original commit message from CVS: 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-05 13:51:26 +00:00
* Returns: the #GstCaps of this pad template. If you intend to keep a reference
* on the caps, make a copy (see gst_caps_copy ()).
*/
const GstCaps *
gst_pad_get_pad_template_caps (GstPad * pad)
{
static GstStaticCaps anycaps = GST_STATIC_CAPS ("ANY");
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
if (GST_PAD_PAD_TEMPLATE (pad))
return GST_PAD_TEMPLATE_CAPS (GST_PAD_PAD_TEMPLATE (pad));
return gst_static_caps_get (&anycaps);
}
/**
* gst_pad_get_peer:
* @pad: a #GstPad to get the peer of.
*
* Gets the peer of @pad. This function refs the peer pad so
* you need to unref it after use.
*
* Returns: the peer #GstPad. Unref after usage.
*
* MT safe.
*/
GstPad *
gst_pad_get_peer (GstPad * pad)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GstPad *result;
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
GST_LOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
result = GST_PAD_PEER (pad);
if (result)
gst_object_ref (result);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
return result;
}
/**
* gst_pad_get_allowed_caps:
* @srcpad: a #GstPad, it must a a source pad.
*
* Gets the capabilities of the allowed media types that can flow through
* @srcpad and its peer. The pad must be a source pad.
gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. Original commit message from CVS: 2004-02-24 Andy Wingo <wingo@pobox.com> * gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. * gst/schedulers/gstoptimalscheduler.c I added a mess of prototypes at the top of the file by way of documentation. Some of the operations on chains and groups were re-organized. (create_group): Added a type argument so if the group is enabled, the setup_group_scheduler knows what to do. (group_elements): Added a type argument here, too, to be passed on to create_group. (group_element_set_enabled): If an unlinked PLAYING element is added to a bin, we have to create a new group to hold the element, and this function will be called before the group is added to the chain. Thus we have a valid case for group->chain==NULL. Instead of calling chain_group_set_enabled, just set the flag on the group (the chain's status will be set when the group is added to it). (gst_opt_scheduler_state_transition, chain_group_set_enabled): Setup the group scheduler when the group is enabled, not specifically when an element goes PAUSED->PLAYING. This means PLAYING elements can be added, linked, and scheduled into a PLAYING pipeline, as was intended. (add_to_group): Don't ref the group twice. I don't know when this double-ref got in here. Removing it has the potential to cause segfaults if other parts of the scheduler are buggy. If you find that the scheduler is segfaulting for you, put in an extra ref here and see if that hacks over the underlying issue. Of course, then find out what code is unreffing a group it doesn't own... (create_group): Make the extra refcount floating, and remove it after adding the element. This means that... (unref_group): Destroy when the refcount reaches 0, not 1, like every other refcounted object in the known universe. (remove_from_group): When a group becomes empty, set it to be not active, and remove it from its chain. Don't unref it again, there's no floating reference any more. (destroy_group): We have to remove the group from the chain in remove_from_group (rather than here) to break refcounting cycles (the chain always has a ref on the group). So assert that group->chain==NULL. (ref_group_by_count): Removed, it was commented out anyway. (merge_chains): Use the remove_from_chain and add_to_chain primitives to do the reparenting, instead of rolling our own implementation. (add_to_chain): The first non-disabled group in the chain's group list will be the entry point for the chain. Because buffers can accumulate in loop elements' peer bufpens, we preferentially schedule loop groups before get groups to avoid unnecessary execution of get-based groups when the bufpens are already full. (gst_opt_scheduler_schedule_run_queue): Debug fixes. (get_group_schedule_function): Ditto. (loop_group_schedule_function): Ditto. (gst_opt_scheduler_loop_wrapper): Ditto. (gst_opt_scheduler_iterate): Ditto. I understand the opt scheduler now, yippee! * gst/gstpad.c: All throughout, added FIXMEs to look at for 0.9. (gst_pad_get_name, gst_pad_set_chain_function) (gst_pad_set_get_function, gst_pad_set_event_function) (gst_pad_set_event_mask_function, gst_pad_get_event_masks) (gst_pad_get_event_masks_default, gst_pad_set_convert_function) (gst_pad_set_query_function, gst_pad_get_query_types) (gst_pad_get_query_types_default) (gst_pad_set_internal_link_function) (gst_pad_set_formats_function, gst_pad_set_link_function) (gst_pad_set_fixate_function, gst_pad_set_getcaps_function) (gst_pad_set_bufferalloc_function, gst_pad_unlink) (gst_pad_renegotiate, gst_pad_set_parent, gst_pad_get_parent) (gst_pad_add_ghost_pad, gst_pad_proxy_getcaps) (gst_pad_proxy_pad_link, gst_pad_proxy_fixate) (gst_pad_get_pad_template_caps, gst_pad_check_compatibility) (gst_pad_get_peer, gst_pad_get_allowed_caps) (gst_pad_alloc_buffer, gst_pad_push, gst_pad_pull) (gst_pad_selectv, gst_pad_select, gst_pad_template_get_caps) (gst_pad_event_default_dispatch, gst_pad_event_default) (gst_pad_dispatcher, gst_pad_send_event, gst_pad_convert_default) (gst_pad_convert, gst_pad_query_default, gst_pad_query) (gst_pad_get_formats_default, gst_pad_get_formats): Better argument checks, and some doc fixes. (gst_pad_custom_new_from_template): Um, does anyone use these functions? Actually make a custom pad instead of a normal one. (gst_pad_try_set_caps): Transpose some checks. (gst_pad_try_set_caps_nonfixed): Same, and use a macro to check if the pad is in negotiation. (gst_pad_try_relink_filtered): Use pad_link_prepare. * gst/gstelement.c: Remove prototypes also defined in gstclock.h. * gst/gstelement.h: * gst/gstclock.h: Un-deprecate the old clocking API, as discussed on the list.
2004-02-25 13:16:12 +00:00
* The caller must free the resulting caps.
*
* Returns: the allowed #GstCaps of the pad link. Free the caps when
* you no longer need it. This function returns NULL when the @srcpad has no
* peer.
*
* MT safe.
*/
GstCaps *
gst_pad_get_allowed_caps (GstPad * srcpad)
{
GstCaps *mycaps;
GstCaps *caps;
GstCaps *peercaps;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GstPad *peer;
g_return_val_if_fail (GST_IS_PAD (srcpad), NULL);
g_return_val_if_fail (GST_PAD_IS_SRC (srcpad), NULL);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (srcpad);
peer = GST_PAD_PEER (srcpad);
if (G_UNLIKELY (peer == NULL))
goto no_peer;
GST_CAT_DEBUG (GST_CAT_PROPERTIES, "%s:%s: getting allowed caps",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (srcpad));
gst_object_ref (peer);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (srcpad);
mycaps = gst_pad_get_caps (srcpad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
peercaps = gst_pad_get_caps (peer);
gst_object_unref (peer);
caps = gst_caps_intersect (mycaps, peercaps);
gst_caps_unref (peercaps);
gst_caps_unref (mycaps);
GST_CAT_DEBUG (GST_CAT_CAPS, "allowed caps %" GST_PTR_FORMAT, caps);
return caps;
no_peer:
{
GST_CAT_DEBUG (GST_CAT_PROPERTIES, "%s:%s: no peer",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (srcpad));
GST_UNLOCK (srcpad);
return NULL;
}
}
gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. Original commit message from CVS: 2004-02-10 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. (gst_pad_get_event_masks, gst_pad_get_event_masks_default) (gst_pad_get_query_types, gst_pad_get_query_types_default): Routine docs. (gst_pad_set_link_function, gst_pad_set_fixate_function) (gst_pad_set_getcaps_function): Doc from Dave's negotation random doc. (gst_pad_unlink, gst_pad_is_linked): Docs. (gst_pad_renegotiate): A brief description of capsnego. (gst_pad_try_set_caps): Document. (gst_pad_try_set_caps_nonfixed): Document. (gst_pad_can_link_filtered, gst_pad_link_filtered): Doc fixes. (gst_pad_set_parent): Deprecated (although not out of the API). (gst_pad_get_parent): Deprecated, although many plugins use this. (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): Doc that these are private and will go away in 0.9. (gst_pad_perform_negotiate): Doc. (gst_pad_link_unnegotiate): I think this is meant to be static. (gst_pad_get_negotiated_caps, gst_pad_get_pad_template_caps) (gst_pad_template_get_caps_by_name, gst_pad_check_compatibility) (gst_pad_get_peer): Doc updates. (gst_pad_caps_change_notify): Doc. (gst_pad_alloc_buffer, gst_pad_push, gst_static_pad_template_get) (gst_ghost_pad_new): Doc fixes. * gst/gstobject.c (gst_object_get_parent, gst_object_unparent) (gst_object_check_uniqueness): * gst/gstelement.c (gst_element_add_pad) (gst_element_add_ghost_pad, gst_element_remove_pad) (gst_element_remove_ghost_pad, gst_element_get_pad) (gst_element_get_static_pad, gst_element_get_pad_list) (gst_element_class_get_pad_template_list) (gst_element_class_get_pad_template): Work on the docs. (gst_element_get_pad_template_list): Uses the class method. (gst_element_get_compatible_pad_template): Docs, and consolidate some test conditions. (gst_element_get_pad_from_template): New static function. (gst_element_request_compatible_pad): Docs, and work with non-request compatible templates. (gst_element_get_compatible_pad_filtered): Docs and remove redundant checks. (gst_element_get_compatible_pad, gst_element_link_pads_filtered) (gst_element_link_filtered, gst_element_link_many) (gst_element_link, gst_element_link_pads) (gst_element_unlink_many): Docs. 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-11 13:26:04 +00:00
/**
* gst_pad_get_negotiated_caps:
* @pad: a #GstPad.
*
* Gets the capabilities of the media type that currently flows through @pad
* and its peer.
*
* This function can be used on both src and sinkpads. Note that srcpads are
* always negotiated before sinkpads so it is possible that the negotiated caps
* on the srcpad do not match the negotiated caps of the peer.
*
* Returns: the negotiated #GstCaps of the pad link. Free the caps when
* you no longer need it. This function returns NULL when the @pad has no
* peer or is not negotiated yet.
gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. Original commit message from CVS: 2004-02-10 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. (gst_pad_get_event_masks, gst_pad_get_event_masks_default) (gst_pad_get_query_types, gst_pad_get_query_types_default): Routine docs. (gst_pad_set_link_function, gst_pad_set_fixate_function) (gst_pad_set_getcaps_function): Doc from Dave's negotation random doc. (gst_pad_unlink, gst_pad_is_linked): Docs. (gst_pad_renegotiate): A brief description of capsnego. (gst_pad_try_set_caps): Document. (gst_pad_try_set_caps_nonfixed): Document. (gst_pad_can_link_filtered, gst_pad_link_filtered): Doc fixes. (gst_pad_set_parent): Deprecated (although not out of the API). (gst_pad_get_parent): Deprecated, although many plugins use this. (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): Doc that these are private and will go away in 0.9. (gst_pad_perform_negotiate): Doc. (gst_pad_link_unnegotiate): I think this is meant to be static. (gst_pad_get_negotiated_caps, gst_pad_get_pad_template_caps) (gst_pad_template_get_caps_by_name, gst_pad_check_compatibility) (gst_pad_get_peer): Doc updates. (gst_pad_caps_change_notify): Doc. (gst_pad_alloc_buffer, gst_pad_push, gst_static_pad_template_get) (gst_ghost_pad_new): Doc fixes. * gst/gstobject.c (gst_object_get_parent, gst_object_unparent) (gst_object_check_uniqueness): * gst/gstelement.c (gst_element_add_pad) (gst_element_add_ghost_pad, gst_element_remove_pad) (gst_element_remove_ghost_pad, gst_element_get_pad) (gst_element_get_static_pad, gst_element_get_pad_list) (gst_element_class_get_pad_template_list) (gst_element_class_get_pad_template): Work on the docs. (gst_element_get_pad_template_list): Uses the class method. (gst_element_get_compatible_pad_template): Docs, and consolidate some test conditions. (gst_element_get_pad_from_template): New static function. (gst_element_request_compatible_pad): Docs, and work with non-request compatible templates. (gst_element_get_compatible_pad_filtered): Docs and remove redundant checks. (gst_element_get_compatible_pad, gst_element_link_pads_filtered) (gst_element_link_filtered, gst_element_link_many) (gst_element_link, gst_element_link_pads) (gst_element_unlink_many): Docs. 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-11 13:26:04 +00:00
*
* MT safe.
gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. Original commit message from CVS: 2004-02-10 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. (gst_pad_get_event_masks, gst_pad_get_event_masks_default) (gst_pad_get_query_types, gst_pad_get_query_types_default): Routine docs. (gst_pad_set_link_function, gst_pad_set_fixate_function) (gst_pad_set_getcaps_function): Doc from Dave's negotation random doc. (gst_pad_unlink, gst_pad_is_linked): Docs. (gst_pad_renegotiate): A brief description of capsnego. (gst_pad_try_set_caps): Document. (gst_pad_try_set_caps_nonfixed): Document. (gst_pad_can_link_filtered, gst_pad_link_filtered): Doc fixes. (gst_pad_set_parent): Deprecated (although not out of the API). (gst_pad_get_parent): Deprecated, although many plugins use this. (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): Doc that these are private and will go away in 0.9. (gst_pad_perform_negotiate): Doc. (gst_pad_link_unnegotiate): I think this is meant to be static. (gst_pad_get_negotiated_caps, gst_pad_get_pad_template_caps) (gst_pad_template_get_caps_by_name, gst_pad_check_compatibility) (gst_pad_get_peer): Doc updates. (gst_pad_caps_change_notify): Doc. (gst_pad_alloc_buffer, gst_pad_push, gst_static_pad_template_get) (gst_ghost_pad_new): Doc fixes. * gst/gstobject.c (gst_object_get_parent, gst_object_unparent) (gst_object_check_uniqueness): * gst/gstelement.c (gst_element_add_pad) (gst_element_add_ghost_pad, gst_element_remove_pad) (gst_element_remove_ghost_pad, gst_element_get_pad) (gst_element_get_static_pad, gst_element_get_pad_list) (gst_element_class_get_pad_template_list) (gst_element_class_get_pad_template): Work on the docs. (gst_element_get_pad_template_list): Uses the class method. (gst_element_get_compatible_pad_template): Docs, and consolidate some test conditions. (gst_element_get_pad_from_template): New static function. (gst_element_request_compatible_pad): Docs, and work with non-request compatible templates. (gst_element_get_compatible_pad_filtered): Docs and remove redundant checks. (gst_element_get_compatible_pad, gst_element_link_pads_filtered) (gst_element_link_filtered, gst_element_link_many) (gst_element_link, gst_element_link_pads) (gst_element_unlink_many): Docs. 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-11 13:26:04 +00:00
*/
GstCaps *
gst_pad_get_negotiated_caps (GstPad * pad)
{
GstCaps *caps;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GstPad *peer;
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY ((peer = GST_PAD_PEER (pad)) == NULL))
goto no_peer;
GST_CAT_DEBUG (GST_CAT_PROPERTIES, "%s:%s: getting negotiated caps",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
caps = GST_PAD_CAPS (pad);
if (caps)
gst_caps_ref (caps);
GST_UNLOCK (pad);
GST_CAT_DEBUG (GST_CAT_CAPS, "negotiated caps %" GST_PTR_FORMAT, caps);
return caps;
no_peer:
{
GST_CAT_DEBUG (GST_CAT_PROPERTIES, "%s:%s: no peer",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (pad));
GST_UNLOCK (pad);
return NULL;
}
}
/**
* gst_pad_alloc_buffer:
* @pad: a source #GstPad
* @offset: the offset of the new buffer in the stream
* @size: the size of the new buffer
* @caps: the caps of the new buffer
* @buf: a newly allocated buffer
*
gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. Original commit message from CVS: 2004-02-10 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (gst_pad_custom_new): Add a FIXME, this is a hacky way to do inheritance. (gst_pad_get_event_masks, gst_pad_get_event_masks_default) (gst_pad_get_query_types, gst_pad_get_query_types_default): Routine docs. (gst_pad_set_link_function, gst_pad_set_fixate_function) (gst_pad_set_getcaps_function): Doc from Dave's negotation random doc. (gst_pad_unlink, gst_pad_is_linked): Docs. (gst_pad_renegotiate): A brief description of capsnego. (gst_pad_try_set_caps): Document. (gst_pad_try_set_caps_nonfixed): Document. (gst_pad_can_link_filtered, gst_pad_link_filtered): Doc fixes. (gst_pad_set_parent): Deprecated (although not out of the API). (gst_pad_get_parent): Deprecated, although many plugins use this. (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): Doc that these are private and will go away in 0.9. (gst_pad_perform_negotiate): Doc. (gst_pad_link_unnegotiate): I think this is meant to be static. (gst_pad_get_negotiated_caps, gst_pad_get_pad_template_caps) (gst_pad_template_get_caps_by_name, gst_pad_check_compatibility) (gst_pad_get_peer): Doc updates. (gst_pad_caps_change_notify): Doc. (gst_pad_alloc_buffer, gst_pad_push, gst_static_pad_template_get) (gst_ghost_pad_new): Doc fixes. * gst/gstobject.c (gst_object_get_parent, gst_object_unparent) (gst_object_check_uniqueness): * gst/gstelement.c (gst_element_add_pad) (gst_element_add_ghost_pad, gst_element_remove_pad) (gst_element_remove_ghost_pad, gst_element_get_pad) (gst_element_get_static_pad, gst_element_get_pad_list) (gst_element_class_get_pad_template_list) (gst_element_class_get_pad_template): Work on the docs. (gst_element_get_pad_template_list): Uses the class method. (gst_element_get_compatible_pad_template): Docs, and consolidate some test conditions. (gst_element_get_pad_from_template): New static function. (gst_element_request_compatible_pad): Docs, and work with non-request compatible templates. (gst_element_get_compatible_pad_filtered): Docs and remove redundant checks. (gst_element_get_compatible_pad, gst_element_link_pads_filtered) (gst_element_link_filtered, gst_element_link_many) (gst_element_link, gst_element_link_pads) (gst_element_unlink_many): Docs. 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-11 13:26:04 +00:00
* Allocates a new, empty buffer optimized to push to pad @pad. This
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* function only works if @pad is a source pad and has a peer.
*
* You need to check the caps of the buffer after performing this
* function and renegotiate to the format if needed.
*
* A new, empty #GstBuffer will be put in the @buf argument.
*
* Returns: a result code indicating success of the operation. Any
* result code other than GST_FLOW_OK is an error and @buf should
* not be used.
* An error can occur if the pad is not connected or when the downstream
* peer elements cannot provide an acceptable buffer.
*
* MT safe.
*/
GstFlowReturn
gst_pad_alloc_buffer (GstPad * pad, guint64 offset, gint size, GstCaps * caps,
GstBuffer ** buf)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GstPad *peer;
GstFlowReturn ret;
GstPadBufferAllocFunction bufferallocfunc;
gboolean caps_changed;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_val_if_fail (GST_IS_PAD (pad), GST_FLOW_ERROR);
g_return_val_if_fail (GST_PAD_IS_SRC (pad), GST_FLOW_ERROR);
g_return_val_if_fail (buf != NULL, GST_FLOW_ERROR);
GST_LOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY ((peer = GST_PAD_PEER (pad)) == NULL))
goto no_peer;
gst_object_ref (peer);
GST_UNLOCK (pad);
if (G_LIKELY ((bufferallocfunc = peer->bufferallocfunc) == NULL))
goto fallback;
GST_LOCK (peer);
/* when the peer is flushing we cannot give a buffer */
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY (GST_PAD_IS_FLUSHING (peer)))
goto flushing;
GST_CAT_DEBUG (GST_CAT_PADS,
"calling bufferallocfunc &%s (@%p) of peer pad %s:%s",
GST_DEBUG_FUNCPTR_NAME (bufferallocfunc),
&bufferallocfunc, GST_DEBUG_PAD_NAME (peer));
GST_UNLOCK (peer);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
ret = bufferallocfunc (peer, offset, size, caps, buf);
if (G_UNLIKELY (ret != GST_FLOW_OK))
goto peer_error;
if (G_UNLIKELY (*buf == NULL))
goto fallback;
do_caps:
gst_object_unref (peer);
/* FIXME, move capnego this into a base class? */
caps = GST_BUFFER_CAPS (*buf);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
caps_changed = caps && caps != GST_PAD_CAPS (pad);
/* we got a new datatype on the pad, see if it can handle it */
if (G_UNLIKELY (caps_changed)) {
GST_DEBUG ("caps changed to %" GST_PTR_FORMAT, caps);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY (!gst_pad_configure_src (pad, caps)))
goto not_negotiated;
}
return ret;
no_peer:
{
/* pad has no peer */
GST_CAT_DEBUG (GST_CAT_PADS,
"%s:%s called bufferallocfunc but had no peer",
GST_DEBUG_PAD_NAME (pad));
GST_UNLOCK (pad);
return GST_FLOW_NOT_LINKED;
}
flushing:
{
/* peer was flushing */
GST_UNLOCK (peer);
gst_object_unref (peer);
GST_CAT_DEBUG (GST_CAT_PADS,
"%s:%s called bufferallocfunc but peer was flushing",
GST_DEBUG_PAD_NAME (pad));
return GST_FLOW_WRONG_STATE;
}
/* fallback case, allocate a buffer of our own, add pad caps. */
fallback:
{
GST_CAT_DEBUG (GST_CAT_PADS,
"%s:%s fallback buffer alloc", GST_DEBUG_PAD_NAME (pad));
*buf = gst_buffer_new_and_alloc (size);
gst_buffer_set_caps (*buf, caps);
ret = GST_FLOW_OK;
goto do_caps;
}
not_negotiated:
{
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad,
"alloc function retured unacceptable buffer");
return GST_FLOW_NOT_NEGOTIATED;
}
peer_error:
{
gst_object_unref (peer);
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad,
"alloc function retured error %d", ret);
return ret;
}
}
/**
* gst_pad_get_internal_links_default:
* @pad: the #GstPad to get the internal links of.
*
* Gets a list of pads to which the given pad is linked to
* inside of the parent element.
* This is the default handler, and thus returns a list of all of the
* pads inside the parent element with opposite direction.
* The caller must free this list after use.
*
* Returns: a newly allocated #GList of pads, or NULL if the pad has no parent.
*
* Not MT safe.
*/
GList *
gst_pad_get_internal_links_default (GstPad * pad)
{
GList *res = NULL;
GstElement *parent;
GList *parent_pads;
GstPadDirection direction;
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
direction = pad->direction;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
parent = GST_PAD_PARENT (pad);
if (!parent)
return NULL;
parent_pads = parent->pads;
while (parent_pads) {
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GstPad *parent_pad = GST_PAD_CAST (parent_pads->data);
if (parent_pad->direction != direction) {
res = g_list_prepend (res, parent_pad);
}
parent_pads = g_list_next (parent_pads);
}
return res;
}
/**
* gst_pad_get_internal_links:
* @pad: the #GstPad to get the internal links of.
*
* Gets a list of pads to which the given pad is linked to
* inside of the parent element.
* The caller must free this list after use.
*
* Returns: a newly allocated #GList of pads.
*
* Not MT safe.
*/
GList *
gst_pad_get_internal_links (GstPad * pad)
{
GList *res = NULL;
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (GST_PAD_INTLINKFUNC (pad))
res = GST_PAD_INTLINKFUNC (pad) (pad);
return res;
}
static gboolean
gst_pad_event_default_dispatch (GstPad * pad, GstEvent * event)
{
GList *orig, *pads;
gboolean result;
GST_INFO_OBJECT (pad, "Sending event %p to all internally linked pads",
event);
result = (GST_PAD_DIRECTION (pad) == GST_PAD_SINK);
orig = pads = gst_pad_get_internal_links (pad);
while (pads) {
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
GstPad *eventpad = GST_PAD_CAST (pads->data);
pads = g_list_next (pads);
/* for all of the internally-linked pads that are actually linked */
if (GST_PAD_IS_LINKED (eventpad)) {
if (GST_PAD_DIRECTION (eventpad) == GST_PAD_SRC) {
/* for each pad we send to, we should ref the event; it's up
* to downstream to unref again when handled. */
GST_LOG_OBJECT (pad, "Reffing and sending event %p to %s:%s", event,
GST_DEBUG_PAD_NAME (eventpad));
gst_event_ref (event);
gst_pad_push_event (eventpad, event);
} else {
/* we only send the event on one pad, multi-sinkpad elements
* should implement a handler */
GST_LOG_OBJECT (pad, "sending event %p to one sink pad %s:%s", event,
GST_DEBUG_PAD_NAME (eventpad));
result = gst_pad_push_event (eventpad, event);
goto done;
}
}
}
/* we handled the incoming event so we unref once */
GST_LOG_OBJECT (pad, "handled event %p, unreffing", event);
gst_event_unref (event);
done:
g_list_free (orig);
return result;
}
/**
* gst_pad_event_default:
* @pad: a #GstPad to call the default event handler on.
* @event: the #GstEvent to handle.
*
* Invokes the default event handler for the given pad. End-of-stream and
* discontinuity events are handled specially, and then the event is sent to all
* pads internally linked to @pad. Note that if there are many possible sink
* pads that are internally linked to @pad, only one will be sent an event.
* Multi-sinkpad elements should implement custom event handlers.
*
* Returns: TRUE if the event was sent succesfully.
*/
gboolean
gst_pad_event_default (GstPad * pad, GstEvent * event)
{
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
g_return_val_if_fail (event != NULL, FALSE);
switch (GST_EVENT_TYPE (event)) {
case GST_EVENT_EOS:
{
GST_DEBUG_OBJECT (pad, "pausing task because of eos");
gst_pad_pause_task (pad);
}
default:
break;
}
return gst_pad_event_default_dispatch (pad, event);
}
/**
* gst_pad_dispatcher:
* @pad: a #GstPad to dispatch.
* @dispatch: the #GstDispatcherFunction to call.
* @data: gpointer user data passed to the dispatcher function.
*
* Invokes the given dispatcher function on all pads that are
* internally linked to the given pad.
* The GstPadDispatcherFunction should return TRUE when no further pads
* need to be processed.
*
* Returns: TRUE if one of the dispatcher functions returned TRUE.
*/
gboolean
gst_pad_dispatcher (GstPad * pad, GstPadDispatcherFunction dispatch,
gpointer data)
{
gboolean res = FALSE;
GList *int_pads, *orig;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
g_return_val_if_fail (dispatch != NULL, FALSE);
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
orig = int_pads = gst_pad_get_internal_links (pad);
while (int_pads) {
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GstPad *int_pad = GST_PAD_CAST (int_pads->data);
GstPad *int_peer = GST_PAD_PEER (int_pad);
if (int_peer) {
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
res = dispatch (int_peer, data);
if (res)
break;
}
int_pads = g_list_next (int_pads);
}
g_list_free (orig);
return res;
}
/**
* gst_pad_query:
* @pad: a #GstPad to invoke the default query on.
* @query: the #GstQuery to perform.
*
* Dispatches a query to a pad. The query should have been allocated by the
* caller via one of the type-specific allocation functions in gstquery.h. The
* element is responsible for filling the query with an appropriate response,
* which should then be parsed with a type-specific query parsing function.
*
* Again, the caller is responsible for both the allocation and deallocation of
* the query structure.
*
* Returns: TRUE if the query could be performed.
*/
gboolean
gst_pad_query (GstPad * pad, GstQuery * query)
{
GstPadQueryFunction func;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
g_return_val_if_fail (GST_IS_QUERY (query), FALSE);
GST_DEBUG ("sending query %p to pad %s:%s", query, GST_DEBUG_PAD_NAME (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if ((func = GST_PAD_QUERYFUNC (pad)) == NULL)
goto no_func;
2004-04-04 02:36:12 +00:00
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
return func (pad, query);
2004-04-04 02:36:12 +00:00
no_func:
{
GST_DEBUG ("pad had no query function");
return FALSE;
2004-04-04 02:36:12 +00:00
}
}
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
gboolean
gst_pad_query_default (GstPad * pad, GstQuery * query)
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
{
switch (GST_QUERY_TYPE (query)) {
case GST_QUERY_POSITION:
case GST_QUERY_SEEKING:
case GST_QUERY_FORMATS:
case GST_QUERY_LATENCY:
case GST_QUERY_JITTER:
case GST_QUERY_RATE:
case GST_QUERY_CONVERT:
default:
return gst_pad_dispatcher
(pad, (GstPadDispatcherFunction) gst_pad_query, query);
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
}
}
#ifndef GST_DISABLE_LOADSAVE
/* FIXME: why isn't this on a GstElement ? */
/**
* gst_pad_load_and_link:
* @self: an #xmlNodePtr to read the description from.
* @parent: the #GstObject element that owns the pad.
*
* Reads the pad definition from the XML node and links the given pad
* in the element to a pad of an element up in the hierarchy.
*/
void
gst_pad_load_and_link (xmlNodePtr self, GstObject * parent)
{
xmlNodePtr field = self->xmlChildrenNode;
GstPad *pad = NULL, *targetpad;
gchar *peer = NULL;
gchar **split;
GstElement *target;
GstObject *grandparent;
gchar *name = NULL;
while (field) {
if (!strcmp ((char *) field->name, "name")) {
name = (gchar *) xmlNodeGetContent (field);
pad = gst_element_get_pad (GST_ELEMENT (parent), name);
g_free (name);
} else if (!strcmp ((char *) field->name, "peer")) {
peer = (gchar *) xmlNodeGetContent (field);
}
field = field->next;
}
g_return_if_fail (pad != NULL);
if (peer == NULL)
return;
split = g_strsplit (peer, ".", 2);
if (split[0] == NULL || split[1] == NULL) {
GST_CAT_DEBUG (GST_CAT_XML,
"Could not parse peer '%s' for pad %s:%s, leaving unlinked",
peer, GST_DEBUG_PAD_NAME (pad));
g_free (peer);
return;
}
g_free (peer);
g_return_if_fail (split[0] != NULL);
g_return_if_fail (split[1] != NULL);
grandparent = gst_object_get_parent (parent);
if (grandparent && GST_IS_BIN (grandparent)) {
target = gst_bin_get_by_name_recurse_up (GST_BIN (grandparent), split[0]);
} else
goto cleanup;
if (target == NULL)
goto cleanup;
targetpad = gst_element_get_pad (target, split[1]);
if (targetpad == NULL)
goto cleanup;
gst_pad_link (pad, targetpad);
cleanup:
g_strfreev (split);
}
/**
* gst_pad_save_thyself:
* @pad: a #GstPad to save.
* @parent: the parent #xmlNodePtr to save the description in.
*
* Saves the pad into an xml representation.
*
* Returns: the #xmlNodePtr representation of the pad.
*/
static xmlNodePtr
gst_pad_save_thyself (GstObject * object, xmlNodePtr parent)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GstPad *pad;
GstPad *peer;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_val_if_fail (GST_IS_PAD (object), NULL);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
pad = GST_PAD (object);
xmlNewChild (parent, NULL, (xmlChar *) "name",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
(xmlChar *) GST_PAD_NAME (pad));
if (GST_PAD_PEER (pad) != NULL) {
gchar *content;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
peer = GST_PAD_PEER (pad);
/* first check to see if the peer's parent's parent is the same */
/* we just save it off */
content = g_strdup_printf ("%s.%s",
GST_OBJECT_NAME (GST_PAD_PARENT (peer)), GST_PAD_NAME (peer));
xmlNewChild (parent, NULL, (xmlChar *) "peer", (xmlChar *) content);
g_free (content);
} else
xmlNewChild (parent, NULL, (xmlChar *) "peer", NULL);
return parent;
}
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
#if 0
/**
* gst_ghost_pad_save_thyself:
* @pad: a ghost #GstPad to save.
* @parent: the parent #xmlNodePtr to save the description in.
*
* Saves the ghost pad into an xml representation.
*
* Returns: the #xmlNodePtr representation of the pad.
*/
xmlNodePtr
gst_ghost_pad_save_thyself (GstPad * pad, xmlNodePtr parent)
{
xmlNodePtr self;
g_return_val_if_fail (GST_IS_GHOST_PAD (pad), NULL);
self = xmlNewChild (parent, NULL, (xmlChar *) "ghostpad", NULL);
xmlNewChild (self, NULL, (xmlChar *) "name", (xmlChar *) GST_PAD_NAME (pad));
xmlNewChild (self, NULL, (xmlChar *) "parent",
(xmlChar *) GST_OBJECT_NAME (GST_PAD_PARENT (pad)));
gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. Original commit message from CVS: 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-05 13:51:26 +00:00
/* FIXME FIXME FIXME! */
return self;
}
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
#endif /* 0 */
#endif /* GST_DISABLE_LOADSAVE */
gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. Original commit message from CVS: 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-05 13:51:26 +00:00
/*
* should be called with pad lock held
*
* MT safe.
*/
static void
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
handle_pad_block (GstPad * pad)
{
GstPadBlockCallback callback;
gpointer user_data;
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad,
"signal block taken on pad %s:%s", GST_DEBUG_PAD_NAME (pad));
/* need to grab extra ref for the callbacks */
gst_object_ref (pad);
callback = pad->block_callback;
if (callback) {
user_data = pad->block_data;
GST_UNLOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
callback (pad, TRUE, user_data);
GST_LOCK (pad);
} else {
GST_PAD_BLOCK_SIGNAL (pad);
}
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
while (GST_PAD_IS_BLOCKED (pad))
GST_PAD_BLOCK_WAIT (pad);
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad, "got unblocked");
gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. Original commit message from CVS: 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-05 13:51:26 +00:00
callback = pad->block_callback;
if (callback) {
user_data = pad->block_data;
GST_UNLOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
callback (pad, FALSE, user_data);
GST_LOCK (pad);
} else {
GST_PAD_BLOCK_SIGNAL (pad);
gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. Original commit message from CVS: 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-05 13:51:26 +00:00
}
gst_object_unref (pad);
gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. Original commit message from CVS: 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-05 13:51:26 +00:00
}
/**********************************************************************
* Data passing functions
*/
gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. Original commit message from CVS: 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstpad.c (_gst_real_pad_fixate_accumulator): s/pointer/boxed/. * gst/gstmarshal.list (VOID:BOXED, BOXED:BOXED): New marshallers. * gst/gstpad.c (gst_real_pad_class_init): Use a BOXED:BOXED marshaller for ::fixate, and VOID:BOXED for ::caps-nego-failed, with the type=GST_TYPE_CAPS. This allows language bindings to know what kind of data they're dealing with. * gst/gstcaps.c (_gst_caps_value_init): GBoxed values initialize to NULL when g_value_init is called. GstCaps, which rolls its own type implementation, now does the same instead of allocating empty caps. (_gst_caps_initialize, _gst_caps_collect_value, _gst_caps_lcopy_value): Provide collect_value and lcopy_value type table methods. This allows G_VALUE_COLLECT to work. 2004-02-05 Andy Wingo <wingo@pobox.com> * configure.ac: * testsuite/Makefile.am (SUBDIRS): * testsuite/ghostpads/Makefile.am: * testsuite/ghostpads/ghostpads.c: A new test for ghost pads. * gst/gstpad.c (gst_pad_add_ghost_pad, gst_pad_remove_ghost_pad): These two routines are the only ones that set GST_GPAD_REALPAD(gpad), the ghost pad list, and the ghost pad's pad template. They should be made static, depending on ABI needs. (gst_real_pad_dispose): Handle the case of ghost pads without a parent. Assert after dealing with ghost pads that the ghost pad list is empty. (gst_ghost_pad_class_init): New property added, ::real-pad. Can be set after creation. (gst_ghost_pad_dispose): Set ::real-pad to NULL. (gst_ghost_pad_set_property, gst_ghost_pad_get_property): New functions. set_property will call add_ghost_pad/remove_ghost_pad as appropriate. (gst_ghost_pad_new): All the work is offloaded to g_object_new. * gst/gstelement.c (gst_element_add_pad): Handle ghost pads as well. (gst_element_add_ghost_pad): Remove code duplicated from _add_pad. (gst_element_remove_pad): Handle ghost pads as well. (gst_element_remove_ghost_pad): Deprecated (could be removed, depending on API-stability needs). 2004-02-05 Andy Wingo <wingo@pobox.com> * gst/gstbin.[ch]: (gst_bin_get_by_interface): GTypes are scalars, of course they're const
2004-02-05 13:51:26 +00:00
static gboolean
gst_pad_emit_have_data_signal (GstPad * pad, GstMiniObject * obj)
{
GValue ret = { 0 };
GValue args[2] = { {0}, {0} };
gboolean res;
/* init */
g_value_init (&ret, G_TYPE_BOOLEAN);
g_value_set_boolean (&ret, TRUE);
g_value_init (&args[0], GST_TYPE_PAD);
g_value_set_object (&args[0], pad);
g_value_init (&args[1], GST_TYPE_MINI_OBJECT); // G_TYPE_POINTER);
gst_value_set_mini_object (&args[1], obj);
/* actually emit */
g_signal_emitv (args, gst_pad_signals[PAD_HAVE_DATA], 0, &ret);
res = g_value_get_boolean (&ret);
/* clean up */
g_value_unset (&ret);
g_value_unset (&args[0]);
g_value_unset (&args[1]);
return res;
}
/**
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
* gst_pad_chain:
* @pad: a sink #GstPad.
* @buffer: the #GstBuffer to send.
*
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* Chain a buffer to @pad.
*
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
* Returns: a #GstFlowReturn from the pad.
*
* MT safe.
*/
GstFlowReturn
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
gst_pad_chain (GstPad * pad, GstBuffer * buffer)
{
GstCaps *caps;
gboolean caps_changed;
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
GstPadChainFunction chainfunc;
GstFlowReturn ret;
gboolean emit_signal;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_val_if_fail (GST_IS_PAD (pad), GST_FLOW_ERROR);
g_return_val_if_fail (GST_PAD_DIRECTION (pad) == GST_PAD_SINK,
GST_FLOW_ERROR);
g_return_val_if_fail (buffer != NULL, GST_FLOW_ERROR);
g_return_val_if_fail (GST_IS_BUFFER (buffer), GST_FLOW_ERROR);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
GST_STREAM_LOCK (pad);
GST_LOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY (GST_PAD_IS_FLUSHING (pad)))
goto flushing;
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
caps = GST_BUFFER_CAPS (buffer);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
caps_changed = caps && caps != GST_PAD_CAPS (pad);
emit_signal = GST_PAD_DO_BUFFER_SIGNALS (pad) > 0;
GST_UNLOCK (pad);
/* see if the signal should be emited, we emit before caps nego as
* we might drop the buffer and do capsnego for nothing. */
if (G_UNLIKELY (emit_signal)) {
if (!gst_pad_emit_have_data_signal (pad, GST_MINI_OBJECT (buffer)))
goto dropping;
}
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
/* we got a new datatype on the pad, see if it can handle it */
if (G_UNLIKELY (caps_changed)) {
GST_DEBUG ("caps changed to %" GST_PTR_FORMAT, caps);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
if (G_UNLIKELY (!gst_pad_configure_sink (pad, caps)))
goto not_negotiated;
}
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
/* NOTE: we read the chainfunc unlocked.
* we cannot hold the lock for the pad so we might send
* the data to the wrong function. This is not really a
* problem since functions are assigned at creation time
* and don't change that often... */
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY ((chainfunc = GST_PAD_CHAINFUNC (pad)) == NULL))
goto no_function;
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad,
"calling chainfunction &%s of pad %s:%s",
GST_DEBUG_FUNCPTR_NAME (chainfunc), GST_DEBUG_PAD_NAME (pad));
ret = chainfunc (pad, buffer);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
GST_STREAM_UNLOCK (pad);
return ret;
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
/* ERRORS */
flushing:
{
gst_buffer_unref (buffer);
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad,
"pushing, but pad was flushing");
GST_UNLOCK (pad);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
GST_STREAM_UNLOCK (pad);
return GST_FLOW_WRONG_STATE;
}
dropping:
{
gst_buffer_unref (buffer);
GST_DEBUG ("Dropping buffer due to FALSE probe return");
GST_STREAM_UNLOCK (pad);
return GST_FLOW_OK;
}
not_negotiated:
{
gst_buffer_unref (buffer);
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad,
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
"pushing buffer but pad did not accept");
GST_STREAM_UNLOCK (pad);
return GST_FLOW_NOT_NEGOTIATED;
}
no_function:
{
gst_buffer_unref (buffer);
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad,
"pushing, but not chainhandler");
GST_ELEMENT_ERROR (GST_PAD_PARENT (pad), CORE, PAD, (NULL),
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
("push on pad %s:%s but it has no chainfunction",
GST_DEBUG_PAD_NAME (pad)));
GST_STREAM_UNLOCK (pad);
return GST_FLOW_ERROR;
}
}
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
/**
* gst_pad_push:
* @pad: a source #GstPad.
* @buffer: the #GstBuffer to push.
*
* Pushes a buffer to the peer of @pad. @pad must be linked.
*
* Returns: a #GstFlowReturn from the peer pad.
*
* MT safe.
*/
GstFlowReturn
gst_pad_push (GstPad * pad, GstBuffer * buffer)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GstPad *peer;
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
GstFlowReturn ret;
gboolean emit_signal;
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_val_if_fail (GST_IS_PAD (pad), GST_FLOW_ERROR);
g_return_val_if_fail (GST_PAD_DIRECTION (pad) == GST_PAD_SRC, GST_FLOW_ERROR);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
g_return_val_if_fail (buffer != NULL, GST_FLOW_ERROR);
g_return_val_if_fail (GST_IS_BUFFER (buffer), GST_FLOW_ERROR);
GST_LOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
while (G_UNLIKELY (GST_PAD_IS_BLOCKED (pad)))
handle_pad_block (pad);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY ((peer = GST_PAD_PEER (pad)) == NULL))
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
goto not_linked;
/* we emit signals on the pad areg, the peer will have a chance to
* emit in the _chain() function */
emit_signal = GST_PAD_DO_BUFFER_SIGNALS (pad) > 0;
gst_object_ref (peer);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
GST_UNLOCK (pad);
if (G_UNLIKELY (emit_signal)) {
if (!gst_pad_emit_have_data_signal (pad, GST_MINI_OBJECT (buffer)))
goto dropping;
}
ret = gst_pad_chain (peer, buffer);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
gst_object_unref (peer);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
return ret;
/* ERROR recovery here */
not_linked:
{
gst_buffer_unref (buffer);
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad,
"pushing, but it was not linked");
GST_UNLOCK (pad);
return GST_FLOW_NOT_LINKED;
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
}
dropping:
{
gst_buffer_unref (buffer);
gst_object_unref (peer);
GST_DEBUG ("Dropping buffer due to FALSE probe return");
return GST_FLOW_OK;
}
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
}
/**
* gst_pad_check_pull_range:
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
* @pad: a sink #GstPad.
*
* Checks if a #gst_pad_pull_range() can be performed on the peer
* source pad. This function is used by plugins that want to check
* if they can use random access on the peer source pad.
*
* The peer sourcepad can implement a custom #GstPadCheckGetRangeFunction
* if it needs to perform some logic to determine if pull_range is
* possible.
gst/: Simplify pad activation. Original commit message from CVS: * gst/base/Makefile.am: * gst/base/README: * gst/base/gstbasesink.c: (gst_basesink_get_type), (gst_basesink_base_init), (gst_basesink_class_init), (gst_basesink_pad_getcaps), (gst_basesink_init), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesink.h: * gst/base/gstbasetransform.c: (gst_base_transform_get_type), (gst_base_transform_base_init), (gst_base_transform_finalize), (gst_base_transform_class_init), (gst_base_transform_init), (gst_base_transform_proxy_getcaps), (gst_base_transform_setcaps), (gst_base_transform_event), (gst_base_transform_getrange), (gst_base_transform_chain), (gst_base_transform_handle_buffer), (gst_base_transform_set_property), (gst_base_transform_get_property), (gst_base_transform_sink_activate), (gst_base_transform_src_activate), (gst_base_transform_change_state): * gst/base/gstbasetransform.h: * gst/elements/gstidentity.c: (gst_identity_finalize), (gst_identity_class_init), (gst_identity_init), (gst_identity_event), (gst_identity_check_perfect), (gst_identity_transform), (gst_identity_set_property), (gst_identity_get_property), (gst_identity_change_state): * gst/elements/gstidentity.h: * gst/gstelement.c: (gst_element_get_state_func), (gst_element_lost_state), (gst_element_pads_activate): * gst/gstpad.c: (gst_pad_set_active), (gst_pad_peer_set_active), (gst_pad_check_pull_range), (gst_pad_pull_range): * gst/gstpad.h: Simplify pad activation. Added function to check if pull_range can be performed. Error out when pulling inactive or flushing pads. Removed const from refcounted types as it does not make sense. Simplify pad templates in basesink Added base class for simple 1-to-1 transforms. Make identity subclass the base transform.
2005-03-29 16:18:12 +00:00
*
* Returns: a gboolean with the result.
*
* MT safe.
*/
gboolean
gst_pad_check_pull_range (GstPad * pad)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GstPad *peer;
gboolean ret;
GstPadCheckGetRangeFunction checkgetrangefunc;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
GST_LOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (GST_PAD_DIRECTION (pad) != GST_PAD_SINK)
goto wrong_direction;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY ((peer = GST_PAD_PEER (pad)) == NULL))
goto not_connected;
gst_object_ref (peer);
GST_UNLOCK (pad);
gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. Original commit message from CVS: 2004-02-24 Andy Wingo <wingo@pobox.com> * gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. * gst/schedulers/gstoptimalscheduler.c I added a mess of prototypes at the top of the file by way of documentation. Some of the operations on chains and groups were re-organized. (create_group): Added a type argument so if the group is enabled, the setup_group_scheduler knows what to do. (group_elements): Added a type argument here, too, to be passed on to create_group. (group_element_set_enabled): If an unlinked PLAYING element is added to a bin, we have to create a new group to hold the element, and this function will be called before the group is added to the chain. Thus we have a valid case for group->chain==NULL. Instead of calling chain_group_set_enabled, just set the flag on the group (the chain's status will be set when the group is added to it). (gst_opt_scheduler_state_transition, chain_group_set_enabled): Setup the group scheduler when the group is enabled, not specifically when an element goes PAUSED->PLAYING. This means PLAYING elements can be added, linked, and scheduled into a PLAYING pipeline, as was intended. (add_to_group): Don't ref the group twice. I don't know when this double-ref got in here. Removing it has the potential to cause segfaults if other parts of the scheduler are buggy. If you find that the scheduler is segfaulting for you, put in an extra ref here and see if that hacks over the underlying issue. Of course, then find out what code is unreffing a group it doesn't own... (create_group): Make the extra refcount floating, and remove it after adding the element. This means that... (unref_group): Destroy when the refcount reaches 0, not 1, like every other refcounted object in the known universe. (remove_from_group): When a group becomes empty, set it to be not active, and remove it from its chain. Don't unref it again, there's no floating reference any more. (destroy_group): We have to remove the group from the chain in remove_from_group (rather than here) to break refcounting cycles (the chain always has a ref on the group). So assert that group->chain==NULL. (ref_group_by_count): Removed, it was commented out anyway. (merge_chains): Use the remove_from_chain and add_to_chain primitives to do the reparenting, instead of rolling our own implementation. (add_to_chain): The first non-disabled group in the chain's group list will be the entry point for the chain. Because buffers can accumulate in loop elements' peer bufpens, we preferentially schedule loop groups before get groups to avoid unnecessary execution of get-based groups when the bufpens are already full. (gst_opt_scheduler_schedule_run_queue): Debug fixes. (get_group_schedule_function): Ditto. (loop_group_schedule_function): Ditto. (gst_opt_scheduler_loop_wrapper): Ditto. (gst_opt_scheduler_iterate): Ditto. I understand the opt scheduler now, yippee! * gst/gstpad.c: All throughout, added FIXMEs to look at for 0.9. (gst_pad_get_name, gst_pad_set_chain_function) (gst_pad_set_get_function, gst_pad_set_event_function) (gst_pad_set_event_mask_function, gst_pad_get_event_masks) (gst_pad_get_event_masks_default, gst_pad_set_convert_function) (gst_pad_set_query_function, gst_pad_get_query_types) (gst_pad_get_query_types_default) (gst_pad_set_internal_link_function) (gst_pad_set_formats_function, gst_pad_set_link_function) (gst_pad_set_fixate_function, gst_pad_set_getcaps_function) (gst_pad_set_bufferalloc_function, gst_pad_unlink) (gst_pad_renegotiate, gst_pad_set_parent, gst_pad_get_parent) (gst_pad_add_ghost_pad, gst_pad_proxy_getcaps) (gst_pad_proxy_pad_link, gst_pad_proxy_fixate) (gst_pad_get_pad_template_caps, gst_pad_check_compatibility) (gst_pad_get_peer, gst_pad_get_allowed_caps) (gst_pad_alloc_buffer, gst_pad_push, gst_pad_pull) (gst_pad_selectv, gst_pad_select, gst_pad_template_get_caps) (gst_pad_event_default_dispatch, gst_pad_event_default) (gst_pad_dispatcher, gst_pad_send_event, gst_pad_convert_default) (gst_pad_convert, gst_pad_query_default, gst_pad_query) (gst_pad_get_formats_default, gst_pad_get_formats): Better argument checks, and some doc fixes. (gst_pad_custom_new_from_template): Um, does anyone use these functions? Actually make a custom pad instead of a normal one. (gst_pad_try_set_caps): Transpose some checks. (gst_pad_try_set_caps_nonfixed): Same, and use a macro to check if the pad is in negotiation. (gst_pad_try_relink_filtered): Use pad_link_prepare. * gst/gstelement.c: Remove prototypes also defined in gstclock.h. * gst/gstelement.h: * gst/gstclock.h: Un-deprecate the old clocking API, as discussed on the list.
2004-02-25 13:16:12 +00:00
/* see note in above function */
if (G_LIKELY ((checkgetrangefunc = peer->checkgetrangefunc) == NULL)) {
/* FIXME, kindoff ghetto */
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
ret = GST_PAD_GETRANGEFUNC (peer) != NULL;
} else {
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad,
"calling checkgetrangefunc %s of peer pad %s:%s",
GST_DEBUG_FUNCPTR_NAME (checkgetrangefunc), GST_DEBUG_PAD_NAME (peer));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
ret = checkgetrangefunc (peer);
}
gst_object_unref (peer);
return ret;
/* ERROR recovery here */
wrong_direction:
{
GST_UNLOCK (pad);
return FALSE;
}
not_connected:
{
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad,
"checking pull range, but it was not linked");
GST_UNLOCK (pad);
return FALSE;
}
}
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
/**
* gst_pad_get_range:
* @pad: a src #GstPad.
* @buffer: a pointer to hold the #GstBuffer.
* @offset: The start offset of the buffer
* @length: The length of the buffer
*
* Calls the getrange function of @pad.
*
* Returns: a #GstFlowReturn from the pad.
*
* MT safe.
*/
GstFlowReturn
gst_pad_get_range (GstPad * pad, guint64 offset, guint size,
GstBuffer ** buffer)
{
GstFlowReturn ret;
GstPadGetRangeFunction getrangefunc;
gboolean emit_signal;
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_val_if_fail (GST_IS_PAD (pad), GST_FLOW_ERROR);
g_return_val_if_fail (GST_PAD_DIRECTION (pad) == GST_PAD_SRC, GST_FLOW_ERROR);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
g_return_val_if_fail (buffer != NULL, GST_FLOW_ERROR);
GST_STREAM_LOCK (pad);
GST_LOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY (GST_PAD_IS_FLUSHING (pad)))
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
goto flushing;
emit_signal = GST_PAD_DO_BUFFER_SIGNALS (pad) > 0;
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
GST_UNLOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY ((getrangefunc = GST_PAD_GETRANGEFUNC (pad)) == NULL))
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
goto no_function;
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad,
"calling getrangefunc %s of peer pad %s:%s, offset %"
G_GUINT64_FORMAT ", size %u",
GST_DEBUG_FUNCPTR_NAME (getrangefunc), GST_DEBUG_PAD_NAME (pad),
offset, size);
ret = getrangefunc (pad, offset, size, buffer);
/* can only fire the signal if we have a valid buffer */
if (G_UNLIKELY (emit_signal) && (ret == GST_FLOW_OK)) {
if (!gst_pad_emit_have_data_signal (pad, GST_MINI_OBJECT (*buffer)))
goto dropping;
}
GST_STREAM_UNLOCK (pad);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
return ret;
/* ERRORS */
flushing:
{
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad,
"pulling range, but pad was flushing");
GST_UNLOCK (pad);
GST_STREAM_UNLOCK (pad);
return GST_FLOW_WRONG_STATE;
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
}
no_function:
{
GST_ELEMENT_ERROR (GST_PAD_PARENT (pad), CORE, PAD, (NULL),
("pullrange on pad %s:%s but it has no getrangefunction",
GST_DEBUG_PAD_NAME (pad)));
GST_STREAM_UNLOCK (pad);
return GST_FLOW_ERROR;
}
dropping:
{
GST_DEBUG ("Dropping data after FALSE probe return");
GST_STREAM_UNLOCK (pad);
gst_buffer_unref (*buffer);
*buffer = NULL;
return GST_FLOW_UNEXPECTED;
}
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
}
/**
* gst_pad_pull_range:
* @pad: a sink #GstPad.
* @buffer: a pointer to hold the #GstBuffer.
* @offset: The start offset of the buffer
* @length: The length of the buffer
*
* Pulls a buffer from the peer pad. @pad must be linked.
*
* Returns: a #GstFlowReturn from the peer pad.
*
* MT safe.
*/
GstFlowReturn
gst_pad_pull_range (GstPad * pad, guint64 offset, guint size,
GstBuffer ** buffer)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GstPad *peer;
GstFlowReturn ret;
gboolean emit_signal;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_val_if_fail (GST_IS_PAD (pad), GST_FLOW_ERROR);
g_return_val_if_fail (GST_PAD_DIRECTION (pad) == GST_PAD_SINK,
GST_FLOW_ERROR);
g_return_val_if_fail (buffer != NULL, GST_FLOW_ERROR);
GST_LOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
while (G_UNLIKELY (GST_PAD_IS_BLOCKED (pad)))
handle_pad_block (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (G_UNLIKELY ((peer = GST_PAD_PEER (pad)) == NULL))
goto not_connected;
/* signal emision for the pad, peer has chance to emit when
* we call _get_range() */
emit_signal = GST_PAD_DO_BUFFER_SIGNALS (pad) > 0;
gst_object_ref (peer);
GST_UNLOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
ret = gst_pad_get_range (peer, offset, size, buffer);
gst_object_unref (peer);
/* can only fire the signal if we have a valid buffer */
if (G_UNLIKELY (emit_signal) && (ret == GST_FLOW_OK)) {
if (!gst_pad_emit_have_data_signal (pad, GST_MINI_OBJECT (*buffer)))
goto dropping;
}
return ret;
/* ERROR recovery here */
not_connected:
{
GST_CAT_LOG_OBJECT (GST_CAT_SCHEDULING, pad,
"pulling range, but it was not linked");
GST_UNLOCK (pad);
return GST_FLOW_NOT_LINKED;
}
dropping:
{
GST_DEBUG ("Dropping data after FALSE probe return");
gst_buffer_unref (*buffer);
*buffer = NULL;
return GST_FLOW_UNEXPECTED;
}
}
/**
* gst_pad_push_event:
* @pad: a #GstPad to push the event to.
* @event: the #GstEvent to send to the pad.
*
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
* Sends the event to the peer of the given pad. This function is
* mainly used by elements to send events to their peer
* elements.
*
* Returns: TRUE if the event was handled.
*
* MT safe.
*/
gboolean
gst_pad_push_event (GstPad * pad, GstEvent * event)
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GstPad *peerpad;
gboolean result;
gboolean emit_signal;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
g_return_val_if_fail (event != NULL, FALSE);
GST_LOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
peerpad = GST_PAD_PEER (pad);
if (peerpad == NULL)
goto not_linked;
emit_signal = GST_PAD_DO_EVENT_SIGNALS (pad) > 0;
gst_object_ref (peerpad);
GST_UNLOCK (pad);
if (G_UNLIKELY (emit_signal)) {
if (!gst_pad_emit_have_data_signal (pad, GST_MINI_OBJECT (event)))
goto dropping;
}
result = gst_pad_send_event (peerpad, event);
gst_object_unref (peerpad);
return result;
/* ERROR handling */
not_linked:
{
gst_event_unref (event);
GST_UNLOCK (pad);
return FALSE;
}
dropping:
{
GST_DEBUG ("Dropping event after FALSE probe return");
gst_object_unref (peerpad);
gst_event_unref (event);
return FALSE;
}
}
/**
* gst_pad_send_event:
* @pad: a #GstPad to send the event to.
* @event: the #GstEvent to send to the pad.
*
* Sends the event to the pad. This function can be used
* by applications to send events in the pipeline.
*
* Returns: TRUE if the event was handled.
*/
gboolean
gst_pad_send_event (GstPad * pad, GstEvent * event)
{
gboolean result = FALSE;
GstPadEventFunction eventfunc;
gboolean emit_signal;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. Original commit message from CVS: 2004-02-24 Andy Wingo <wingo@pobox.com> * gst/gstelement.c (gst_element_dispose): Protect against multiple invocations. * gst/schedulers/gstoptimalscheduler.c I added a mess of prototypes at the top of the file by way of documentation. Some of the operations on chains and groups were re-organized. (create_group): Added a type argument so if the group is enabled, the setup_group_scheduler knows what to do. (group_elements): Added a type argument here, too, to be passed on to create_group. (group_element_set_enabled): If an unlinked PLAYING element is added to a bin, we have to create a new group to hold the element, and this function will be called before the group is added to the chain. Thus we have a valid case for group->chain==NULL. Instead of calling chain_group_set_enabled, just set the flag on the group (the chain's status will be set when the group is added to it). (gst_opt_scheduler_state_transition, chain_group_set_enabled): Setup the group scheduler when the group is enabled, not specifically when an element goes PAUSED->PLAYING. This means PLAYING elements can be added, linked, and scheduled into a PLAYING pipeline, as was intended. (add_to_group): Don't ref the group twice. I don't know when this double-ref got in here. Removing it has the potential to cause segfaults if other parts of the scheduler are buggy. If you find that the scheduler is segfaulting for you, put in an extra ref here and see if that hacks over the underlying issue. Of course, then find out what code is unreffing a group it doesn't own... (create_group): Make the extra refcount floating, and remove it after adding the element. This means that... (unref_group): Destroy when the refcount reaches 0, not 1, like every other refcounted object in the known universe. (remove_from_group): When a group becomes empty, set it to be not active, and remove it from its chain. Don't unref it again, there's no floating reference any more. (destroy_group): We have to remove the group from the chain in remove_from_group (rather than here) to break refcounting cycles (the chain always has a ref on the group). So assert that group->chain==NULL. (ref_group_by_count): Removed, it was commented out anyway. (merge_chains): Use the remove_from_chain and add_to_chain primitives to do the reparenting, instead of rolling our own implementation. (add_to_chain): The first non-disabled group in the chain's group list will be the entry point for the chain. Because buffers can accumulate in loop elements' peer bufpens, we preferentially schedule loop groups before get groups to avoid unnecessary execution of get-based groups when the bufpens are already full. (gst_opt_scheduler_schedule_run_queue): Debug fixes. (get_group_schedule_function): Ditto. (loop_group_schedule_function): Ditto. (gst_opt_scheduler_loop_wrapper): Ditto. (gst_opt_scheduler_iterate): Ditto. I understand the opt scheduler now, yippee! * gst/gstpad.c: All throughout, added FIXMEs to look at for 0.9. (gst_pad_get_name, gst_pad_set_chain_function) (gst_pad_set_get_function, gst_pad_set_event_function) (gst_pad_set_event_mask_function, gst_pad_get_event_masks) (gst_pad_get_event_masks_default, gst_pad_set_convert_function) (gst_pad_set_query_function, gst_pad_get_query_types) (gst_pad_get_query_types_default) (gst_pad_set_internal_link_function) (gst_pad_set_formats_function, gst_pad_set_link_function) (gst_pad_set_fixate_function, gst_pad_set_getcaps_function) (gst_pad_set_bufferalloc_function, gst_pad_unlink) (gst_pad_renegotiate, gst_pad_set_parent, gst_pad_get_parent) (gst_pad_add_ghost_pad, gst_pad_proxy_getcaps) (gst_pad_proxy_pad_link, gst_pad_proxy_fixate) (gst_pad_get_pad_template_caps, gst_pad_check_compatibility) (gst_pad_get_peer, gst_pad_get_allowed_caps) (gst_pad_alloc_buffer, gst_pad_push, gst_pad_pull) (gst_pad_selectv, gst_pad_select, gst_pad_template_get_caps) (gst_pad_event_default_dispatch, gst_pad_event_default) (gst_pad_dispatcher, gst_pad_send_event, gst_pad_convert_default) (gst_pad_convert, gst_pad_query_default, gst_pad_query) (gst_pad_get_formats_default, gst_pad_get_formats): Better argument checks, and some doc fixes. (gst_pad_custom_new_from_template): Um, does anyone use these functions? Actually make a custom pad instead of a normal one. (gst_pad_try_set_caps): Transpose some checks. (gst_pad_try_set_caps_nonfixed): Same, and use a macro to check if the pad is in negotiation. (gst_pad_try_relink_filtered): Use pad_link_prepare. * gst/gstelement.c: Remove prototypes also defined in gstclock.h. * gst/gstelement.h: * gst/gstclock.h: Un-deprecate the old clocking API, as discussed on the list.
2004-02-25 13:16:12 +00:00
g_return_val_if_fail (event != NULL, FALSE);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_LOCK (pad);
if (GST_EVENT_SRC (event) == NULL)
GST_EVENT_SRC (event) = gst_object_ref (pad);
switch (GST_EVENT_TYPE (event)) {
case GST_EVENT_FLUSH:
GST_CAT_DEBUG (GST_CAT_EVENT, "have event type %d (FLUSH) on pad %s:%s",
GST_EVENT_TYPE (event), GST_DEBUG_PAD_NAME (pad));
if (GST_EVENT_FLUSH_DONE (event)) {
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_UNSET_FLUSHING (pad);
GST_CAT_DEBUG (GST_CAT_EVENT, "cleared flush flag");
} else {
/* can't even accept a flush begin event when flushing */
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (GST_PAD_IS_FLUSHING (pad))
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
goto flushing;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_SET_FLUSHING (pad);
GST_CAT_DEBUG (GST_CAT_EVENT, "set flush flag");
}
break;
default:
GST_CAT_DEBUG (GST_CAT_EVENT, "have event type %d on pad %s:%s",
GST_EVENT_TYPE (event), GST_DEBUG_PAD_NAME (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if (GST_PAD_IS_FLUSHING (pad))
goto flushing;
break;
}
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
if ((eventfunc = GST_PAD_EVENTFUNC (pad)) == NULL)
goto no_function;
emit_signal = GST_PAD_DO_EVENT_SIGNALS (pad) > 0;
gst_object_ref (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (pad);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
if (G_UNLIKELY (emit_signal)) {
if (!gst_pad_emit_have_data_signal (pad, GST_MINI_OBJECT (event)))
goto dropping;
}
result = eventfunc (GST_PAD_CAST (pad), event);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
gst_object_unref (pad);
return result;
/* ERROR handling */
no_function:
{
g_warning ("pad %s:%s has no event handler, file a bug.",
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_DEBUG_PAD_NAME (pad));
GST_UNLOCK (pad);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
gst_event_unref (event);
return FALSE;
}
flushing:
{
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_UNLOCK (pad);
GST_CAT_DEBUG (GST_CAT_EVENT, "Received event on flushing pad. Discarding");
gst_event_unref (event);
return FALSE;
}
dropping:
{
GST_DEBUG ("Dropping event after FALSE probe return");
gst_object_unref (pad);
gst_event_unref (event);
return FALSE;
}
}
/************************************************************************
*
* templates
*
*/
static void gst_pad_template_class_init (GstPadTemplateClass * klass);
static void gst_pad_template_init (GstPadTemplate * templ);
static void gst_pad_template_dispose (GObject * object);
GType
gst_pad_template_get_type (void)
{
static GType padtemplate_type = 0;
if (!padtemplate_type) {
static const GTypeInfo padtemplate_info = {
sizeof (GstPadTemplateClass), NULL, NULL,
(GClassInitFunc) gst_pad_template_class_init, NULL, NULL,
sizeof (GstPadTemplate),
0,
(GInstanceInitFunc) gst_pad_template_init, NULL
};
padtemplate_type =
g_type_register_static (GST_TYPE_OBJECT, "GstPadTemplate",
&padtemplate_info, 0);
}
return padtemplate_type;
}
static void
gst_pad_template_class_init (GstPadTemplateClass * klass)
{
GObjectClass *gobject_class;
GstObjectClass *gstobject_class;
gobject_class = (GObjectClass *) klass;
gstobject_class = (GstObjectClass *) klass;
padtemplate_parent_class = g_type_class_ref (GST_TYPE_OBJECT);
gst_pad_template_signals[TEMPL_PAD_CREATED] =
g_signal_new ("pad-created", G_TYPE_FROM_CLASS (klass), G_SIGNAL_RUN_LAST,
G_STRUCT_OFFSET (GstPadTemplateClass, pad_created),
NULL, NULL, gst_marshal_VOID__OBJECT, G_TYPE_NONE, 1, GST_TYPE_PAD);
gobject_class->dispose = gst_pad_template_dispose;
gstobject_class->path_string_separator = "*";
}
static void
gst_pad_template_init (GstPadTemplate * templ)
{
}
static void
gst_pad_template_dispose (GObject * object)
{
GstPadTemplate *templ = GST_PAD_TEMPLATE (object);
g_free (GST_PAD_TEMPLATE_NAME_TEMPLATE (templ));
if (GST_PAD_TEMPLATE_CAPS (templ)) {
gst_caps_unref (GST_PAD_TEMPLATE_CAPS (templ));
}
G_OBJECT_CLASS (padtemplate_parent_class)->dispose (object);
}
/* ALWAYS padtemplates cannot have conversion specifications, it doesn't make
* sense.
* SOMETIMES padtemplates can do whatever they want, they are provided by the
* element.
* REQUEST padtemplates can be reverse-parsed (the user asks for 'sink1', the
* 'sink%d' template is automatically selected), so we need to restrict their
* naming.
*/
static gboolean
name_is_valid (const gchar * name, GstPadPresence presence)
{
const gchar *str;
if (presence == GST_PAD_ALWAYS) {
if (strchr (name, '%')) {
g_warning ("invalid name template %s: conversion specifications are not"
" allowed for GST_PAD_ALWAYS padtemplates", name);
return FALSE;
}
} else if (presence == GST_PAD_REQUEST) {
if ((str = strchr (name, '%')) && strchr (str + 1, '%')) {
g_warning ("invalid name template %s: only one conversion specification"
" allowed in GST_PAD_REQUEST padtemplate", name);
return FALSE;
}
if (str && (*(str + 1) != 's' && *(str + 1) != 'd')) {
g_warning ("invalid name template %s: conversion specification must be of"
" type '%%d' or '%%s' for GST_PAD_REQUEST padtemplate", name);
return FALSE;
}
if (str && (*(str + 2) != '\0')) {
g_warning ("invalid name template %s: conversion specification must"
" appear at the end of the GST_PAD_REQUEST padtemplate name", name);
return FALSE;
}
}
return TRUE;
}
/**
* gst_static_pad_template_get:
* @pad_template: the static pad template
*
* Converts a #GstStaticPadTemplate into a #GstPadTemplate.
*
* Returns: a new #GstPadTemplate.
*/
GstPadTemplate *
gst_static_pad_template_get (GstStaticPadTemplate * pad_template)
{
GstPadTemplate *new;
if (!name_is_valid (pad_template->name_template, pad_template->presence))
return NULL;
new = g_object_new (gst_pad_template_get_type (),
"name", pad_template->name_template, NULL);
GST_PAD_TEMPLATE_NAME_TEMPLATE (new) = g_strdup (pad_template->name_template);
GST_PAD_TEMPLATE_DIRECTION (new) = pad_template->direction;
GST_PAD_TEMPLATE_PRESENCE (new) = pad_template->presence;
GST_PAD_TEMPLATE_CAPS (new) =
gst_caps_copy (gst_static_caps_get (&pad_template->static_caps));
return new;
}
/**
* gst_pad_template_new:
* @name_template: the name template.
* @direction: the #GstPadDirection of the template.
* @presence: the #GstPadPresence of the pad.
* @caps: a #GstCaps set for the template. The caps are taken ownership of.
*
* Creates a new pad template with a name according to the given template
* and with the given arguments. This functions takes ownership of the provided
* caps, so be sure to not use them afterwards.
*
* Returns: a new #GstPadTemplate.
*/
GstPadTemplate *
gst_pad_template_new (const gchar * name_template,
GstPadDirection direction, GstPadPresence presence, GstCaps * caps)
{
GstPadTemplate *new;
g_return_val_if_fail (name_template != NULL, NULL);
g_return_val_if_fail (caps != NULL, NULL);
g_return_val_if_fail (direction == GST_PAD_SRC
|| direction == GST_PAD_SINK, NULL);
g_return_val_if_fail (presence == GST_PAD_ALWAYS
|| presence == GST_PAD_SOMETIMES || presence == GST_PAD_REQUEST, NULL);
if (!name_is_valid (name_template, presence))
return NULL;
new = g_object_new (gst_pad_template_get_type (),
"name", name_template, NULL);
GST_PAD_TEMPLATE_NAME_TEMPLATE (new) = g_strdup (name_template);
GST_PAD_TEMPLATE_DIRECTION (new) = direction;
GST_PAD_TEMPLATE_PRESENCE (new) = presence;
GST_PAD_TEMPLATE_CAPS (new) = caps;
return new;
}
/**
* gst_static_pad_template_get_caps:
* @templ: a #GstStaticPadTemplate to get capabilities of.
*
* Gets the capabilities of the static pad template.
*
* Returns: the #GstCaps of the static pad template. If you need to keep a
* reference to the caps, take a ref (see gst_caps_ref ()).
*/
GstCaps *
gst_static_pad_template_get_caps (GstStaticPadTemplate * templ)
{
g_return_val_if_fail (templ, NULL);
return (GstCaps *) gst_static_caps_get (&templ->static_caps);
}
/**
* gst_pad_template_get_caps:
* @templ: a #GstPadTemplate to get capabilities of.
*
* Gets the capabilities of the pad template.
*
* Returns: the #GstCaps of the pad template. If you need to keep a reference to
* the caps, take a ref (see gst_caps_ref ()).
*/
GstCaps *
gst_pad_template_get_caps (GstPadTemplate * templ)
{
g_return_val_if_fail (GST_IS_PAD_TEMPLATE (templ), NULL);
return GST_PAD_TEMPLATE_CAPS (templ);
}
/**
* gst_pad_set_element_private:
* @pad: the #GstPad to set the private data of.
* @priv: The private data to attach to the pad.
*
* Set the given private data gpointer on the pad.
* This function can only be used by the element that owns the pad.
*/
void
gst_pad_set_element_private (GstPad * pad, gpointer priv)
{
pad->element_private = priv;
}
/**
* gst_pad_get_element_private:
* @pad: the #GstPad to get the private data of.
*
* Gets the private data of a pad.
*
* Returns: a #gpointer to the private data.
*/
gpointer
gst_pad_get_element_private (GstPad * pad)
{
return pad->element_private;
}
/**
* gst_pad_start_task:
* @pad: the #GstPad to start the task of
* @func: the task function to call
* @data: data passed to the task function
*
* Starts a task that repeadedly calls @func with @data. This function
* is nostly used in the pad activation function to start the
* dataflow. This function will automatically acauire the STREAM_LOCK of
* the pad before calling @func.
*
* Returns: a TRUE if the task could be started.
*/
gboolean
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
gst_pad_start_task (GstPad * pad, GstTaskFunction func, gpointer data)
{
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
GstTask *task;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
g_return_val_if_fail (func != NULL, FALSE);
GST_LOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
task = GST_PAD_TASK (pad);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
if (task == NULL) {
task = gst_task_create (func, data);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
gst_task_set_lock (task, GST_STREAM_GET_LOCK (pad));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_TASK (pad) = task;
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
}
gst_task_start (task);
GST_UNLOCK (pad);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
return TRUE;
}
/**
* gst_pad_pause_task:
* @pad: the #GstPad to pause the task of
*
* Pause the task of @pad. This function will also make sure that the
* function executed by the task will effectively stop.
*
* Returns: a TRUE if the task could be paused or FALSE when the pad
* has no task.
*/
gboolean
gst_pad_pause_task (GstPad * pad)
{
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
GstTask *task;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
GST_LOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
task = GST_PAD_TASK (pad);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
if (task == NULL)
goto no_task;
gst_task_pause (task);
GST_UNLOCK (pad);
GST_STREAM_LOCK (pad);
GST_STREAM_UNLOCK (pad);
return TRUE;
no_task:
{
GST_UNLOCK (pad);
return TRUE;
}
}
/**
* gst_pad_stop_task:
* @pad: the #GstPad to stop the task of
*
* Stop the task of @pad. This function will also make sure that the
* function executed by the task will effectively stop.
*
* Returns: a TRUE if the task could be stopped or FALSE when the pad
* has no task.
*/
gst/gstquery.h Original commit message from CVS: 2005-05-06 Andy Wingo <wingo@pobox.com> * gst/gstquery.h * gst/gstquery.c (_gst_query_initialize): Extend GstQuery from GstData, init a memchunk. (standard_definitions): Add a few query types, deprecate a few. (gst_query_get_type): New proc. (_gst_query_copy, _gst_query_free, gst_query_new): GstData implementation. (gst_query_new_application, gst_query_get_structure): New public procs. * docs/design/draft-query.txt: Removed LINKS from the query types, because all the rest can be dispatched to other pads -- seemed ugly to have a query that couldn't be dispatched. internal_links is fine as a pad method. * gst/gstpad.h: Add query2 as a pad method, add the new functions in gstpad.c, but maintain binary compatibility for the moment. Will fix before 0.9 is out. * gst/gstqueryutils.c: * gst/gstqueryutils.h: New files, implement 3 methods for each query type: parse_query, parse_response, and set. Probably need an allocator as well. * gst/gst.h: Add gstquery.h and gstqueryutils.h to the list. * gst/elements/gstfilesink.c (gst_filesink_query2): * gst/base/gstbasesrc.c (gst_basesrc_query2): Replace old query, query_types, and formats methods. * gst/gstpad.c (gst_pad_query2, gst_pad_query2_default) (gst_pad_set_query2_function): New functions. (gst_real_pad_init): Set query2_default as the default query2 function. Basically just dispatches to internally linked pads. Needs review! * gst/gstdata_private.h (_GST_DATA_INIT): Set data->refcount to 1 without using the atomic operations. Only one thread can possibly be accessing the data at this point. Changed so as to avoid gst_atomic operations.
2005-05-06 21:41:22 +00:00
gboolean
gst_pad_stop_task (GstPad * pad)
gst/gstquery.h Original commit message from CVS: 2005-05-06 Andy Wingo <wingo@pobox.com> * gst/gstquery.h * gst/gstquery.c (_gst_query_initialize): Extend GstQuery from GstData, init a memchunk. (standard_definitions): Add a few query types, deprecate a few. (gst_query_get_type): New proc. (_gst_query_copy, _gst_query_free, gst_query_new): GstData implementation. (gst_query_new_application, gst_query_get_structure): New public procs. * docs/design/draft-query.txt: Removed LINKS from the query types, because all the rest can be dispatched to other pads -- seemed ugly to have a query that couldn't be dispatched. internal_links is fine as a pad method. * gst/gstpad.h: Add query2 as a pad method, add the new functions in gstpad.c, but maintain binary compatibility for the moment. Will fix before 0.9 is out. * gst/gstqueryutils.c: * gst/gstqueryutils.h: New files, implement 3 methods for each query type: parse_query, parse_response, and set. Probably need an allocator as well. * gst/gst.h: Add gstquery.h and gstqueryutils.h to the list. * gst/elements/gstfilesink.c (gst_filesink_query2): * gst/base/gstbasesrc.c (gst_basesrc_query2): Replace old query, query_types, and formats methods. * gst/gstpad.c (gst_pad_query2, gst_pad_query2_default) (gst_pad_set_query2_function): New functions. (gst_real_pad_init): Set query2_default as the default query2 function. Basically just dispatches to internally linked pads. Needs review! * gst/gstdata_private.h (_GST_DATA_INIT): Set data->refcount to 1 without using the atomic operations. Only one thread can possibly be accessing the data at this point. Changed so as to avoid gst_atomic operations.
2005-05-06 21:41:22 +00:00
{
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
GstTask *task;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
GST_LOCK (pad);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
task = GST_PAD_TASK (pad);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
if (task == NULL)
goto no_task;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
GST_PAD_TASK (pad) = NULL;
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
gst_task_stop (task);
GST_UNLOCK (pad);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
GST_STREAM_LOCK (pad);
GST_STREAM_UNLOCK (pad);
gst_object_unref (task);
gst/: Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Original commit message from CVS: * gst/base/gstadapter.c: (gst_adapter_peek), (gst_adapter_flush): * gst/base/gstbasesink.c: (gst_basesink_preroll_queue_push), (gst_basesink_finish_preroll), (gst_basesink_chain), (gst_basesink_loop), (gst_basesink_activate), (gst_basesink_change_state): * gst/base/gstbasesrc.c: (gst_basesrc_do_seek), (gst_basesrc_get_range), (gst_basesrc_loop), (gst_basesrc_activate): * gst/elements/gsttee.c: (gst_tee_sink_activate): * gst/gstpad.c: (gst_pad_dispose), (gst_real_pad_class_init), (gst_real_pad_init), (gst_real_pad_set_property), (gst_real_pad_get_property), (gst_pad_set_active), (gst_pad_is_active), (gst_pad_get_query_types), (gst_pad_unlink), (gst_pad_link_prepare), (gst_pad_link), (gst_pad_get_real_parent), (gst_real_pad_get_caps_unlocked), (gst_pad_peer_get_caps), (gst_pad_accept_caps), (gst_pad_get_peer), (gst_pad_realize), (gst_pad_event_default_dispatch), (gst_pad_event_default), (gst_pad_dispatcher), (gst_pad_query), (gst_real_pad_dispose), (gst_pad_save_thyself), (handle_pad_block), (gst_pad_chain), (gst_pad_push), (gst_pad_get_range), (gst_pad_pull_range), (gst_pad_send_event), (gst_pad_start_task), (gst_pad_pause_task), (gst_pad_stop_task): * gst/gstpad.h: * gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain), (gst_queue_loop), (gst_queue_src_activate): * gst/gsttask.c: (gst_task_init), (gst_task_set_lock), (gst_task_get_state): * gst/gsttask.h: * gst/schedulers/threadscheduler.c: (gst_thread_scheduler_task_start), (gst_thread_scheduler_func): Implement gst_pad_pause/start/stop_task(), take STREAM lock in task function. Remove ACTIVE pad flag, use FLUSHING everywhere Added _pad_chain(), _pad_get_range() to call chain/getrange functions. Add locks around IS_FLUSHING when reading. Take STREAM lock in chain(), get_range() functions so plugins don't need to take it anymore.
2005-05-25 11:50:11 +00:00
return TRUE;
no_task:
{
GST_UNLOCK (pad);
return TRUE;
}
}