gstreamer/gst/gstghostpad.c

1108 lines
29 KiB
C
Raw Normal View History

gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
/* GStreamer
* Copyright (C) 1999,2000 Erik Walthinsen <omega@cse.ogi.edu>
* 2000 Wim Taymans <wtay@chello.be>
* 2005 Andy Wingo <wingo@pobox.com>
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
* 2006 Edward Hervey <bilboed@bilboed.com>
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
*
* gstghostpad.c: Proxy pads
*
* 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.
*/
/**
* SECTION:gstghostpad
* @short_description: Pseudo link pads
* @see_also: #GstPad
*
* GhostPads are useful when organizing pipelines with #GstBin like elements.
* The idea here is to create hierarchical element graphs. The bin element
* contains a sub-graph. Now one would like to treat the bin-element like other
* #GstElements. This is where GhostPads come into play. A GhostPad acts as a
* proxy for another pad. Thus the bin can have sink and source ghost-pads that
* are associated with sink and source pads of the child elements.
*
* If the target pad is known at creation time, gst_ghost_pad_new() is the
* function to use to get a ghost-pad. Otherwise one can use gst_ghost_pad_new_no_target()
* to create the ghost-pad and use gst_ghost_pad_set_target() to establish the
* association later on.
*
* Note that GhostPads add overhead to the data processing of a pipeline.
*
* Last reviewed on 2005-11-18 (0.9.5)
*/
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into 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 "gst_private.h"
#include "gstghostpad.h"
#define GST_TYPE_PROXY_PAD (gst_proxy_pad_get_type ())
#define GST_IS_PROXY_PAD(obj) (G_TYPE_CHECK_INSTANCE_TYPE ((obj), GST_TYPE_PROXY_PAD))
#define GST_IS_PROXY_PAD_CLASS(klass) (G_TYPE_CHECK_CLASS_TYPE ((klass), GST_TYPE_PROXY_PAD))
#define GST_PROXY_PAD(obj) (G_TYPE_CHECK_INSTANCE_CAST ((obj), GST_TYPE_PROXY_PAD, GstProxyPad))
#define GST_PROXY_PAD_CLASS(klass) (G_TYPE_CHECK_CLASS_CAST ((klass), GST_TYPE_PROXY_PAD, GstProxyPadClass))
#define GST_PROXY_PAD_CAST(obj) ((GstProxyPad *)obj)
#define GST_PROXY_PAD_TARGET(pad) (GST_PROXY_PAD_CAST (pad)->target)
#define GST_PROXY_PAD_INTERNAL(pad) (GST_PROXY_PAD_CAST (pad)->internal)
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
typedef struct _GstProxyPad GstProxyPad;
typedef struct _GstProxyPadClass GstProxyPadClass;
#define GST_PROXY_GET_LOCK(pad) (GST_PROXY_PAD_CAST (pad)->proxy_lock)
#define GST_PROXY_LOCK(pad) (g_mutex_lock (GST_PROXY_GET_LOCK (pad)))
#define GST_PROXY_UNLOCK(pad) (g_mutex_unlock (GST_PROXY_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
struct _GstProxyPad
{
GstPad pad;
/* with PROXY_LOCK */
GMutex *proxy_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
GstPad *target;
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
GstPad *internal;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
};
struct _GstProxyPadClass
{
GstPadClass parent_class;
/*< private > */
gpointer _gst_reserved[1];
};
G_DEFINE_TYPE (GstProxyPad, gst_proxy_pad, GST_TYPE_PAD);
static GstPad *gst_proxy_pad_get_target (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
static void gst_proxy_pad_dispose (GObject * object);
static void gst_proxy_pad_finalize (GObject * object);
#ifndef GST_DISABLE_LOADSAVE
static xmlNodePtr gst_proxy_pad_save_thyself (GstObject * object,
xmlNodePtr parent);
#endif
static void
gst_proxy_pad_class_init (GstProxyPadClass * klass)
{
GObjectClass *gobject_class = (GObjectClass *) klass;
gobject_class->dispose = GST_DEBUG_FUNCPTR (gst_proxy_pad_dispose);
gobject_class->finalize = GST_DEBUG_FUNCPTR (gst_proxy_pad_finalize);
#ifndef GST_DISABLE_LOADSAVE
{
GstObjectClass *gstobject_class = (GstObjectClass *) klass;
gstobject_class->save_thyself =
GST_DEBUG_FUNCPTR (gst_proxy_pad_save_thyself);
}
#endif
}
const GstQueryType *
gst_proxy_pad_do_query_type (GstPad * pad)
{
GstPad *target = gst_proxy_pad_get_target (pad);
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
const GstQueryType *res = NULL;
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
if (target) {
res = gst_pad_get_query_types (target);
gst_object_unref (target);
}
return 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
}
static gboolean
gst_proxy_pad_do_event (GstPad * pad, GstEvent * event)
{
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
gboolean res = FALSE;
GstPad *internal = GST_PROXY_PAD_INTERNAL (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/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
res = gst_pad_push_event (internal, event);
return 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
}
static gboolean
gst_proxy_pad_do_query (GstPad * pad, GstQuery * query)
{
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
gboolean res = FALSE;
GstPad *target = gst_proxy_pad_get_target (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/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
if (target) {
res = gst_pad_query (target, query);
gst_object_unref (target);
}
return 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
}
static GList *
gst_proxy_pad_do_internal_link (GstPad * pad)
{
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
GList *res = NULL;
GstPad *target = gst_proxy_pad_get_target (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/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
if (target) {
res = gst_pad_get_internal_links (target);
gst_object_unref (target);
}
return 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
}
static GstFlowReturn
gst_proxy_pad_do_bufferalloc (GstPad * pad, guint64 offset, guint size,
GstCaps * caps, GstBuffer ** buf)
{
GstFlowReturn result;
GstPad *internal = GST_PROXY_PAD_INTERNAL (pad);
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
result = gst_pad_alloc_buffer (internal, offset, size, caps, buf);
return 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
}
static GstFlowReturn
gst_proxy_pad_do_chain (GstPad * pad, GstBuffer * buffer)
{
GstFlowReturn res;
GstPad *internal = GST_PROXY_PAD_INTERNAL (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/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
res = gst_pad_push (internal, buffer);
return 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
}
static GstFlowReturn
gst_proxy_pad_do_getrange (GstPad * pad, guint64 offset, guint size,
GstBuffer ** buffer)
{
GstFlowReturn res;
GstPad *internal = GST_PROXY_PAD_INTERNAL (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/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
res = gst_pad_pull_range (internal, offset, size, buffer);
return 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
}
static gboolean
gst_proxy_pad_do_checkgetrange (GstPad * pad)
{
gboolean result;
GstPad *internal = GST_PROXY_PAD_INTERNAL (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/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
result = gst_pad_check_pull_range (internal);
return 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
}
static GstCaps *
gst_proxy_pad_do_getcaps (GstPad * pad)
{
GstPad *target = gst_proxy_pad_get_target (pad);
GstCaps *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
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
if (target) {
/* if we have a real target, proxy the call */
GST_DEBUG_OBJECT (pad, "get caps of target");
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
res = gst_pad_get_caps (target);
gst_object_unref (target);
} else {
GstPadTemplate *templ = GST_PAD_PAD_TEMPLATE (pad);
/* else, if we have a template, use that */
if (templ) {
res = GST_PAD_TEMPLATE_CAPS (templ);
GST_DEBUG_OBJECT (pad,
"using pad template %p with caps %p %" GST_PTR_FORMAT, templ, res,
res);
res = gst_caps_ref (res);
goto done;
}
/* last resort, any caps */
GST_DEBUG_OBJECT (pad, "pad has no template, returning ANY");
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
res = gst_caps_new_any ();
}
done:
return 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
}
static gboolean
gst_proxy_pad_do_acceptcaps (GstPad * pad, GstCaps * caps)
{
GstPad *target = gst_proxy_pad_get_target (pad);
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
gboolean res = 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/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
if (target) {
res = gst_pad_accept_caps (target, caps);
gst_object_unref (target);
}
return 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
}
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
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_proxy_pad_do_fixatecaps (GstPad * pad, GstCaps * caps)
{
GstPad *target = gst_proxy_pad_get_target (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/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
if (target) {
gst_pad_fixate_caps (target, caps);
gst_object_unref (target);
}
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into 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 gboolean
gst_proxy_pad_do_setcaps (GstPad * pad, GstCaps * caps)
{
GstPad *target = gst_proxy_pad_get_target (pad);
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
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
if (target) {
res = gst_pad_set_caps (target, caps);
gst_object_unref (target);
} else {
/* We don't have any target, but we shouldn't return FALSE since this
* would stop the actual push of a buffer (which might trigger a pad block
* or probe, or properly return GST_FLOW_NOT_LINKED.
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
*/
res = TRUE;
}
return 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
}
static gboolean
gst_proxy_pad_set_target_unlocked (GstPad * pad, GstPad * target)
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into 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 *oldtarget;
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
if (target) {
GST_LOG_OBJECT (pad, "setting target %s:%s", GST_DEBUG_PAD_NAME (target));
if (G_UNLIKELY (GST_PAD_DIRECTION (pad) != GST_PAD_DIRECTION (target)))
goto wrong_direction;
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
} else
GST_LOG_OBJECT (pad, "clearing target");
/* clear old target */
if ((oldtarget = GST_PROXY_PAD_TARGET (pad))) {
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
gst_object_unref (oldtarget);
}
/* set and ref new target if any */
if (target)
GST_PROXY_PAD_TARGET (pad) = gst_object_ref (target);
else
GST_PROXY_PAD_TARGET (pad) = NULL;
return TRUE;
/* ERRORS */
wrong_direction:
{
GST_ERROR_OBJECT (pad,
"target pad doesn't have the same direction as ourself");
return FALSE;
}
}
static gboolean
gst_proxy_pad_set_target (GstPad * pad, GstPad * target)
{
gboolean result;
GST_PROXY_LOCK (pad);
result = gst_proxy_pad_set_target_unlocked (pad, target);
GST_PROXY_UNLOCK (pad);
return 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
}
static GstPad *
gst_proxy_pad_get_target (GstPad * pad)
{
GstPad *target;
GST_PROXY_LOCK (pad);
target = GST_PROXY_PAD_TARGET (pad);
if (target)
gst_object_ref (target);
GST_PROXY_UNLOCK (pad);
return target;
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into 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_proxy_pad_dispose (GObject * object)
{
GstPad *pad = GST_PAD (object);
GstPad **target_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_PROXY_LOCK (pad);
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
/* remove and unref the target */
target_p = &GST_PROXY_PAD_TARGET (pad);
gst_object_replace ((GstObject **) target_p, NULL);
/* The internal is only cleared by GstGhostPad::dispose, since it is the
* parent of non-ghost GstProxyPad and owns the refcount on the internal.
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
*/
GST_PROXY_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
G_OBJECT_CLASS (gst_proxy_pad_parent_class)->dispose (object);
}
static void
gst_proxy_pad_finalize (GObject * object)
{
GstProxyPad *pad = GST_PROXY_PAD (object);
g_mutex_free (pad->proxy_lock);
pad->proxy_lock = 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
G_OBJECT_CLASS (gst_proxy_pad_parent_class)->finalize (object);
}
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
static void
gst_proxy_pad_init (GstProxyPad * ppad)
{
GstPad *pad = (GstPad *) ppad;
ppad->proxy_lock = g_mutex_new ();
gst_pad_set_query_type_function (pad,
GST_DEBUG_FUNCPTR (gst_proxy_pad_do_query_type));
gst_pad_set_event_function (pad, GST_DEBUG_FUNCPTR (gst_proxy_pad_do_event));
gst_pad_set_query_function (pad, GST_DEBUG_FUNCPTR (gst_proxy_pad_do_query));
gst_pad_set_internal_link_function (pad,
GST_DEBUG_FUNCPTR (gst_proxy_pad_do_internal_link));
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
gst_pad_set_getcaps_function (pad,
GST_DEBUG_FUNCPTR (gst_proxy_pad_do_getcaps));
gst_pad_set_acceptcaps_function (pad,
GST_DEBUG_FUNCPTR (gst_proxy_pad_do_acceptcaps));
gst_pad_set_fixatecaps_function (pad,
GST_DEBUG_FUNCPTR (gst_proxy_pad_do_fixatecaps));
gst_pad_set_setcaps_function (pad,
GST_DEBUG_FUNCPTR (gst_proxy_pad_do_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
#ifndef GST_DISABLE_LOADSAVE
/**
* gst_proxy_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.
*/
static xmlNodePtr
gst_proxy_pad_save_thyself (GstObject * object, xmlNodePtr parent)
{
xmlNodePtr self;
g_return_val_if_fail (GST_IS_PROXY_PAD (object), NULL);
self = xmlNewChild (parent, NULL, (xmlChar *) "ghostpad", NULL);
xmlNewChild (self, NULL, (xmlChar *) "name",
(xmlChar *) GST_OBJECT_NAME (object));
xmlNewChild (self, NULL, (xmlChar *) "parent",
(xmlChar *) GST_OBJECT_NAME (GST_OBJECT_PARENT (object)));
/* FIXME FIXME FIXME! */
return self;
}
#endif /* GST_DISABLE_LOADSAVE */
/***********************************************************************
* Ghost pads, implemented as a pair of proxy pads (sort of)
*/
struct _GstGhostPad
{
GstProxyPad pad;
/* with PROXY_LOCK */
gulong notify_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
/*< private > */
gpointer _gst_reserved[GST_PADDING];
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
};
struct _GstGhostPadClass
{
GstProxyPadClass parent_class;
/*< private > */
gpointer _gst_reserved[GST_PADDING];
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into 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_DEFINE_TYPE (GstGhostPad, gst_ghost_pad, GST_TYPE_PROXY_PAD);
static void gst_ghost_pad_dispose (GObject * object);
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
/*
* The parent_set and parent_unset are used to make sure that:
* _ the internal and target are only linked when the GhostPad has a parent,
* _ the internal and target are unlinked as soon as the GhostPad is removed
* from it's parent.
*/
static void
gst_ghost_pad_parent_set (GstObject * object, GstObject * parent)
{
GstPad *gpad;
GstPad *target;
GstPad *internal;
GstPadLinkReturn ret;
GstObjectClass *klass;
gpad = GST_PAD (object);
/* internal is never NULL */
internal = GST_PROXY_PAD_INTERNAL (gpad);
target = gst_proxy_pad_get_target (gpad);
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
if (target) {
if (GST_PAD_IS_SRC (internal))
ret = gst_pad_link (internal, target);
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
else
ret = gst_pad_link (target, internal);
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
gst_object_unref (target);
if (ret != GST_PAD_LINK_OK)
goto link_failed;
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
}
klass = GST_OBJECT_CLASS (gst_ghost_pad_parent_class);
if (klass->parent_set)
klass->parent_set (object, parent);
return;
/* ERRORS */
link_failed:
{
/* a warning is all we can do */
GST_WARNING_OBJECT (gpad, "could not link internal ghostpad %s:%s",
GST_DEBUG_PAD_NAME (target));
g_warning ("could not link internal ghostpad %s:%s",
GST_DEBUG_PAD_NAME (target));
return;
}
}
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
static void
gst_ghost_pad_parent_unset (GstObject * object, GstObject * parent)
{
GstPad *gpad;
GstPad *target;
GstPad *internal;
GstObjectClass *klass;
gpad = GST_PAD (object);
internal = GST_PROXY_PAD_INTERNAL (gpad);
target = gst_proxy_pad_get_target (gpad);
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
if (target) {
if (GST_PAD_IS_SRC (internal))
gst_pad_unlink (internal, target);
else
gst_pad_unlink (target, internal);
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
gst_object_unref (target);
}
klass = GST_OBJECT_CLASS (gst_ghost_pad_parent_class);
if (klass->parent_unset)
klass->parent_unset (object, parent);
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +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
static void
gst_ghost_pad_class_init (GstGhostPadClass * klass)
{
GObjectClass *gobject_class = (GObjectClass *) klass;
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
GstObjectClass *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
gobject_class->dispose = GST_DEBUG_FUNCPTR (gst_ghost_pad_dispose);
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
gstobject_class->parent_set = GST_DEBUG_FUNCPTR (gst_ghost_pad_parent_set);
gstobject_class->parent_unset =
GST_DEBUG_FUNCPTR (gst_ghost_pad_parent_unset);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
}
/* see gstghostpad design docs */
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_ghost_pad_internal_do_activate_push (GstPad * pad, gboolean 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
{
gboolean ret;
GstPad *other;
GST_LOG_OBJECT (pad, "%sactivate push on %s:%s", (active ? "" : "de"),
GST_DEBUG_PAD_NAME (pad));
if (GST_PAD_DIRECTION (pad) == GST_PAD_SINK) {
other = GST_PROXY_PAD_INTERNAL (pad);
ret = gst_pad_activate_push (other, active);
} else if (G_LIKELY ((other = gst_pad_get_peer (pad)))) {
ret = gst_pad_activate_push (other, active);
gst_object_unref (other);
} else
ret = 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
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
static gboolean
gst_ghost_pad_internal_do_activate_pull (GstPad * pad, gboolean active)
{
gboolean ret;
GstPad *other;
GST_LOG_OBJECT (pad, "%sactivate pull on %s:%s", (active ? "" : "de"),
GST_DEBUG_PAD_NAME (pad));
if (GST_PAD_DIRECTION (pad) == GST_PAD_SRC) {
other = GST_PROXY_PAD_INTERNAL (pad);
ret = gst_pad_activate_pull (other, active);
} else if (G_LIKELY ((other = gst_pad_get_peer (pad)))) {
ret = gst_pad_activate_pull (other, active);
gst_object_unref (other);
} else
ret = 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
return ret;
}
static gboolean
gst_ghost_pad_do_activate_push (GstPad * pad, gboolean active)
{
gboolean ret;
GstPad *other;
GST_LOG_OBJECT (pad, "%sactivate push on %s:%s", (active ? "" : "de"),
GST_DEBUG_PAD_NAME (pad));
if (GST_PAD_DIRECTION (pad) == GST_PAD_SINK) {
other = GST_PROXY_PAD_INTERNAL (pad);
ret = gst_pad_activate_push (other, active);
} else
ret = TRUE;
return ret;
}
static gboolean
gst_ghost_pad_do_activate_pull (GstPad * pad, gboolean active)
{
gboolean ret;
GstPad *other;
GST_LOG_OBJECT (pad, "%sactivate pull on %s:%s", (active ? "" : "de"),
GST_DEBUG_PAD_NAME (pad));
if (GST_PAD_DIRECTION (pad) == GST_PAD_SRC) {
other = GST_PROXY_PAD_INTERNAL (pad);
ret = gst_pad_activate_pull (other, active);
} else if (G_LIKELY ((other = gst_pad_get_peer (pad)))) {
ret = gst_pad_activate_pull (other, active);
gst_object_unref (other);
} else
ret = FALSE;
return ret;
}
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into 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 GstPadLinkReturn
gst_ghost_pad_do_link (GstPad * pad, GstPad * peer)
{
GstPadLinkReturn ret;
GstPad *internal;
GST_DEBUG_OBJECT (pad, "linking ghostpad");
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into a bin creates ghost pads, and that the refcounts are right. Partly moved from gstbin.c.
2005-06-08 22:16:27 +00:00
internal = GST_PROXY_PAD_INTERNAL (pad);
if (!gst_proxy_pad_set_target (internal, peer))
goto target_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
ret = GST_PAD_LINK_OK;
/* if we are a source pad, we should call the peer link function
* if the peer has one */
if (GST_PAD_IS_SRC (pad)) {
if (GST_PAD_LINKFUNC (peer)) {
ret = GST_PAD_LINKFUNC (peer) (peer, pad);
if (ret != GST_PAD_LINK_OK)
goto link_failed;
}
}
return ret;
/* ERRORS */
target_failed:
{
GST_DEBUG_OBJECT (pad, "setting target failed");
return GST_PAD_LINK_REFUSED;
}
link_failed:
{
GST_DEBUG_OBJECT (pad, "linking failed");
/* clear target again */
gst_proxy_pad_set_target (internal, NULL);
return ret;
}
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into 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_ghost_pad_do_unlink (GstPad * pad)
{
GstPad *target;
GstPad *internal;
target = gst_proxy_pad_get_target (pad);
internal = GST_PROXY_PAD_INTERNAL (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_DEBUG_OBJECT (pad, "unlinking ghostpad");
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
/* The target of the internal pad is no longer valid */
gst_proxy_pad_set_target (internal, NULL);
if (target) {
if (GST_PAD_UNLINKFUNC (target))
GST_PAD_UNLINKFUNC (target) (target);
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into 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_object_unref (target);
}
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into 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
on_int_notify (GstPad * internal, GParamSpec * unused, GstGhostPad * pad)
{
GstCaps *caps;
g_object_get (internal, "caps", &caps, NULL);
GST_OBJECT_LOCK (pad);
gst_caps_replace (&(GST_PAD_CAPS (pad)), caps);
GST_OBJECT_UNLOCK (pad);
g_object_notify (G_OBJECT (pad), "caps");
if (caps)
gst_caps_unref (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
static void
gst_ghost_pad_init (GstGhostPad * pad)
{
gst_pad_set_activatepull_function (GST_PAD_CAST (pad),
GST_DEBUG_FUNCPTR (gst_ghost_pad_do_activate_pull));
gst_pad_set_activatepush_function (GST_PAD_CAST (pad),
GST_DEBUG_FUNCPTR (gst_ghost_pad_do_activate_push));
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into 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_ghost_pad_dispose (GObject * object)
{
GstPad *pad;
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
GstPad *internal;
GstPad *intpeer;
pad = GST_PAD (object);
GST_DEBUG_OBJECT (pad, "dispose");
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
GST_PROXY_LOCK (pad);
internal = GST_PROXY_PAD_INTERNAL (pad);
gst_pad_set_activatepull_function (internal, NULL);
gst_pad_set_activatepush_function (internal, NULL);
g_signal_handler_disconnect (internal, GST_GHOST_PAD_CAST (pad)->notify_id);
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
intpeer = gst_pad_get_peer (internal);
if (intpeer) {
if (GST_PAD_IS_SRC (internal))
gst_pad_unlink (internal, intpeer);
else
gst_pad_unlink (intpeer, internal);
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
gst_object_unref (intpeer);
}
GST_PROXY_PAD_INTERNAL (internal) = NULL;
/* disposes of the internal pad, since the ghostpad is the only possible object
* that has a refcount on the internal pad.
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
*/
gst_object_unparent (GST_OBJECT_CAST (internal));
GST_PROXY_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
G_OBJECT_CLASS (gst_ghost_pad_parent_class)->dispose (object);
}
static GstPad *
gst_ghost_pad_new_full (const gchar * name, GstPadDirection dir,
GstPadTemplate * templ)
{
GstPad *ret;
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
GstPad *internal;
g_return_val_if_fail (dir != GST_PAD_UNKNOWN, NULL);
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
/* OBJECT CREATION */
if (templ) {
ret = g_object_new (GST_TYPE_GHOST_PAD, "name", name,
"direction", dir, "template", templ, NULL);
} else {
ret = g_object_new (GST_TYPE_GHOST_PAD, "name", name,
"direction", dir, NULL);
}
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
/* Set directional padfunctions for ghostpad */
if (dir == GST_PAD_SINK) {
gst_pad_set_bufferalloc_function (ret,
GST_DEBUG_FUNCPTR (gst_proxy_pad_do_bufferalloc));
gst_pad_set_chain_function (ret,
GST_DEBUG_FUNCPTR (gst_proxy_pad_do_chain));
} else {
gst_pad_set_getrange_function (ret,
GST_DEBUG_FUNCPTR (gst_proxy_pad_do_getrange));
gst_pad_set_checkgetrange_function (ret,
GST_DEBUG_FUNCPTR (gst_proxy_pad_do_checkgetrange));
}
/* link/unlink functions */
gst_pad_set_link_function (ret, GST_DEBUG_FUNCPTR (gst_ghost_pad_do_link));
gst_pad_set_unlink_function (ret,
GST_DEBUG_FUNCPTR (gst_ghost_pad_do_unlink));
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
/* INTERNAL PAD, it always exists and is child of the ghostpad */
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
internal =
g_object_new (GST_TYPE_PROXY_PAD, "name", NULL,
"direction", (dir == GST_PAD_SRC) ? GST_PAD_SINK : GST_PAD_SRC, NULL);
/* Set directional padfunctions for internal pad */
if (dir == GST_PAD_SRC) {
gst_pad_set_bufferalloc_function (internal,
GST_DEBUG_FUNCPTR (gst_proxy_pad_do_bufferalloc));
gst_pad_set_chain_function (internal,
GST_DEBUG_FUNCPTR (gst_proxy_pad_do_chain));
} else {
gst_pad_set_getrange_function (internal,
GST_DEBUG_FUNCPTR (gst_proxy_pad_do_getrange));
gst_pad_set_checkgetrange_function (internal,
GST_DEBUG_FUNCPTR (gst_proxy_pad_do_checkgetrange));
}
GST_PROXY_LOCK (ret);
/* now make the ghostpad a parent of the internal pad */
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
if (!gst_object_set_parent (GST_OBJECT_CAST (internal),
GST_OBJECT_CAST (ret)))
goto parent_failed;
/* The ghostpad is the parent of the internal pad and is the only object that
* can have a refcount on the internal pad.
* At this point, the GstGhostPad has a refcount of 1, and the internal pad has
* a refcount of 1.
* When the refcount of the GstGhostPad drops to 0, the ghostpad will dispose
* it's refcount on the internal pad in the dispose method by un-parenting it.
* This is why we don't take extra refcounts in the assignments below
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
*/
GST_PROXY_PAD_INTERNAL (ret) = internal;
GST_PROXY_PAD_INTERNAL (internal) = ret;
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
/* could be more general here, iterating over all writable properties...
* taking the short road for now tho */
GST_GHOST_PAD_CAST (ret)->notify_id =
g_signal_connect (internal, "notify::caps", G_CALLBACK (on_int_notify),
ret);
/* call function to init values of the pad caps */
on_int_notify (internal, NULL, GST_GHOST_PAD_CAST (ret));
/* special activation functions for the internal pad */
gst_pad_set_activatepull_function (internal,
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
GST_DEBUG_FUNCPTR (gst_ghost_pad_internal_do_activate_pull));
gst_pad_set_activatepush_function (internal,
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
GST_DEBUG_FUNCPTR (gst_ghost_pad_internal_do_activate_push));
GST_PROXY_UNLOCK (ret);
return ret;
/* ERRORS */
parent_failed:
{
GST_WARNING_OBJECT (ret, "Could not set internal pad %s:%s",
GST_DEBUG_PAD_NAME (internal));
g_critical ("Could not set internal pad %s:%s",
GST_DEBUG_PAD_NAME (internal));
GST_PROXY_UNLOCK (ret);
gst_object_unref (ret);
gst_object_unref (internal);
return NULL;
}
}
/**
* gst_ghost_pad_new_no_target:
* @name: the name of the new pad, or NULL to assign a default name.
* @dir: the direction of the ghostpad
*
* Create a new ghostpad without a target with the given direction.
* A target can be set on the ghostpad later with the
* gst_ghost_pad_set_target() function.
*
* The created ghostpad will not have a padtemplate.
*
* Returns: a new #GstPad, or NULL in case of an error.
*/
GstPad *
gst_ghost_pad_new_no_target (const gchar * name, GstPadDirection dir)
{
GstPad *ret;
g_return_val_if_fail (dir != GST_PAD_UNKNOWN, NULL);
GST_LOG ("name:%s, direction:%d", GST_STR_NULL (name), dir);
ret = gst_ghost_pad_new_full (name, dir, NULL);
return ret;
}
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into 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_ghost_pad_new:
* @name: the name of the new pad, or NULL to assign a default name.
* @target: the pad to ghost.
*
* Create a new ghostpad with @target as the target. The direction will be taken
* from the target pad. @target must be 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
* Will ref the target.
*
* Returns: a new #GstPad, or NULL in case of an error.
*/
GstPad *
gst_ghost_pad_new (const gchar * name, GstPad * target)
{
GstPad *ret;
g_return_val_if_fail (GST_IS_PAD (target), NULL);
g_return_val_if_fail (!gst_pad_is_linked (target), 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_LOG ("name:%s, target:%s:%s", GST_STR_NULL (name),
GST_DEBUG_PAD_NAME (target));
if ((ret = gst_ghost_pad_new_no_target (name, GST_PAD_DIRECTION (target))))
if (!gst_ghost_pad_set_target (GST_GHOST_PAD_CAST (ret), target))
goto set_target_failed;
return ret;
/* ERRORS */
set_target_failed:
{
GST_WARNING_OBJECT (ret, "failed to set target %s:%s",
GST_DEBUG_PAD_NAME (target));
gst_object_unref (ret);
return NULL;
}
}
/**
* gst_ghost_pad_new_from_template:
* @name: the name of the new pad, or NULL to assign a default name.
* @target: the pad to ghost.
* @templ: the #GstPadTemplate to use on the ghostpad.
*
* Create a new ghostpad with @target as the target. The direction will be taken
* from the target pad. The template used on the ghostpad will be @template.
*
* Will ref the target.
*
* Returns: a new #GstPad, or NULL in case of an error.
*
* Since: 0.10.10
*/
GstPad *
gst_ghost_pad_new_from_template (const gchar * name, GstPad * target,
GstPadTemplate * templ)
{
GstPad *ret;
g_return_val_if_fail (GST_IS_PAD (target), NULL);
g_return_val_if_fail (!gst_pad_is_linked (target), NULL);
g_return_val_if_fail (templ != NULL, NULL);
g_return_val_if_fail (GST_PAD_TEMPLATE_DIRECTION (templ) ==
GST_PAD_DIRECTION (target), NULL);
GST_LOG ("name:%s, target:%s:%s, templ:%p", GST_STR_NULL (name),
GST_DEBUG_PAD_NAME (target), templ);
if ((ret = gst_ghost_pad_new_full (name, GST_PAD_DIRECTION (target), templ)))
if (!gst_ghost_pad_set_target (GST_GHOST_PAD_CAST (ret), target))
goto set_target_failed;
return ret;
/* ERRORS */
set_target_failed:
{
GST_WARNING_OBJECT (ret, "failed to set target %s:%s",
GST_DEBUG_PAD_NAME (target));
gst_object_unref (ret);
return NULL;
}
}
/**
* gst_ghost_pad_new_no_target_from_template:
* @name: the name of the new pad, or NULL to assign a default name.
* @templ: the #GstPadTemplate to create the ghostpad from.
*
* Create a new ghostpad based on @templ, without setting a target. The
* direction will be taken from the @templ.
*
* Returns: a new #GstPad, or NULL in case of an error.
*
* Since: 0.10.10
*/
GstPad *
gst_ghost_pad_new_no_target_from_template (const gchar * name,
GstPadTemplate * templ)
{
GstPad *ret;
g_return_val_if_fail (templ != NULL, NULL);
ret =
gst_ghost_pad_new_full (name, GST_PAD_TEMPLATE_DIRECTION (templ), templ);
return ret;
}
gst/gstutils.c: RPAD fixes all around. Original commit message from CVS: 2005-06-08 Andy Wingo <wingo@pobox.com> * gst/gstutils.c: RPAD fixes all around. (gst_element_link_pads): Refcounting fixes. * tools/gst-inspect.c: * tools/gst-xmlinspect.c: * parse/grammar.y: * gst/base/gsttypefindhelper.c: * gst/base/gstbasesink.c: * gst/gstqueue.c: RPAD fixes. * gst/gstghostpad.h: * gst/gstghostpad.c: New ghost pad implementation as full proxy pads. The tricky thing is they provide both source and sink interfaces, since they proxy the internal pad for the external pad, and vice versa. Implement with lower-level ProxyPad objects, with the interior proxy pad as a child of the exterior ghost pad. Should write a doc on this. * gst/gstpad.h: s/RPAD/PAD/, s/RealPad/Pad/. (gst_pad_set_name, gst_pad_set_parent): Macros removed, use gst_object API. * gst/gstpad.c: Big changes. No more stub base GstPad, now all pads are real pads. No ghost pads in this file. Not documenting the myriad s/RPAD/PAD/ and REALIZE fixes. (gst_pad_class_init): Add properties for "direction" and "template". Both are construct-only, so they can't change during the life of the pad. Fixes properly deriving from GstPad. (gst_pad_custom_new, gst_pad_custom_new_from_template): Gone. For derived objects, just set properties when creating the objects via g_object_new. (gst_pad_get_parent): Implement as a function, return NULL if the parent is not an element. (gst_pad_get_real_parent, gst_pad_add_ghost_pad) (gst_pad_remove_ghost_pad, gst_pad_realize): Removed. * gst/gstobject.c (gst_object_class_init): Make name a construct property. Don't set it in the object init. * gst/gstelement.c (gst_element_add_pad): Don't allow adding pads with UNKNOWN direction. (gst_element_add_ghost_pad): Remove non-orthogonal API. Replace with gst_element_add_pad (e, gst_ghost_pad_new (name, pad)). (gst_element_remove_pad): Remove ghost-pad special cases. (gst_element_pads_activate): Remove rpad cruft. * gst/gstbin.c (gst_bin_change_state): Use gst_pad_get_parent to catch the pad's-parent-not-an-element case. * gst/gst.h: Include gstghostpad.h. * gst/gst.c (init_post): No more real, ghost pads. * gst/Makefile.am: Add gstghostpad.[ch]. * check/Makefile.am: * check/gst/gstbin.c: * check/gst/gstghostpad.c (test_ghost_pads): Check that linking into 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_ghost_pad_get_target:
* @gpad: the #GstGhostpad
*
* Get the target pad of #gpad. Unref target pad after usage.
*
* Returns: the target #GstPad, can be NULL if the ghostpad
* has no target set. Unref target pad after usage.
*/
GstPad *
gst_ghost_pad_get_target (GstGhostPad * gpad)
{
GstPad *ret;
g_return_val_if_fail (GST_IS_GHOST_PAD (gpad), 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
ret = gst_proxy_pad_get_target (GST_PAD_CAST (gpad));
GST_DEBUG_OBJECT (gpad, "get target %s:%s", GST_DEBUG_PAD_NAME (ret));
return ret;
}
/**
* gst_ghost_pad_set_target:
* @gpad: the #GstGhostpad
* @newtarget: the new pad target
*
* Set the new target of the ghostpad @gpad. Any existing target
* is unlinked and links to the new target are established.
*
* Returns: TRUE if the new target could be set, FALSE otherwise.
*/
gboolean
gst_ghost_pad_set_target (GstGhostPad * gpad, GstPad * newtarget)
{
GstPad *internal;
GstPad *oldtarget;
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
GstObject *parent;
gboolean result;
GstPadLinkReturn lret;
g_return_val_if_fail (GST_IS_GHOST_PAD (gpad), FALSE);
GST_PROXY_LOCK (gpad);
internal = GST_PROXY_PAD_INTERNAL (gpad);
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
GST_DEBUG_OBJECT (gpad, "set target %s:%s", GST_DEBUG_PAD_NAME (newtarget));
/* clear old target */
if ((oldtarget = GST_PROXY_PAD_TARGET (gpad))) {
/* if we have an internal pad, unlink */
if (internal) {
if (GST_PAD_IS_SRC (internal))
gst_pad_unlink (internal, oldtarget);
else
gst_pad_unlink (oldtarget, internal);
}
}
result = gst_proxy_pad_set_target_unlocked (GST_PAD_CAST (gpad), newtarget);
if (result && newtarget) {
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
/* and link to internal pad if we are not unparent-ed */
if ((parent = gst_object_get_parent (GST_OBJECT (gpad)))) {
gst_object_unref (parent);
GST_DEBUG_OBJECT (gpad,
"GhostPad has parent, connecting internal pad to target");
if (GST_PAD_IS_SRC (internal))
lret = gst_pad_link (internal, newtarget);
else
lret = gst_pad_link (newtarget, internal);
if (lret != GST_PAD_LINK_OK)
goto link_failed;
gst/gstbin.c: (de)activate src pads before calling state_change on the childs. Original commit message from CVS: * gst/gstbin.c: (activate_pads), (iterator_activate_fold_with_resync), (gst_bin_src_pads_activate), (gst_bin_change_state_func): (de)activate src pads before calling state_change on the childs. This is to avoid the case where a src ghostpad is blocked (holding the stream lock), which would block the deactivation of the ghostpad's target pad. * gst/gstghostpad.c: (gst_proxy_pad_do_query_type), (gst_proxy_pad_do_event), (gst_proxy_pad_do_query), (gst_proxy_pad_do_internal_link), (gst_proxy_pad_do_bufferalloc), (gst_proxy_pad_do_chain), (gst_proxy_pad_do_getrange), (gst_proxy_pad_do_checkgetrange), (gst_proxy_pad_do_getcaps), (gst_proxy_pad_do_acceptcaps), (gst_proxy_pad_do_fixatecaps), (gst_proxy_pad_do_setcaps), (gst_proxy_pad_set_target_unlocked), (gst_proxy_pad_set_target), (gst_proxy_pad_get_internal), (gst_proxy_pad_dispose), (gst_proxy_pad_init), (gst_ghost_pad_parent_set), (gst_ghost_pad_parent_unset), (gst_ghost_pad_class_init), (gst_ghost_pad_internal_do_activate_push), (gst_ghost_pad_internal_do_activate_pull), (gst_ghost_pad_do_activate_push), (gst_ghost_pad_do_activate_pull), (gst_ghost_pad_do_link), (gst_ghost_pad_do_unlink), (gst_ghost_pad_dispose), (gst_ghost_pad_new_no_target), (gst_ghost_pad_new), (gst_ghost_pad_set_target): GhostPads now create their internal GstProxyPad at creation (and not when they're linked, as it was being done previously). The internal and target pads are linked straight away. The data will also travel through the other pad in order to make pad blocking and probes non-hackish (the probe/block now really happens on the GhostPad and not on the target). * gst/gstpad.c: (gst_pad_set_blocked_async), (gst_pad_link_prepare), (gst_pad_push_event): Remove previous ghostpad cruft. * gst/gstutils.c: (gst_pad_add_data_probe), (gst_pad_add_event_probe), (gst_pad_add_buffer_probe), (gst_pad_remove_data_probe), (gst_pad_remove_event_probe), (gst_pad_remove_buffer_probe): Remove previous ghost pad cruft. Added more detailed debug statements. * tests/check/gst/gstghostpad.c: (GST_START_TEST): Fix the testsuite for refcounting changes. The comments about who has references were correct, but the refcount being checked wasn't the same (!?!).
2006-07-11 16:20:09 +00:00
} else {
/* we need to connect the internal pad once we have a parent */
GST_DEBUG_OBJECT (gpad,
"GhostPad doesn't have a parent, will connect internal pad later");
}
}
GST_PROXY_UNLOCK (gpad);
return result;
/* ERRORS */
link_failed:
{
GST_WARNING_OBJECT (gpad, "could not link internal and target, reason:%d",
lret);
/* and unset target again */
gst_proxy_pad_set_target_unlocked (GST_PAD_CAST (gpad), NULL);
return 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
}