2000-01-30 09:03:00 +00:00
|
|
|
/* Gnome-Streamer
|
|
|
|
* Copyright (C) <1999> Erik Walthinsen <omega@cse.ogi.edu>
|
|
|
|
*
|
|
|
|
* 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.
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
2000-07-12 22:52:42 +00:00
|
|
|
//#define DEBUG_ENABLED
|
2000-12-15 01:57:34 +00:00
|
|
|
#include "gstdebug.h"
|
|
|
|
#include "gstpad.h"
|
|
|
|
#include "gstelement.h"
|
|
|
|
#include "gsttype.h"
|
2000-01-30 09:03:00 +00:00
|
|
|
|
|
|
|
/* Pad signals and args */
|
|
|
|
enum {
|
2000-11-25 07:02:55 +00:00
|
|
|
SET_ACTIVE,
|
2000-12-11 00:04:25 +00:00
|
|
|
CAPS_CHANGED,
|
2000-01-30 09:03:00 +00:00
|
|
|
/* FILL ME */
|
|
|
|
LAST_SIGNAL
|
|
|
|
};
|
|
|
|
|
|
|
|
enum {
|
|
|
|
ARG_0,
|
2000-11-25 07:02:55 +00:00
|
|
|
ARG_ACTIVE,
|
2000-01-30 09:03:00 +00:00
|
|
|
/* FILL ME */
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
|
|
static void gst_pad_class_init(GstPadClass *klass);
|
|
|
|
static void gst_pad_init(GstPad *pad);
|
2000-11-25 07:02:55 +00:00
|
|
|
|
|
|
|
static void gst_pad_set_arg(GtkObject *object,GtkArg *arg,guint id);
|
|
|
|
static void gst_pad_get_arg(GtkObject *object,GtkArg *arg,guint id);
|
|
|
|
|
2000-01-30 09:03:00 +00:00
|
|
|
static void gst_pad_real_destroy(GtkObject *object);
|
|
|
|
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
static void gst_pad_push_func(GstPad *pad, GstBuffer *buf);
|
|
|
|
|
2000-01-30 09:03:00 +00:00
|
|
|
|
|
|
|
static GstObject *parent_class = NULL;
|
2000-11-25 07:02:55 +00:00
|
|
|
static guint gst_pad_signals[LAST_SIGNAL] = { 0 };
|
2000-01-30 09:03:00 +00:00
|
|
|
|
|
|
|
GtkType
|
|
|
|
gst_pad_get_type(void) {
|
|
|
|
static GtkType pad_type = 0;
|
|
|
|
|
|
|
|
if (!pad_type) {
|
|
|
|
static const GtkTypeInfo pad_info = {
|
|
|
|
"GstPad",
|
|
|
|
sizeof(GstPad),
|
|
|
|
sizeof(GstPadClass),
|
|
|
|
(GtkClassInitFunc)gst_pad_class_init,
|
|
|
|
(GtkObjectInitFunc)gst_pad_init,
|
|
|
|
(GtkArgSetFunc)NULL,
|
|
|
|
(GtkArgGetFunc)NULL,
|
|
|
|
(GtkClassInitFunc)NULL,
|
|
|
|
};
|
|
|
|
pad_type = gtk_type_unique(GST_TYPE_OBJECT,&pad_info);
|
|
|
|
}
|
|
|
|
return pad_type;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
2000-11-01 13:49:41 +00:00
|
|
|
gst_pad_class_init (GstPadClass *klass)
|
|
|
|
{
|
2000-01-30 09:03:00 +00:00
|
|
|
GtkObjectClass *gtkobject_class;
|
|
|
|
|
|
|
|
gtkobject_class = (GtkObjectClass*)klass;
|
|
|
|
|
|
|
|
parent_class = gtk_type_class(GST_TYPE_OBJECT);
|
|
|
|
|
2000-11-25 07:02:55 +00:00
|
|
|
gst_pad_signals[SET_ACTIVE] =
|
|
|
|
gtk_signal_new ("set_active", GTK_RUN_LAST, gtkobject_class->type,
|
|
|
|
GTK_SIGNAL_OFFSET (GstPadClass, set_active),
|
|
|
|
gtk_marshal_NONE__BOOL, GTK_TYPE_NONE, 1,
|
|
|
|
GTK_TYPE_BOOL);
|
2000-12-11 00:04:25 +00:00
|
|
|
gst_pad_signals[CAPS_CHANGED] =
|
|
|
|
gtk_signal_new ("caps_changed", GTK_RUN_LAST, gtkobject_class->type,
|
|
|
|
GTK_SIGNAL_OFFSET (GstPadClass, caps_changed),
|
|
|
|
gtk_marshal_NONE__POINTER, GTK_TYPE_NONE, 1,
|
|
|
|
GTK_TYPE_POINTER);
|
2000-11-25 07:02:55 +00:00
|
|
|
|
|
|
|
gtk_object_add_arg_type ("GstPad::active", GTK_TYPE_BOOL,
|
|
|
|
GTK_ARG_READWRITE, ARG_ACTIVE);
|
|
|
|
|
2000-01-30 09:03:00 +00:00
|
|
|
gtkobject_class->destroy = gst_pad_real_destroy;
|
2000-11-25 07:02:55 +00:00
|
|
|
gtkobject_class->set_arg = gst_pad_set_arg;
|
|
|
|
gtkobject_class->get_arg = gst_pad_get_arg;
|
2000-01-30 09:03:00 +00:00
|
|
|
}
|
|
|
|
|
2000-11-01 13:49:41 +00:00
|
|
|
static void
|
|
|
|
gst_pad_init (GstPad *pad)
|
|
|
|
{
|
2000-01-30 09:03:00 +00:00
|
|
|
pad->direction = GST_PAD_UNKNOWN;
|
|
|
|
pad->peer = NULL;
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
|
2000-07-17 17:14:15 +00:00
|
|
|
pad->chainfunc = NULL;
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
pad->getfunc = NULL;
|
|
|
|
pad->getregionfunc = NULL;
|
|
|
|
pad->qosfunc = NULL;
|
|
|
|
|
|
|
|
pad->pushfunc = NULL; //GST_DEBUG_FUNCPTR(gst_pad_push_func);
|
2000-07-17 17:14:15 +00:00
|
|
|
pad->pullfunc = NULL;
|
2000-11-06 00:15:51 +00:00
|
|
|
pad->pullregionfunc = NULL;
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
|
2000-01-30 09:03:00 +00:00
|
|
|
pad->parent = NULL;
|
|
|
|
pad->ghostparents = NULL;
|
2000-12-11 00:04:25 +00:00
|
|
|
pad->caps = NULL;
|
2000-01-30 09:03:00 +00:00
|
|
|
}
|
|
|
|
|
2000-11-25 07:02:55 +00:00
|
|
|
static void
|
|
|
|
gst_pad_set_arg (GtkObject *object,GtkArg *arg,guint id) {
|
|
|
|
g_return_if_fail(GST_IS_PAD(object));
|
|
|
|
|
|
|
|
switch (id) {
|
|
|
|
case ARG_ACTIVE:
|
|
|
|
if (GTK_VALUE_BOOL(*arg)) {
|
|
|
|
gst_info("gstpad: activating pad\n");
|
|
|
|
GST_FLAG_UNSET(object,GST_PAD_DISABLED);
|
|
|
|
} else {
|
|
|
|
gst_info("gstpad: de-activating pad\n");
|
|
|
|
GST_FLAG_SET(object,GST_PAD_DISABLED);
|
|
|
|
}
|
|
|
|
gtk_signal_emit(GTK_OBJECT(object), gst_pad_signals[SET_ACTIVE],
|
|
|
|
! GST_FLAG_IS_SET(object,GST_PAD_DISABLED));
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
gst_pad_get_arg (GtkObject *object,
|
|
|
|
GtkArg *arg,
|
|
|
|
guint id)
|
|
|
|
{
|
|
|
|
/* it's not null if we got it, but it might not be ours */
|
|
|
|
g_return_if_fail (GST_IS_PAD (object));
|
|
|
|
|
|
|
|
switch (id) {
|
|
|
|
case ARG_ACTIVE:
|
|
|
|
GTK_VALUE_BOOL (*arg) = ! GST_FLAG_IS_SET (object, GST_PAD_DISABLED);
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2000-01-30 09:03:00 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_new:
|
|
|
|
* @name: name of new pad
|
|
|
|
* @direction: either GST_PAD_SRC or GST_PAD_SINK
|
|
|
|
*
|
|
|
|
* Create a new pad with given name.
|
|
|
|
*
|
|
|
|
* Returns: new pad
|
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
GstPad*
|
|
|
|
gst_pad_new (gchar *name,
|
|
|
|
GstPadDirection direction)
|
|
|
|
{
|
2000-01-30 09:03:00 +00:00
|
|
|
GstPad *pad;
|
|
|
|
|
2000-11-01 13:49:41 +00:00
|
|
|
g_return_val_if_fail (name != NULL, NULL);
|
|
|
|
g_return_val_if_fail (direction != GST_PAD_UNKNOWN, NULL);
|
2000-01-30 09:03:00 +00:00
|
|
|
|
2000-11-01 13:49:41 +00:00
|
|
|
pad = GST_PAD (gtk_type_new (gst_pad_get_type ()));
|
|
|
|
pad->name = g_strdup (name);
|
2000-01-30 09:03:00 +00:00
|
|
|
pad->direction = direction;
|
2000-12-13 19:29:35 +00:00
|
|
|
|
|
|
|
return pad;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* gst_pad_new_from_template:
|
|
|
|
* @temp: the pad template to use
|
|
|
|
* @name: the name of the element
|
|
|
|
*
|
|
|
|
* Create a new pad with given name from the given template.
|
|
|
|
*
|
|
|
|
* Returns: new pad
|
|
|
|
*/
|
|
|
|
GstPad*
|
|
|
|
gst_pad_new_from_template (GstPadTemplate *temp,
|
|
|
|
gchar *name)
|
|
|
|
{
|
|
|
|
GstPad *pad;
|
|
|
|
|
|
|
|
g_return_val_if_fail (name != NULL, NULL);
|
|
|
|
g_return_val_if_fail (temp != NULL, NULL);
|
|
|
|
|
|
|
|
pad = gst_pad_new (name, temp->direction);
|
|
|
|
pad->caps = temp->caps;
|
|
|
|
|
2000-01-30 09:03:00 +00:00
|
|
|
return pad;
|
|
|
|
}
|
|
|
|
|
2000-03-27 19:53:43 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_get_direction:
|
|
|
|
* @pad: the Pad to get the direction from
|
|
|
|
*
|
|
|
|
* get the direction of the pad
|
|
|
|
*
|
|
|
|
* Returns: the direction of the pad
|
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
GstPadDirection
|
|
|
|
gst_pad_get_direction (GstPad *pad)
|
|
|
|
{
|
|
|
|
g_return_val_if_fail (pad != NULL, GST_PAD_UNKNOWN);
|
|
|
|
g_return_val_if_fail (GST_IS_PAD (pad), GST_PAD_UNKNOWN);
|
2000-01-30 09:03:00 +00:00
|
|
|
|
|
|
|
return pad->direction;
|
|
|
|
}
|
|
|
|
|
2000-03-27 19:53:43 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_set_name:
|
|
|
|
* @pad: the pad to set the name of
|
|
|
|
* @name: the name of the pad
|
|
|
|
*
|
|
|
|
* set the name of a pad
|
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
void
|
|
|
|
gst_pad_set_name (GstPad *pad,
|
|
|
|
const gchar *name)
|
|
|
|
{
|
|
|
|
g_return_if_fail (pad != NULL);
|
|
|
|
g_return_if_fail (GST_IS_PAD (pad));
|
2000-01-30 09:03:00 +00:00
|
|
|
|
|
|
|
if (pad->name != NULL)
|
2000-11-01 13:49:41 +00:00
|
|
|
g_free (pad->name);
|
2000-01-30 09:03:00 +00:00
|
|
|
|
2000-11-01 13:49:41 +00:00
|
|
|
pad->name = g_strdup (name);
|
2000-01-30 09:03:00 +00:00
|
|
|
}
|
|
|
|
|
2000-03-27 19:53:43 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_get_name:
|
|
|
|
* @pad: the pad to get the name of
|
|
|
|
*
|
|
|
|
* get the name of a pad
|
|
|
|
*
|
2000-11-01 13:49:41 +00:00
|
|
|
* Returns: the name of the pad, don't free.
|
2000-03-27 19:53:43 +00:00
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
const gchar*
|
|
|
|
gst_pad_get_name (GstPad *pad)
|
|
|
|
{
|
|
|
|
g_return_val_if_fail (pad != NULL, NULL);
|
|
|
|
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
|
2000-01-30 09:03:00 +00:00
|
|
|
|
|
|
|
return pad->name;
|
|
|
|
}
|
|
|
|
|
2000-10-25 19:09:53 +00:00
|
|
|
/**
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
* gst_pad_set_chain_function:
|
|
|
|
* @pad: the pad to set the chain function for
|
|
|
|
* @chain: the chain function
|
2000-10-25 19:09:53 +00:00
|
|
|
*
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
* Set the given chain function for the pad
|
2000-10-25 19:09:53 +00:00
|
|
|
*/
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
void gst_pad_set_chain_function (GstPad *pad,
|
|
|
|
GstPadChainFunction chain)
|
2000-11-01 13:49:41 +00:00
|
|
|
{
|
|
|
|
g_return_if_fail (pad != NULL);
|
|
|
|
g_return_if_fail (GST_IS_PAD (pad));
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
|
|
|
|
pad->chainfunc = chain;
|
2000-02-26 18:55:14 +00:00
|
|
|
}
|
|
|
|
|
2000-12-06 23:04:12 +00:00
|
|
|
/**
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
* gst_pad_set_pull_function:
|
|
|
|
* @pad: the pad to set the get function for
|
|
|
|
* @get: the get function
|
2000-12-06 23:04:12 +00:00
|
|
|
*
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
* Set the given get function for the pad
|
2000-12-06 23:04:12 +00:00
|
|
|
*/
|
|
|
|
void
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
gst_pad_set_get_function (GstPad *pad,
|
|
|
|
GstPadGetFunction get)
|
2000-12-06 23:04:12 +00:00
|
|
|
{
|
|
|
|
g_return_if_fail (pad != NULL);
|
|
|
|
g_return_if_fail (GST_IS_PAD (pad));
|
|
|
|
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
// the if and such should optimize out when DEBUG is off
|
|
|
|
DEBUG("setting get function for %s:%s\n",GST_DEBUG_PAD_NAME(pad));
|
2000-12-06 23:04:12 +00:00
|
|
|
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
pad->getfunc = get;
|
|
|
|
DEBUG("getfunc for %s:%s(@%p) at %p is set to %p\n",GST_DEBUG_PAD_NAME(pad),pad,&pad->getfunc,get);
|
2000-12-06 23:04:12 +00:00
|
|
|
}
|
|
|
|
|
2000-10-25 19:09:53 +00:00
|
|
|
/**
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
* gst_pad_set_getregion_function:
|
|
|
|
* @pad: the pad to set the getregion function for
|
|
|
|
* @getregion: the getregion function
|
2000-10-25 19:09:53 +00:00
|
|
|
*
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
* Set the given getregion function for the pad
|
2000-10-25 19:09:53 +00:00
|
|
|
*/
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
void
|
|
|
|
gst_pad_set_getregion_function (GstPad *pad,
|
|
|
|
GstPadGetRegionFunction getregion)
|
2000-11-01 13:49:41 +00:00
|
|
|
{
|
|
|
|
g_return_if_fail (pad != NULL);
|
|
|
|
g_return_if_fail (GST_IS_PAD (pad));
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
|
|
|
|
g_print("gstpad: pad setting getregion function\n");
|
|
|
|
|
|
|
|
pad->getregionfunc = getregion;
|
2000-01-30 09:03:00 +00:00
|
|
|
}
|
|
|
|
|
2000-10-25 19:09:53 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_set_qos_function:
|
|
|
|
* @pad: the pad to set the qos function for
|
|
|
|
* @qos: the qos function
|
|
|
|
*
|
|
|
|
* Set the given qos function for the pad
|
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
void
|
|
|
|
gst_pad_set_qos_function (GstPad *pad,
|
|
|
|
GstPadQoSFunction qos)
|
|
|
|
{
|
|
|
|
g_return_if_fail (pad != NULL);
|
|
|
|
g_return_if_fail (GST_IS_PAD (pad));
|
2000-07-12 22:52:42 +00:00
|
|
|
|
2000-07-17 17:14:15 +00:00
|
|
|
pad->qosfunc = qos;
|
2000-07-12 22:52:42 +00:00
|
|
|
}
|
|
|
|
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
static void
|
|
|
|
gst_pad_push_func(GstPad *pad, GstBuffer *buf)
|
|
|
|
{
|
|
|
|
if (pad->peer->chainfunc != NULL) {
|
|
|
|
DEBUG("calling chain function\n");
|
|
|
|
(pad->peer->chainfunc)(pad,buf);
|
|
|
|
} else {
|
|
|
|
DEBUG("got a problem here: default pad_push handler in place, no chain function\n");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2000-10-25 19:09:53 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_push:
|
|
|
|
* @pad: the pad to push
|
|
|
|
* @buffer: the buffer to push
|
|
|
|
*
|
|
|
|
* pushes a buffer along a src pad
|
|
|
|
*/
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
/*
|
2000-11-01 13:49:41 +00:00
|
|
|
void
|
|
|
|
gst_pad_push (GstPad *pad,
|
|
|
|
GstBuffer *buffer)
|
|
|
|
{
|
Changed the way things are scheduled, especially sources. A Src used to have a push() function, and optionally a pus...
Original commit message from CVS:
Changed the way things are scheduled, especially sources. A Src used to
have a push() function, and optionally a pushregion() to deal with async
reads, etc. That whole thing has gone away, in favor of providing a
pull() function for the output (Src) pad instead, ala chain functions.
This makes constructing cothreaded schedules out of non-loop elements
somewhat easier. Basically there was always a question as to which pad
was being dealt with. In the pullregion case, cothread-specific data was
used to try to pass the region struct to the right place, which is a slow
hack. And in general, the push function severely limited the kind of
tricks that could be played when there's more than one output pad, such as
a multi-out file reader with async capabilities on each pad independently.
This changes the way cothread scheduling occurs. Instead of the hack to
deal with Src's by calling their push() function (or optionally the
pushregion(), in certain cases), we now are working towards a general
mechanism where pads are the only thing that are dealt with directly.
An optimization was made in the process of doing this: the loopfunction
actually run as the outer [stack] frame of the cothread is now set more
intelligently in create_plan() based on what kind of element it is. We
now have:
loopfunc_wrapper: used for loop-based elements, it simply calls the
loopfunc in a loop, paying attention to COTHREAD_STOPPING (see
below). It currently does other, soon to be depracated, stuff.
pullsrc_wrapper: wraps a Src that's not loop-based (since your options
are now loop- or pull-based)
There will be a couple more to deal with other cases, such as Connections
and chain-based elements. The general idea is that it's a lot more
efficient to make the decisions once in create_plan than to keep doing
this huge if/else chain in the wrapper. Just choose the right wrapper up
front. It'll be most apparent performance-wise in the case of whichever
element context is switched to first for each iteration, since the whole
wrapper setup is done for every iteration.
The tricky part is that there is now a bit of overloading of the function
pointers in a pad. The current meanings (possibly to change a bit more
soon) are:
chainfunc: as always, chainfunc pointer is mirrored between peer pads
(this may change, and the chain func may end up in pushfunc)
pushfunc: SrcPad: gst_pad_pushfunc_proxy, cothread_switch to peer
SinkPad: none (may take over chainfunc, see below) pullfunc:
SrcPad: Src or Connection's function to construct buffers
SinkPad: gst_pad_pullfunc_proxy, cothread_switch to peer
There are a number of issues remaining with the scheduling, not the least
of which is the fact that Connections are still dealt with the old way,
with _push() functions and such. I'm trying to figure out a way to unify
the system so it makes sense. Following the scheduling system is hard
enough, trying to change it is murder.
Another useful scheduling addition, mentioned above, is COTHREAD_STOPPING.
It's an element flag that's used to signal whatever code is running in
cothread context that it should be finishing up and exiting soon. An
example of this is in plugins/cobin/spindentity.c. All the loops should
now be composed of do/while loops, rather than while(1) loops:
do {
buf = gst_pad_pull(spindentity->sinkpad);
gst_pad_push(spindentity->srcpad,buf);
} while (!GST_ELEMENT_IS_COTHREAD_STOPPING(element));
The reason for this is that COTHREAD_STOPPING may be set before the above
loop ever gets started. It wouldn't do for the body of the loop to never
once get called, that would simply stall the pipeline. Note that only the
core library code is ever responsible for setting and unsetting this flag.
All elements have to do is respond to it by cleanly exiting the loop and
the function holding it.
This is needed primarily to allow iterations to occur properly.
Basically, there's a single entry point in the cothread scheduling loop,
gst_bin_iterate_func() simply switches to this cothread. If the element
in this context is allowed to loop infinitely, nothing would even switch
back to the context from which the iterate() was originally called. This
is a bit of a problem. The solution is for there to be an implicit switch
back to the originating context. Now, even I'm not sure exactly how this
works, but if the cothread that's switched to actually returns, execution
returns back to the calling context, i.e. iterate_func().
COTHREAD_STOPPING is therefore set just before switching into this
(currently randomly chosen) context, on the assumption that it will return
promptly after finishing its duties. The burden of clearing the flag
falls to the various wrapper functions provided by the Bin code, thus
element writers don't have to worry about doing that at all (and simply
shouldn't).
Related changes:
All the sources in elements/ have been changed to reflect the new system.
FIXMEs:
1) gstpipeline.c calls gst_src_push at some point, dunno why, it's
commented out now.
2) any other sources, including vcdsrc, dvdsrc, and v4lsrc will break
badly and need to be modified to work as pull-based sources.
2000-12-04 10:52:30 +00:00
|
|
|
GstPad *peer;
|
|
|
|
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
DEBUG_ENTER("(pad:'%s'(@%p),buffer:%p)",gst_pad_get_name(pad),pad,buffer);
|
Changed the way things are scheduled, especially sources. A Src used to have a push() function, and optionally a pus...
Original commit message from CVS:
Changed the way things are scheduled, especially sources. A Src used to
have a push() function, and optionally a pushregion() to deal with async
reads, etc. That whole thing has gone away, in favor of providing a
pull() function for the output (Src) pad instead, ala chain functions.
This makes constructing cothreaded schedules out of non-loop elements
somewhat easier. Basically there was always a question as to which pad
was being dealt with. In the pullregion case, cothread-specific data was
used to try to pass the region struct to the right place, which is a slow
hack. And in general, the push function severely limited the kind of
tricks that could be played when there's more than one output pad, such as
a multi-out file reader with async capabilities on each pad independently.
This changes the way cothread scheduling occurs. Instead of the hack to
deal with Src's by calling their push() function (or optionally the
pushregion(), in certain cases), we now are working towards a general
mechanism where pads are the only thing that are dealt with directly.
An optimization was made in the process of doing this: the loopfunction
actually run as the outer [stack] frame of the cothread is now set more
intelligently in create_plan() based on what kind of element it is. We
now have:
loopfunc_wrapper: used for loop-based elements, it simply calls the
loopfunc in a loop, paying attention to COTHREAD_STOPPING (see
below). It currently does other, soon to be depracated, stuff.
pullsrc_wrapper: wraps a Src that's not loop-based (since your options
are now loop- or pull-based)
There will be a couple more to deal with other cases, such as Connections
and chain-based elements. The general idea is that it's a lot more
efficient to make the decisions once in create_plan than to keep doing
this huge if/else chain in the wrapper. Just choose the right wrapper up
front. It'll be most apparent performance-wise in the case of whichever
element context is switched to first for each iteration, since the whole
wrapper setup is done for every iteration.
The tricky part is that there is now a bit of overloading of the function
pointers in a pad. The current meanings (possibly to change a bit more
soon) are:
chainfunc: as always, chainfunc pointer is mirrored between peer pads
(this may change, and the chain func may end up in pushfunc)
pushfunc: SrcPad: gst_pad_pushfunc_proxy, cothread_switch to peer
SinkPad: none (may take over chainfunc, see below) pullfunc:
SrcPad: Src or Connection's function to construct buffers
SinkPad: gst_pad_pullfunc_proxy, cothread_switch to peer
There are a number of issues remaining with the scheduling, not the least
of which is the fact that Connections are still dealt with the old way,
with _push() functions and such. I'm trying to figure out a way to unify
the system so it makes sense. Following the scheduling system is hard
enough, trying to change it is murder.
Another useful scheduling addition, mentioned above, is COTHREAD_STOPPING.
It's an element flag that's used to signal whatever code is running in
cothread context that it should be finishing up and exiting soon. An
example of this is in plugins/cobin/spindentity.c. All the loops should
now be composed of do/while loops, rather than while(1) loops:
do {
buf = gst_pad_pull(spindentity->sinkpad);
gst_pad_push(spindentity->srcpad,buf);
} while (!GST_ELEMENT_IS_COTHREAD_STOPPING(element));
The reason for this is that COTHREAD_STOPPING may be set before the above
loop ever gets started. It wouldn't do for the body of the loop to never
once get called, that would simply stall the pipeline. Note that only the
core library code is ever responsible for setting and unsetting this flag.
All elements have to do is respond to it by cleanly exiting the loop and
the function holding it.
This is needed primarily to allow iterations to occur properly.
Basically, there's a single entry point in the cothread scheduling loop,
gst_bin_iterate_func() simply switches to this cothread. If the element
in this context is allowed to loop infinitely, nothing would even switch
back to the context from which the iterate() was originally called. This
is a bit of a problem. The solution is for there to be an implicit switch
back to the originating context. Now, even I'm not sure exactly how this
works, but if the cothread that's switched to actually returns, execution
returns back to the calling context, i.e. iterate_func().
COTHREAD_STOPPING is therefore set just before switching into this
(currently randomly chosen) context, on the assumption that it will return
promptly after finishing its duties. The burden of clearing the flag
falls to the various wrapper functions provided by the Bin code, thus
element writers don't have to worry about doing that at all (and simply
shouldn't).
Related changes:
All the sources in elements/ have been changed to reflect the new system.
FIXMEs:
1) gstpipeline.c calls gst_src_push at some point, dunno why, it's
commented out now.
2) any other sources, including vcdsrc, dvdsrc, and v4lsrc will break
badly and need to be modified to work as pull-based sources.
2000-12-04 10:52:30 +00:00
|
|
|
|
2000-01-30 09:03:00 +00:00
|
|
|
g_return_if_fail(pad != NULL);
|
|
|
|
g_return_if_fail(GST_IS_PAD(pad));
|
2000-09-22 23:35:14 +00:00
|
|
|
g_return_if_fail(GST_PAD_CONNECTED(pad));
|
2000-01-30 09:03:00 +00:00
|
|
|
g_return_if_fail(buffer != NULL);
|
|
|
|
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
// if the pad has been disabled, unreference the pad and let it drop
|
2000-11-25 07:02:55 +00:00
|
|
|
if (GST_FLAG_IS_SET(pad,GST_PAD_DISABLED)) {
|
|
|
|
g_print("gst_pad_push: pad disabled, dropping buffer\n");
|
|
|
|
gst_buffer_unref(buffer);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2000-01-30 09:03:00 +00:00
|
|
|
gst_trace_add_entry(NULL,0,buffer,"push buffer");
|
2000-07-17 17:14:15 +00:00
|
|
|
|
Changed the way things are scheduled, especially sources. A Src used to have a push() function, and optionally a pus...
Original commit message from CVS:
Changed the way things are scheduled, especially sources. A Src used to
have a push() function, and optionally a pushregion() to deal with async
reads, etc. That whole thing has gone away, in favor of providing a
pull() function for the output (Src) pad instead, ala chain functions.
This makes constructing cothreaded schedules out of non-loop elements
somewhat easier. Basically there was always a question as to which pad
was being dealt with. In the pullregion case, cothread-specific data was
used to try to pass the region struct to the right place, which is a slow
hack. And in general, the push function severely limited the kind of
tricks that could be played when there's more than one output pad, such as
a multi-out file reader with async capabilities on each pad independently.
This changes the way cothread scheduling occurs. Instead of the hack to
deal with Src's by calling their push() function (or optionally the
pushregion(), in certain cases), we now are working towards a general
mechanism where pads are the only thing that are dealt with directly.
An optimization was made in the process of doing this: the loopfunction
actually run as the outer [stack] frame of the cothread is now set more
intelligently in create_plan() based on what kind of element it is. We
now have:
loopfunc_wrapper: used for loop-based elements, it simply calls the
loopfunc in a loop, paying attention to COTHREAD_STOPPING (see
below). It currently does other, soon to be depracated, stuff.
pullsrc_wrapper: wraps a Src that's not loop-based (since your options
are now loop- or pull-based)
There will be a couple more to deal with other cases, such as Connections
and chain-based elements. The general idea is that it's a lot more
efficient to make the decisions once in create_plan than to keep doing
this huge if/else chain in the wrapper. Just choose the right wrapper up
front. It'll be most apparent performance-wise in the case of whichever
element context is switched to first for each iteration, since the whole
wrapper setup is done for every iteration.
The tricky part is that there is now a bit of overloading of the function
pointers in a pad. The current meanings (possibly to change a bit more
soon) are:
chainfunc: as always, chainfunc pointer is mirrored between peer pads
(this may change, and the chain func may end up in pushfunc)
pushfunc: SrcPad: gst_pad_pushfunc_proxy, cothread_switch to peer
SinkPad: none (may take over chainfunc, see below) pullfunc:
SrcPad: Src or Connection's function to construct buffers
SinkPad: gst_pad_pullfunc_proxy, cothread_switch to peer
There are a number of issues remaining with the scheduling, not the least
of which is the fact that Connections are still dealt with the old way,
with _push() functions and such. I'm trying to figure out a way to unify
the system so it makes sense. Following the scheduling system is hard
enough, trying to change it is murder.
Another useful scheduling addition, mentioned above, is COTHREAD_STOPPING.
It's an element flag that's used to signal whatever code is running in
cothread context that it should be finishing up and exiting soon. An
example of this is in plugins/cobin/spindentity.c. All the loops should
now be composed of do/while loops, rather than while(1) loops:
do {
buf = gst_pad_pull(spindentity->sinkpad);
gst_pad_push(spindentity->srcpad,buf);
} while (!GST_ELEMENT_IS_COTHREAD_STOPPING(element));
The reason for this is that COTHREAD_STOPPING may be set before the above
loop ever gets started. It wouldn't do for the body of the loop to never
once get called, that would simply stall the pipeline. Note that only the
core library code is ever responsible for setting and unsetting this flag.
All elements have to do is respond to it by cleanly exiting the loop and
the function holding it.
This is needed primarily to allow iterations to occur properly.
Basically, there's a single entry point in the cothread scheduling loop,
gst_bin_iterate_func() simply switches to this cothread. If the element
in this context is allowed to loop infinitely, nothing would even switch
back to the context from which the iterate() was originally called. This
is a bit of a problem. The solution is for there to be an implicit switch
back to the originating context. Now, even I'm not sure exactly how this
works, but if the cothread that's switched to actually returns, execution
returns back to the calling context, i.e. iterate_func().
COTHREAD_STOPPING is therefore set just before switching into this
(currently randomly chosen) context, on the assumption that it will return
promptly after finishing its duties. The burden of clearing the flag
falls to the various wrapper functions provided by the Bin code, thus
element writers don't have to worry about doing that at all (and simply
shouldn't).
Related changes:
All the sources in elements/ have been changed to reflect the new system.
FIXMEs:
1) gstpipeline.c calls gst_src_push at some point, dunno why, it's
commented out now.
2) any other sources, including vcdsrc, dvdsrc, and v4lsrc will break
badly and need to be modified to work as pull-based sources.
2000-12-04 10:52:30 +00:00
|
|
|
peer = pad->peer;
|
|
|
|
g_return_if_fail(peer != NULL);
|
|
|
|
|
2000-07-17 17:14:15 +00:00
|
|
|
// first check to see if there's a push handler
|
|
|
|
if (pad->pushfunc != NULL) {
|
2000-12-11 00:24:32 +00:00
|
|
|
DEBUG("putting the buffer in the pen and calling pushfunc\n");
|
2000-07-17 17:14:15 +00:00
|
|
|
// put the buffer in peer's holding pen
|
Changed the way things are scheduled, especially sources. A Src used to have a push() function, and optionally a pus...
Original commit message from CVS:
Changed the way things are scheduled, especially sources. A Src used to
have a push() function, and optionally a pushregion() to deal with async
reads, etc. That whole thing has gone away, in favor of providing a
pull() function for the output (Src) pad instead, ala chain functions.
This makes constructing cothreaded schedules out of non-loop elements
somewhat easier. Basically there was always a question as to which pad
was being dealt with. In the pullregion case, cothread-specific data was
used to try to pass the region struct to the right place, which is a slow
hack. And in general, the push function severely limited the kind of
tricks that could be played when there's more than one output pad, such as
a multi-out file reader with async capabilities on each pad independently.
This changes the way cothread scheduling occurs. Instead of the hack to
deal with Src's by calling their push() function (or optionally the
pushregion(), in certain cases), we now are working towards a general
mechanism where pads are the only thing that are dealt with directly.
An optimization was made in the process of doing this: the loopfunction
actually run as the outer [stack] frame of the cothread is now set more
intelligently in create_plan() based on what kind of element it is. We
now have:
loopfunc_wrapper: used for loop-based elements, it simply calls the
loopfunc in a loop, paying attention to COTHREAD_STOPPING (see
below). It currently does other, soon to be depracated, stuff.
pullsrc_wrapper: wraps a Src that's not loop-based (since your options
are now loop- or pull-based)
There will be a couple more to deal with other cases, such as Connections
and chain-based elements. The general idea is that it's a lot more
efficient to make the decisions once in create_plan than to keep doing
this huge if/else chain in the wrapper. Just choose the right wrapper up
front. It'll be most apparent performance-wise in the case of whichever
element context is switched to first for each iteration, since the whole
wrapper setup is done for every iteration.
The tricky part is that there is now a bit of overloading of the function
pointers in a pad. The current meanings (possibly to change a bit more
soon) are:
chainfunc: as always, chainfunc pointer is mirrored between peer pads
(this may change, and the chain func may end up in pushfunc)
pushfunc: SrcPad: gst_pad_pushfunc_proxy, cothread_switch to peer
SinkPad: none (may take over chainfunc, see below) pullfunc:
SrcPad: Src or Connection's function to construct buffers
SinkPad: gst_pad_pullfunc_proxy, cothread_switch to peer
There are a number of issues remaining with the scheduling, not the least
of which is the fact that Connections are still dealt with the old way,
with _push() functions and such. I'm trying to figure out a way to unify
the system so it makes sense. Following the scheduling system is hard
enough, trying to change it is murder.
Another useful scheduling addition, mentioned above, is COTHREAD_STOPPING.
It's an element flag that's used to signal whatever code is running in
cothread context that it should be finishing up and exiting soon. An
example of this is in plugins/cobin/spindentity.c. All the loops should
now be composed of do/while loops, rather than while(1) loops:
do {
buf = gst_pad_pull(spindentity->sinkpad);
gst_pad_push(spindentity->srcpad,buf);
} while (!GST_ELEMENT_IS_COTHREAD_STOPPING(element));
The reason for this is that COTHREAD_STOPPING may be set before the above
loop ever gets started. It wouldn't do for the body of the loop to never
once get called, that would simply stall the pipeline. Note that only the
core library code is ever responsible for setting and unsetting this flag.
All elements have to do is respond to it by cleanly exiting the loop and
the function holding it.
This is needed primarily to allow iterations to occur properly.
Basically, there's a single entry point in the cothread scheduling loop,
gst_bin_iterate_func() simply switches to this cothread. If the element
in this context is allowed to loop infinitely, nothing would even switch
back to the context from which the iterate() was originally called. This
is a bit of a problem. The solution is for there to be an implicit switch
back to the originating context. Now, even I'm not sure exactly how this
works, but if the cothread that's switched to actually returns, execution
returns back to the calling context, i.e. iterate_func().
COTHREAD_STOPPING is therefore set just before switching into this
(currently randomly chosen) context, on the assumption that it will return
promptly after finishing its duties. The burden of clearing the flag
falls to the various wrapper functions provided by the Bin code, thus
element writers don't have to worry about doing that at all (and simply
shouldn't).
Related changes:
All the sources in elements/ have been changed to reflect the new system.
FIXMEs:
1) gstpipeline.c calls gst_src_push at some point, dunno why, it's
commented out now.
2) any other sources, including vcdsrc, dvdsrc, and v4lsrc will break
badly and need to be modified to work as pull-based sources.
2000-12-04 10:52:30 +00:00
|
|
|
peer->bufpen = buffer;
|
2000-07-17 17:14:15 +00:00
|
|
|
// now inform the handler that the peer pad has something
|
Changed the way things are scheduled, especially sources. A Src used to have a push() function, and optionally a pus...
Original commit message from CVS:
Changed the way things are scheduled, especially sources. A Src used to
have a push() function, and optionally a pushregion() to deal with async
reads, etc. That whole thing has gone away, in favor of providing a
pull() function for the output (Src) pad instead, ala chain functions.
This makes constructing cothreaded schedules out of non-loop elements
somewhat easier. Basically there was always a question as to which pad
was being dealt with. In the pullregion case, cothread-specific data was
used to try to pass the region struct to the right place, which is a slow
hack. And in general, the push function severely limited the kind of
tricks that could be played when there's more than one output pad, such as
a multi-out file reader with async capabilities on each pad independently.
This changes the way cothread scheduling occurs. Instead of the hack to
deal with Src's by calling their push() function (or optionally the
pushregion(), in certain cases), we now are working towards a general
mechanism where pads are the only thing that are dealt with directly.
An optimization was made in the process of doing this: the loopfunction
actually run as the outer [stack] frame of the cothread is now set more
intelligently in create_plan() based on what kind of element it is. We
now have:
loopfunc_wrapper: used for loop-based elements, it simply calls the
loopfunc in a loop, paying attention to COTHREAD_STOPPING (see
below). It currently does other, soon to be depracated, stuff.
pullsrc_wrapper: wraps a Src that's not loop-based (since your options
are now loop- or pull-based)
There will be a couple more to deal with other cases, such as Connections
and chain-based elements. The general idea is that it's a lot more
efficient to make the decisions once in create_plan than to keep doing
this huge if/else chain in the wrapper. Just choose the right wrapper up
front. It'll be most apparent performance-wise in the case of whichever
element context is switched to first for each iteration, since the whole
wrapper setup is done for every iteration.
The tricky part is that there is now a bit of overloading of the function
pointers in a pad. The current meanings (possibly to change a bit more
soon) are:
chainfunc: as always, chainfunc pointer is mirrored between peer pads
(this may change, and the chain func may end up in pushfunc)
pushfunc: SrcPad: gst_pad_pushfunc_proxy, cothread_switch to peer
SinkPad: none (may take over chainfunc, see below) pullfunc:
SrcPad: Src or Connection's function to construct buffers
SinkPad: gst_pad_pullfunc_proxy, cothread_switch to peer
There are a number of issues remaining with the scheduling, not the least
of which is the fact that Connections are still dealt with the old way,
with _push() functions and such. I'm trying to figure out a way to unify
the system so it makes sense. Following the scheduling system is hard
enough, trying to change it is murder.
Another useful scheduling addition, mentioned above, is COTHREAD_STOPPING.
It's an element flag that's used to signal whatever code is running in
cothread context that it should be finishing up and exiting soon. An
example of this is in plugins/cobin/spindentity.c. All the loops should
now be composed of do/while loops, rather than while(1) loops:
do {
buf = gst_pad_pull(spindentity->sinkpad);
gst_pad_push(spindentity->srcpad,buf);
} while (!GST_ELEMENT_IS_COTHREAD_STOPPING(element));
The reason for this is that COTHREAD_STOPPING may be set before the above
loop ever gets started. It wouldn't do for the body of the loop to never
once get called, that would simply stall the pipeline. Note that only the
core library code is ever responsible for setting and unsetting this flag.
All elements have to do is respond to it by cleanly exiting the loop and
the function holding it.
This is needed primarily to allow iterations to occur properly.
Basically, there's a single entry point in the cothread scheduling loop,
gst_bin_iterate_func() simply switches to this cothread. If the element
in this context is allowed to loop infinitely, nothing would even switch
back to the context from which the iterate() was originally called. This
is a bit of a problem. The solution is for there to be an implicit switch
back to the originating context. Now, even I'm not sure exactly how this
works, but if the cothread that's switched to actually returns, execution
returns back to the calling context, i.e. iterate_func().
COTHREAD_STOPPING is therefore set just before switching into this
(currently randomly chosen) context, on the assumption that it will return
promptly after finishing its duties. The burden of clearing the flag
falls to the various wrapper functions provided by the Bin code, thus
element writers don't have to worry about doing that at all (and simply
shouldn't).
Related changes:
All the sources in elements/ have been changed to reflect the new system.
FIXMEs:
1) gstpipeline.c calls gst_src_push at some point, dunno why, it's
commented out now.
2) any other sources, including vcdsrc, dvdsrc, and v4lsrc will break
badly and need to be modified to work as pull-based sources.
2000-12-04 10:52:30 +00:00
|
|
|
(pad->pushfunc)(peer);
|
2000-07-17 17:14:15 +00:00
|
|
|
// otherwise we assume we're chaining directly
|
Changed the way things are scheduled, especially sources. A Src used to have a push() function, and optionally a pus...
Original commit message from CVS:
Changed the way things are scheduled, especially sources. A Src used to
have a push() function, and optionally a pushregion() to deal with async
reads, etc. That whole thing has gone away, in favor of providing a
pull() function for the output (Src) pad instead, ala chain functions.
This makes constructing cothreaded schedules out of non-loop elements
somewhat easier. Basically there was always a question as to which pad
was being dealt with. In the pullregion case, cothread-specific data was
used to try to pass the region struct to the right place, which is a slow
hack. And in general, the push function severely limited the kind of
tricks that could be played when there's more than one output pad, such as
a multi-out file reader with async capabilities on each pad independently.
This changes the way cothread scheduling occurs. Instead of the hack to
deal with Src's by calling their push() function (or optionally the
pushregion(), in certain cases), we now are working towards a general
mechanism where pads are the only thing that are dealt with directly.
An optimization was made in the process of doing this: the loopfunction
actually run as the outer [stack] frame of the cothread is now set more
intelligently in create_plan() based on what kind of element it is. We
now have:
loopfunc_wrapper: used for loop-based elements, it simply calls the
loopfunc in a loop, paying attention to COTHREAD_STOPPING (see
below). It currently does other, soon to be depracated, stuff.
pullsrc_wrapper: wraps a Src that's not loop-based (since your options
are now loop- or pull-based)
There will be a couple more to deal with other cases, such as Connections
and chain-based elements. The general idea is that it's a lot more
efficient to make the decisions once in create_plan than to keep doing
this huge if/else chain in the wrapper. Just choose the right wrapper up
front. It'll be most apparent performance-wise in the case of whichever
element context is switched to first for each iteration, since the whole
wrapper setup is done for every iteration.
The tricky part is that there is now a bit of overloading of the function
pointers in a pad. The current meanings (possibly to change a bit more
soon) are:
chainfunc: as always, chainfunc pointer is mirrored between peer pads
(this may change, and the chain func may end up in pushfunc)
pushfunc: SrcPad: gst_pad_pushfunc_proxy, cothread_switch to peer
SinkPad: none (may take over chainfunc, see below) pullfunc:
SrcPad: Src or Connection's function to construct buffers
SinkPad: gst_pad_pullfunc_proxy, cothread_switch to peer
There are a number of issues remaining with the scheduling, not the least
of which is the fact that Connections are still dealt with the old way,
with _push() functions and such. I'm trying to figure out a way to unify
the system so it makes sense. Following the scheduling system is hard
enough, trying to change it is murder.
Another useful scheduling addition, mentioned above, is COTHREAD_STOPPING.
It's an element flag that's used to signal whatever code is running in
cothread context that it should be finishing up and exiting soon. An
example of this is in plugins/cobin/spindentity.c. All the loops should
now be composed of do/while loops, rather than while(1) loops:
do {
buf = gst_pad_pull(spindentity->sinkpad);
gst_pad_push(spindentity->srcpad,buf);
} while (!GST_ELEMENT_IS_COTHREAD_STOPPING(element));
The reason for this is that COTHREAD_STOPPING may be set before the above
loop ever gets started. It wouldn't do for the body of the loop to never
once get called, that would simply stall the pipeline. Note that only the
core library code is ever responsible for setting and unsetting this flag.
All elements have to do is respond to it by cleanly exiting the loop and
the function holding it.
This is needed primarily to allow iterations to occur properly.
Basically, there's a single entry point in the cothread scheduling loop,
gst_bin_iterate_func() simply switches to this cothread. If the element
in this context is allowed to loop infinitely, nothing would even switch
back to the context from which the iterate() was originally called. This
is a bit of a problem. The solution is for there to be an implicit switch
back to the originating context. Now, even I'm not sure exactly how this
works, but if the cothread that's switched to actually returns, execution
returns back to the calling context, i.e. iterate_func().
COTHREAD_STOPPING is therefore set just before switching into this
(currently randomly chosen) context, on the assumption that it will return
promptly after finishing its duties. The burden of clearing the flag
falls to the various wrapper functions provided by the Bin code, thus
element writers don't have to worry about doing that at all (and simply
shouldn't).
Related changes:
All the sources in elements/ have been changed to reflect the new system.
FIXMEs:
1) gstpipeline.c calls gst_src_push at some point, dunno why, it's
commented out now.
2) any other sources, including vcdsrc, dvdsrc, and v4lsrc will break
badly and need to be modified to work as pull-based sources.
2000-12-04 10:52:30 +00:00
|
|
|
} else if (peer->chainfunc != NULL) {
|
2000-12-11 00:24:32 +00:00
|
|
|
DEBUG("calling chain function\n");
|
2000-08-14 10:55:35 +00:00
|
|
|
//g_print("-- gst_pad_push(): calling chain handler\n");
|
Changed the way things are scheduled, especially sources. A Src used to have a push() function, and optionally a pus...
Original commit message from CVS:
Changed the way things are scheduled, especially sources. A Src used to
have a push() function, and optionally a pushregion() to deal with async
reads, etc. That whole thing has gone away, in favor of providing a
pull() function for the output (Src) pad instead, ala chain functions.
This makes constructing cothreaded schedules out of non-loop elements
somewhat easier. Basically there was always a question as to which pad
was being dealt with. In the pullregion case, cothread-specific data was
used to try to pass the region struct to the right place, which is a slow
hack. And in general, the push function severely limited the kind of
tricks that could be played when there's more than one output pad, such as
a multi-out file reader with async capabilities on each pad independently.
This changes the way cothread scheduling occurs. Instead of the hack to
deal with Src's by calling their push() function (or optionally the
pushregion(), in certain cases), we now are working towards a general
mechanism where pads are the only thing that are dealt with directly.
An optimization was made in the process of doing this: the loopfunction
actually run as the outer [stack] frame of the cothread is now set more
intelligently in create_plan() based on what kind of element it is. We
now have:
loopfunc_wrapper: used for loop-based elements, it simply calls the
loopfunc in a loop, paying attention to COTHREAD_STOPPING (see
below). It currently does other, soon to be depracated, stuff.
pullsrc_wrapper: wraps a Src that's not loop-based (since your options
are now loop- or pull-based)
There will be a couple more to deal with other cases, such as Connections
and chain-based elements. The general idea is that it's a lot more
efficient to make the decisions once in create_plan than to keep doing
this huge if/else chain in the wrapper. Just choose the right wrapper up
front. It'll be most apparent performance-wise in the case of whichever
element context is switched to first for each iteration, since the whole
wrapper setup is done for every iteration.
The tricky part is that there is now a bit of overloading of the function
pointers in a pad. The current meanings (possibly to change a bit more
soon) are:
chainfunc: as always, chainfunc pointer is mirrored between peer pads
(this may change, and the chain func may end up in pushfunc)
pushfunc: SrcPad: gst_pad_pushfunc_proxy, cothread_switch to peer
SinkPad: none (may take over chainfunc, see below) pullfunc:
SrcPad: Src or Connection's function to construct buffers
SinkPad: gst_pad_pullfunc_proxy, cothread_switch to peer
There are a number of issues remaining with the scheduling, not the least
of which is the fact that Connections are still dealt with the old way,
with _push() functions and such. I'm trying to figure out a way to unify
the system so it makes sense. Following the scheduling system is hard
enough, trying to change it is murder.
Another useful scheduling addition, mentioned above, is COTHREAD_STOPPING.
It's an element flag that's used to signal whatever code is running in
cothread context that it should be finishing up and exiting soon. An
example of this is in plugins/cobin/spindentity.c. All the loops should
now be composed of do/while loops, rather than while(1) loops:
do {
buf = gst_pad_pull(spindentity->sinkpad);
gst_pad_push(spindentity->srcpad,buf);
} while (!GST_ELEMENT_IS_COTHREAD_STOPPING(element));
The reason for this is that COTHREAD_STOPPING may be set before the above
loop ever gets started. It wouldn't do for the body of the loop to never
once get called, that would simply stall the pipeline. Note that only the
core library code is ever responsible for setting and unsetting this flag.
All elements have to do is respond to it by cleanly exiting the loop and
the function holding it.
This is needed primarily to allow iterations to occur properly.
Basically, there's a single entry point in the cothread scheduling loop,
gst_bin_iterate_func() simply switches to this cothread. If the element
in this context is allowed to loop infinitely, nothing would even switch
back to the context from which the iterate() was originally called. This
is a bit of a problem. The solution is for there to be an implicit switch
back to the originating context. Now, even I'm not sure exactly how this
works, but if the cothread that's switched to actually returns, execution
returns back to the calling context, i.e. iterate_func().
COTHREAD_STOPPING is therefore set just before switching into this
(currently randomly chosen) context, on the assumption that it will return
promptly after finishing its duties. The burden of clearing the flag
falls to the various wrapper functions provided by the Bin code, thus
element writers don't have to worry about doing that at all (and simply
shouldn't).
Related changes:
All the sources in elements/ have been changed to reflect the new system.
FIXMEs:
1) gstpipeline.c calls gst_src_push at some point, dunno why, it's
commented out now.
2) any other sources, including vcdsrc, dvdsrc, and v4lsrc will break
badly and need to be modified to work as pull-based sources.
2000-12-04 10:52:30 +00:00
|
|
|
(peer->chainfunc)(peer,buffer);
|
2000-07-17 17:14:15 +00:00
|
|
|
// else we squawk
|
|
|
|
} else {
|
2000-09-24 14:29:49 +00:00
|
|
|
g_print("-- gst_pad_push(): houston, we have a problem, no way of talking to peer\n");
|
2000-07-17 17:14:15 +00:00
|
|
|
}
|
2000-01-30 09:03:00 +00:00
|
|
|
}
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
*/
|
2000-01-30 09:03:00 +00:00
|
|
|
|
2000-10-25 19:09:53 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_pull:
|
|
|
|
* @pad: the pad to pull
|
|
|
|
*
|
|
|
|
* pulls a buffer along a sink pad
|
|
|
|
*
|
|
|
|
* Returns: the buffer that was pulled
|
|
|
|
*/
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
/*
|
2000-11-01 13:49:41 +00:00
|
|
|
GstBuffer*
|
|
|
|
gst_pad_pull (GstPad *pad)
|
|
|
|
{
|
2000-01-30 09:03:00 +00:00
|
|
|
GstBuffer *buf;
|
|
|
|
|
2000-12-11 00:24:32 +00:00
|
|
|
DEBUG_ENTER("(%s:%s)",GST_DEBUG_PAD_NAME(pad));
|
|
|
|
|
2000-02-27 23:18:38 +00:00
|
|
|
g_return_val_if_fail(pad != NULL, NULL);
|
|
|
|
g_return_val_if_fail(GST_IS_PAD(pad), NULL);
|
2000-01-30 09:03:00 +00:00
|
|
|
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
// check to see if the peer pad is disabled. return NULL if it is
|
|
|
|
// FIXME: this may be the wrong way to go about it
|
2000-11-25 07:02:55 +00:00
|
|
|
if (GST_FLAG_IS_SET(pad->peer,GST_PAD_DISABLED)) {
|
2000-12-11 00:24:32 +00:00
|
|
|
DEBUG("pad disabled, returning NULL\n");
|
2000-11-25 07:02:55 +00:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2000-07-17 17:14:15 +00:00
|
|
|
// if no buffer in pen and there's a pull handler, fire it
|
|
|
|
if (pad->bufpen == NULL) {
|
|
|
|
if (pad->pullfunc != NULL) {
|
2000-12-11 00:24:32 +00:00
|
|
|
DEBUG("calling pullfunc to fill buffer pen\n");
|
2000-07-17 17:14:15 +00:00
|
|
|
(pad->pullfunc)(pad->peer);
|
|
|
|
} else {
|
2000-12-11 00:24:32 +00:00
|
|
|
DEBUG("no buffer in pen, and no handler (# %p) to get one there!!!\n",&pad->pullfunc);
|
2000-12-12 19:29:43 +00:00
|
|
|
g_return_val_if_fail(pad->pullfunc != NULL, NULL);
|
2000-07-17 17:14:15 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// if there's a buffer in the holding pen, use it
|
|
|
|
if (pad->bufpen != NULL) {
|
|
|
|
buf = pad->bufpen;
|
|
|
|
pad->bufpen = NULL;
|
|
|
|
return buf;
|
|
|
|
// else we have a big problem...
|
|
|
|
} else {
|
2000-12-11 00:24:32 +00:00
|
|
|
DEBUG("no buffer in pen, and no handler\n");
|
2000-12-12 19:29:43 +00:00
|
|
|
g_return_val_if_fail(pad->pullfunc != NULL, NULL);
|
2000-07-17 17:14:15 +00:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
return NULL;
|
2000-01-30 09:03:00 +00:00
|
|
|
}
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
*/
|
2000-01-30 09:03:00 +00:00
|
|
|
|
2000-11-01 13:49:41 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_pull_region:
|
|
|
|
* @pad: the pad to pull
|
|
|
|
* @offset: the offset to pull
|
|
|
|
* @size: the size to pull
|
|
|
|
*
|
|
|
|
* pulls a buffer along a sink pad with a given offset and size
|
|
|
|
*
|
|
|
|
* Returns: the buffer that was pulled
|
|
|
|
*/
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
/*
|
2000-11-01 13:49:41 +00:00
|
|
|
GstBuffer*
|
|
|
|
gst_pad_pull_region (GstPad *pad,
|
|
|
|
gulong offset,
|
|
|
|
gulong size)
|
|
|
|
{
|
2000-11-06 00:15:51 +00:00
|
|
|
GstBuffer *buf;
|
|
|
|
|
|
|
|
g_return_val_if_fail(pad != NULL, NULL);
|
|
|
|
g_return_val_if_fail(GST_IS_PAD(pad), NULL);
|
|
|
|
|
|
|
|
DEBUG("-- gst_pad_pull_region(%s:%s): region (%lu,%lu)\n",
|
|
|
|
GST_ELEMENT(pad->parent)->name, pad->peer->name,
|
|
|
|
offset, size);
|
|
|
|
|
|
|
|
// if no buffer in pen and there's a pull handler, fire it
|
|
|
|
if (pad->bufpen == NULL) {
|
|
|
|
if (pad->pullregionfunc != NULL) {
|
|
|
|
(pad->pullregionfunc)(pad->peer, offset, size);
|
|
|
|
} else {
|
|
|
|
g_print("-- gst_pad_pull_region(%s:%s): no buffer in pen, and no handler to get one there!!!\n",
|
|
|
|
GST_ELEMENT(pad->parent)->name, pad->name);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// if there's a buffer in the holding pen, use it
|
|
|
|
if (pad->bufpen != NULL) {
|
|
|
|
buf = pad->bufpen;
|
|
|
|
pad->bufpen = NULL;
|
|
|
|
return buf;
|
|
|
|
// else we have a big problem...
|
|
|
|
} else {
|
|
|
|
g_print("-- gst_pad_pull_region(%s:%s): no buffer in pen, and no handler\n",
|
|
|
|
GST_ELEMENT(pad->parent)->name, pad->peer->name);
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
return NULL;
|
2000-11-01 13:49:41 +00:00
|
|
|
}
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
|
2000-10-25 19:09:53 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_chain:
|
|
|
|
* @pad: the pad to chain
|
|
|
|
*
|
|
|
|
* call the chain function of the given pad
|
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
void
|
|
|
|
gst_pad_chain (GstPad *pad)
|
|
|
|
{
|
|
|
|
g_return_if_fail (pad != NULL);
|
|
|
|
g_return_if_fail (GST_IS_PAD (pad));
|
|
|
|
g_return_if_fail (pad->peer != NULL);
|
|
|
|
g_return_if_fail (pad->chainfunc != NULL);
|
2000-01-30 09:03:00 +00:00
|
|
|
|
2000-07-17 17:14:15 +00:00
|
|
|
if (pad->bufpen && pad->chainfunc)
|
2000-11-01 13:49:41 +00:00
|
|
|
(pad->chainfunc) (pad,pad->bufpen);
|
2000-01-30 09:03:00 +00:00
|
|
|
}
|
|
|
|
|
2000-07-17 17:14:15 +00:00
|
|
|
|
2000-07-12 22:52:42 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_handle_qos:
|
2000-09-14 20:31:03 +00:00
|
|
|
* @pad: the pad to handle the QoS message
|
|
|
|
* @qos_message: the QoS message to handle
|
2000-07-12 22:52:42 +00:00
|
|
|
*
|
2000-10-25 19:09:53 +00:00
|
|
|
* pass the qos message downstream
|
2000-07-12 22:52:42 +00:00
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
void
|
|
|
|
gst_pad_handle_qos(GstPad *pad,
|
|
|
|
glong qos_message)
|
2000-07-12 22:52:42 +00:00
|
|
|
{
|
|
|
|
GstElement *element;
|
|
|
|
GList *pads;
|
|
|
|
GstPad *target_pad;
|
|
|
|
|
|
|
|
DEBUG("gst_pad_handle_qos(\"%s\",%08ld)\n", GST_ELEMENT(pad->parent)->name,qos_message);
|
|
|
|
|
2000-07-17 17:14:15 +00:00
|
|
|
if (pad->qosfunc) {
|
2000-11-01 13:49:41 +00:00
|
|
|
(pad->qosfunc) (pad,qos_message);
|
2000-07-12 22:52:42 +00:00
|
|
|
}
|
|
|
|
else {
|
2000-11-01 13:49:41 +00:00
|
|
|
element = GST_ELEMENT (pad->peer->parent);
|
2000-07-12 22:52:42 +00:00
|
|
|
|
|
|
|
pads = element->pads;
|
|
|
|
DEBUG("gst_pad_handle_qos recurse(\"%s\",%08ld)\n", element->name,qos_message);
|
|
|
|
while (pads) {
|
2000-11-01 13:49:41 +00:00
|
|
|
target_pad = GST_PAD (pads->data);
|
2000-07-12 22:52:42 +00:00
|
|
|
if (target_pad->direction == GST_PAD_SINK) {
|
2000-11-01 13:49:41 +00:00
|
|
|
gst_pad_handle_qos (target_pad, qos_message);
|
2000-07-12 22:52:42 +00:00
|
|
|
}
|
2000-11-01 13:49:41 +00:00
|
|
|
pads = g_list_next (pads);
|
2000-07-12 22:52:42 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2000-10-25 19:09:53 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_disconnect:
|
|
|
|
* @srcpad: the source pad to disconnect
|
|
|
|
* @sinkpad: the sink pad to disconnect
|
|
|
|
*
|
|
|
|
* disconnects the source pad from the sink pad
|
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
void
|
|
|
|
gst_pad_disconnect (GstPad *srcpad,
|
|
|
|
GstPad *sinkpad)
|
|
|
|
{
|
2000-06-25 21:38:00 +00:00
|
|
|
/* generic checks */
|
2000-11-01 13:49:41 +00:00
|
|
|
g_return_if_fail (srcpad != NULL);
|
|
|
|
g_return_if_fail (GST_IS_PAD (srcpad));
|
|
|
|
g_return_if_fail (srcpad->peer != NULL);
|
|
|
|
g_return_if_fail (sinkpad != NULL);
|
|
|
|
g_return_if_fail (GST_IS_PAD (sinkpad));
|
|
|
|
g_return_if_fail (sinkpad->peer != NULL);
|
2000-06-25 21:38:00 +00:00
|
|
|
|
2000-11-01 13:49:41 +00:00
|
|
|
g_return_if_fail ((srcpad->direction == GST_PAD_SRC) &&
|
|
|
|
(sinkpad->direction == GST_PAD_SINK));
|
2000-06-25 21:38:00 +00:00
|
|
|
|
|
|
|
/* first clear peers */
|
|
|
|
srcpad->peer = NULL;
|
|
|
|
sinkpad->peer = NULL;
|
|
|
|
|
|
|
|
}
|
|
|
|
|
2000-10-25 19:09:53 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_connect:
|
|
|
|
* @srcpad: the source pad to connect
|
|
|
|
* @sinkpad: the sink pad to connect
|
|
|
|
*
|
|
|
|
* connects the source pad to the sink pad
|
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
void
|
|
|
|
gst_pad_connect (GstPad *srcpad,
|
|
|
|
GstPad *sinkpad)
|
|
|
|
{
|
2000-01-30 09:03:00 +00:00
|
|
|
GstPad *temppad;
|
|
|
|
|
|
|
|
/* generic checks */
|
|
|
|
g_return_if_fail(srcpad != NULL);
|
|
|
|
g_return_if_fail(GST_IS_PAD(srcpad));
|
|
|
|
g_return_if_fail(srcpad->peer == NULL);
|
|
|
|
g_return_if_fail(sinkpad != NULL);
|
|
|
|
g_return_if_fail(GST_IS_PAD(sinkpad));
|
|
|
|
g_return_if_fail(sinkpad->peer == NULL);
|
|
|
|
// g_return_if_fail(sinkpad->chain != NULL);
|
|
|
|
|
|
|
|
/* check for reversed directions and swap if necessary */
|
|
|
|
if ((srcpad->direction == GST_PAD_SINK) &&
|
|
|
|
(sinkpad->direction == GST_PAD_SRC)) {
|
|
|
|
temppad = srcpad;
|
|
|
|
srcpad = sinkpad;
|
|
|
|
sinkpad = temppad;
|
|
|
|
}
|
|
|
|
g_return_if_fail((srcpad->direction == GST_PAD_SRC) &&
|
|
|
|
(sinkpad->direction == GST_PAD_SINK));
|
2000-12-14 17:21:29 +00:00
|
|
|
|
|
|
|
/* chack pad compatibility */
|
|
|
|
if (srcpad->caps && sinkpad->caps) {
|
|
|
|
if (!gst_caps_check_compatibility (srcpad->caps, sinkpad->caps))
|
2000-12-22 23:23:10 +00:00
|
|
|
g_warning ("gstpad: connecting incompatible pads (%s:%s) and (%s:%s)\n",
|
|
|
|
GST_DEBUG_PAD_NAME (srcpad), GST_DEBUG_PAD_NAME (sinkpad));
|
2000-12-14 17:21:29 +00:00
|
|
|
else
|
2000-12-22 23:23:10 +00:00
|
|
|
g_print ("gstpad: connecting compatible pads (%s:%s) and (%s:%s)\n",
|
|
|
|
GST_DEBUG_PAD_NAME (srcpad), GST_DEBUG_PAD_NAME (sinkpad));
|
2000-12-14 17:21:29 +00:00
|
|
|
}
|
2000-12-17 16:24:14 +00:00
|
|
|
else
|
2000-12-22 23:23:10 +00:00
|
|
|
g_print ("gstpad: could not check capabilities of pads (%s:%s) and (%s:%s)\n",
|
|
|
|
GST_DEBUG_PAD_NAME (srcpad), GST_DEBUG_PAD_NAME (sinkpad));
|
2000-01-30 09:03:00 +00:00
|
|
|
|
|
|
|
/* first set peers */
|
|
|
|
srcpad->peer = sinkpad;
|
|
|
|
sinkpad->peer = srcpad;
|
|
|
|
|
|
|
|
/* now copy the chain pointer from sink to src */
|
2000-07-17 17:14:15 +00:00
|
|
|
srcpad->chainfunc = sinkpad->chainfunc;
|
2000-02-26 18:55:14 +00:00
|
|
|
/* and the pull function */
|
2000-09-24 14:29:49 +00:00
|
|
|
//srcpad->pullfunc = sinkpad->pullfunc;
|
2000-01-30 09:03:00 +00:00
|
|
|
|
|
|
|
/* set the connected flag */
|
|
|
|
/* FIXME: set connected flag */
|
|
|
|
}
|
|
|
|
|
2000-10-25 19:09:53 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_set_parent:
|
|
|
|
* @pad: the pad to set the parent
|
|
|
|
* @parent: the object to set the parent to
|
|
|
|
*
|
|
|
|
* sets the parent object of a pad.
|
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
void
|
|
|
|
gst_pad_set_parent (GstPad *pad,
|
|
|
|
GstObject *parent)
|
|
|
|
{
|
|
|
|
g_return_if_fail (pad != NULL);
|
|
|
|
g_return_if_fail (GST_IS_PAD (pad));
|
|
|
|
g_return_if_fail (pad->parent == NULL);
|
|
|
|
g_return_if_fail (parent != NULL);
|
|
|
|
g_return_if_fail (GTK_IS_OBJECT (parent));
|
|
|
|
g_return_if_fail ((gpointer)pad != (gpointer)parent);
|
2000-01-30 09:03:00 +00:00
|
|
|
|
2000-06-25 21:38:00 +00:00
|
|
|
//g_print("set parent %s\n", gst_element_get_name(parent));
|
2000-03-22 21:18:15 +00:00
|
|
|
|
2000-01-30 09:03:00 +00:00
|
|
|
pad->parent = parent;
|
|
|
|
}
|
|
|
|
|
2000-10-25 19:09:53 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_add_ghost_parent:
|
|
|
|
* @pad: the pad to set the ghost parent
|
|
|
|
* @parent: the object to set the ghost parent to
|
|
|
|
*
|
|
|
|
* add a ghost parent object to a pad.
|
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
void
|
|
|
|
gst_pad_add_ghost_parent (GstPad *pad,
|
|
|
|
GstObject *parent)
|
|
|
|
{
|
|
|
|
g_return_if_fail (pad != NULL);
|
|
|
|
g_return_if_fail (GST_IS_PAD (pad));
|
|
|
|
g_return_if_fail (parent != NULL);
|
|
|
|
g_return_if_fail (GTK_IS_OBJECT (parent));
|
2000-01-30 09:03:00 +00:00
|
|
|
|
2000-11-01 13:49:41 +00:00
|
|
|
pad->ghostparents = g_list_prepend (pad->ghostparents, parent);
|
2000-01-30 09:03:00 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2000-10-25 19:09:53 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_remove_ghost_parent:
|
|
|
|
* @pad: the pad to remove the ghost parent
|
|
|
|
* @parent: the object to remove the ghost parent from
|
|
|
|
*
|
|
|
|
* remove a ghost parent object from a pad.
|
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
void
|
|
|
|
gst_pad_remove_ghost_parent (GstPad *pad,
|
|
|
|
GstObject *parent)
|
|
|
|
{
|
|
|
|
g_return_if_fail (pad != NULL);
|
|
|
|
g_return_if_fail (GST_IS_PAD (pad));
|
|
|
|
g_return_if_fail (parent != NULL);
|
|
|
|
g_return_if_fail (GTK_IS_OBJECT (parent));
|
2000-01-30 09:03:00 +00:00
|
|
|
|
2000-11-01 13:49:41 +00:00
|
|
|
pad->ghostparents = g_list_remove (pad->ghostparents, parent);
|
2000-01-30 09:03:00 +00:00
|
|
|
}
|
|
|
|
|
2000-10-25 19:09:53 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_get_parent:
|
|
|
|
* @pad: the pad to get the parent from
|
|
|
|
*
|
|
|
|
* get the parent object of this pad
|
|
|
|
*
|
|
|
|
* Returns: the parent object
|
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
GstObject*
|
|
|
|
gst_pad_get_parent (GstPad *pad)
|
|
|
|
{
|
|
|
|
g_return_val_if_fail (pad != NULL, NULL);
|
|
|
|
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
|
2000-01-30 09:03:00 +00:00
|
|
|
|
|
|
|
return pad->parent;
|
|
|
|
}
|
|
|
|
|
2000-10-25 19:09:53 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_get_ghost_parents:
|
|
|
|
* @pad: the pad to get the ghost parents from
|
|
|
|
*
|
|
|
|
* get the ghost parents of this pad
|
|
|
|
*
|
|
|
|
* Returns: a list of ghost parent objects
|
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
GList*
|
|
|
|
gst_pad_get_ghost_parents (GstPad *pad)
|
|
|
|
{
|
|
|
|
g_return_val_if_fail (pad != NULL, NULL);
|
|
|
|
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
|
2000-01-30 09:03:00 +00:00
|
|
|
|
|
|
|
return pad->ghostparents;
|
|
|
|
}
|
|
|
|
|
2000-12-08 18:24:16 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_set_caps:
|
|
|
|
* @pad: the pad to set the caps to
|
|
|
|
* @caps: the caps to attach to this pad
|
|
|
|
*
|
|
|
|
* set the capabilities of this pad
|
|
|
|
*/
|
|
|
|
void
|
|
|
|
gst_pad_set_caps (GstPad *pad,
|
|
|
|
GstCaps *caps)
|
|
|
|
{
|
|
|
|
g_return_if_fail (pad != NULL);
|
|
|
|
g_return_if_fail (GST_IS_PAD (pad));
|
|
|
|
|
|
|
|
pad->caps = caps;
|
|
|
|
}
|
|
|
|
/**
|
|
|
|
* gst_pad_get_caps:
|
|
|
|
* @pad: the pad to get the capabilities from
|
|
|
|
*
|
|
|
|
* get the capabilities of this pad
|
|
|
|
*
|
|
|
|
* Return; the capabilities of this pad
|
|
|
|
*/
|
|
|
|
GstCaps *
|
|
|
|
gst_pad_get_caps (GstPad *pad)
|
|
|
|
{
|
|
|
|
g_return_val_if_fail (pad != NULL, NULL);
|
|
|
|
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
|
|
|
|
|
|
|
|
return pad->caps;
|
|
|
|
}
|
|
|
|
|
2000-10-25 19:09:53 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_get_peer:
|
|
|
|
* @pad: the pad to get the peer from
|
|
|
|
*
|
|
|
|
* Get the peer pad of this pad
|
|
|
|
*
|
|
|
|
* Returns: the peer pad
|
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
GstPad*
|
|
|
|
gst_pad_get_peer (GstPad *pad)
|
|
|
|
{
|
|
|
|
g_return_val_if_fail (pad != NULL, NULL);
|
|
|
|
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
|
2000-01-30 09:03:00 +00:00
|
|
|
|
|
|
|
return pad->peer;
|
|
|
|
}
|
|
|
|
|
2000-11-01 13:49:41 +00:00
|
|
|
static void
|
|
|
|
gst_pad_real_destroy (GtkObject *object)
|
|
|
|
{
|
|
|
|
GstPad *pad = GST_PAD (object);
|
2000-01-30 09:03:00 +00:00
|
|
|
|
|
|
|
// g_print("in gst_pad_real_destroy()\n");
|
|
|
|
|
|
|
|
if (pad->name)
|
2000-11-01 13:49:41 +00:00
|
|
|
g_free (pad->name);
|
|
|
|
g_list_free (pad->ghostparents);
|
2000-01-30 09:03:00 +00:00
|
|
|
}
|
|
|
|
|
2000-09-27 19:33:10 +00:00
|
|
|
|
|
|
|
/**
|
2000-10-25 19:09:53 +00:00
|
|
|
* gst_pad_load_and_connect:
|
2000-09-27 19:33:10 +00:00
|
|
|
* @parent: the parent XML node to read the description from
|
|
|
|
* @element: the element that has the source pad
|
|
|
|
* @elements: a hashtable with elements
|
|
|
|
*
|
|
|
|
* Read the pad definition from the XML node and connect the given pad
|
|
|
|
* in element to a pad of an element in the hashtable.
|
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
void
|
|
|
|
gst_pad_load_and_connect (xmlNodePtr parent,
|
|
|
|
GstObject *element,
|
|
|
|
GHashTable *elements)
|
|
|
|
{
|
2000-09-27 19:33:10 +00:00
|
|
|
xmlNodePtr field = parent->childs;
|
2000-10-08 22:17:11 +00:00
|
|
|
GstPad *pad = NULL, *targetpad;
|
|
|
|
guchar *peer = NULL;
|
2000-09-27 19:33:10 +00:00
|
|
|
gchar **split;
|
|
|
|
GstElement *target;
|
|
|
|
|
|
|
|
while (field) {
|
|
|
|
if (!strcmp(field->name, "name")) {
|
|
|
|
pad = gst_element_get_pad(GST_ELEMENT(element), xmlNodeGetContent(field));
|
|
|
|
}
|
|
|
|
else if (!strcmp(field->name, "peer")) {
|
|
|
|
peer = g_strdup(xmlNodeGetContent(field));
|
|
|
|
}
|
|
|
|
field = field->next;
|
|
|
|
}
|
|
|
|
g_return_if_fail(pad != NULL);
|
2000-12-16 17:12:28 +00:00
|
|
|
|
|
|
|
if (peer == NULL) return;
|
2000-09-27 19:33:10 +00:00
|
|
|
|
|
|
|
split = g_strsplit(peer, ".", 2);
|
|
|
|
|
|
|
|
g_return_if_fail(split[0] != NULL);
|
|
|
|
g_return_if_fail(split[1] != NULL);
|
|
|
|
|
|
|
|
target = (GstElement *)g_hash_table_lookup(elements, split[0]);
|
|
|
|
|
|
|
|
if (target == NULL) goto cleanup;
|
|
|
|
|
|
|
|
targetpad = gst_element_get_pad(target, split[1]);
|
|
|
|
|
2000-12-08 18:24:16 +00:00
|
|
|
if (targetpad == NULL) goto cleanup;
|
2000-09-27 19:33:10 +00:00
|
|
|
|
|
|
|
gst_pad_connect(pad, targetpad);
|
|
|
|
|
|
|
|
cleanup:
|
|
|
|
g_strfreev(split);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2000-10-25 19:09:53 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_save_thyself:
|
|
|
|
* @pad: the pad to save
|
|
|
|
* @parent: the parent XML node to save the description in
|
|
|
|
*
|
|
|
|
* Saves the pad into an xml representation
|
|
|
|
*
|
|
|
|
* Returns: the xml representation of the pad
|
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
xmlNodePtr
|
|
|
|
gst_pad_save_thyself (GstPad *pad,
|
|
|
|
xmlNodePtr parent)
|
|
|
|
{
|
2000-01-30 09:03:00 +00:00
|
|
|
GstPad *peer;
|
|
|
|
|
2000-09-27 19:33:10 +00:00
|
|
|
xmlNewChild(parent,NULL,"name",pad->name);
|
2000-01-30 09:03:00 +00:00
|
|
|
if (pad->peer != NULL) {
|
|
|
|
peer = pad->peer;
|
|
|
|
// first check to see if the peer's parent's parent is the same
|
2000-09-27 19:33:10 +00:00
|
|
|
//if (pad->parent->parent == peer->parent->parent)
|
2000-01-30 09:03:00 +00:00
|
|
|
// we just save it off
|
2000-09-27 19:33:10 +00:00
|
|
|
xmlNewChild(parent,NULL,"peer",g_strdup_printf("%s.%s",
|
2000-01-30 09:03:00 +00:00
|
|
|
GST_ELEMENT(peer->parent)->name,peer->name));
|
|
|
|
} else
|
2000-09-27 19:33:10 +00:00
|
|
|
xmlNewChild(parent,NULL,"peer","");
|
2000-01-30 09:03:00 +00:00
|
|
|
|
2000-09-27 19:33:10 +00:00
|
|
|
return parent;
|
2000-01-30 09:03:00 +00:00
|
|
|
}
|
|
|
|
|
2000-10-25 19:09:53 +00:00
|
|
|
/**
|
|
|
|
* gst_pad_ghost_save_thyself:
|
|
|
|
* @pad: the pad to save
|
|
|
|
* @bin: the bin
|
|
|
|
* @parent: the parent XML node to save the description in
|
|
|
|
*
|
|
|
|
* Saves the ghost pad into an xml representation
|
|
|
|
*
|
|
|
|
* Returns: the xml representation of the pad
|
|
|
|
*/
|
2000-11-01 13:49:41 +00:00
|
|
|
xmlNodePtr
|
|
|
|
gst_pad_ghost_save_thyself (GstPad *pad,
|
|
|
|
GstElement *bin,
|
|
|
|
xmlNodePtr parent)
|
|
|
|
{
|
2000-01-30 09:03:00 +00:00
|
|
|
xmlNodePtr self;
|
|
|
|
|
|
|
|
self = xmlNewChild(parent,NULL,"ghostpad",NULL);
|
|
|
|
xmlNewChild(self,NULL,"name",pad->name);
|
|
|
|
xmlNewChild(self,NULL,"parent",GST_ELEMENT(pad->parent)->name);
|
|
|
|
|
|
|
|
return self;
|
|
|
|
}
|
2000-12-12 19:29:43 +00:00
|
|
|
|
2000-12-13 19:29:35 +00:00
|
|
|
GstPadTemplate*
|
|
|
|
gst_padtemplate_new (GstPadFactory *factory)
|
2000-12-12 19:29:43 +00:00
|
|
|
{
|
2000-12-13 19:29:35 +00:00
|
|
|
GstPadTemplate *new;
|
|
|
|
GstPadFactoryEntry tag;
|
|
|
|
gint i = 0;
|
2000-12-12 19:29:43 +00:00
|
|
|
|
2000-12-13 19:29:35 +00:00
|
|
|
g_return_val_if_fail (factory != NULL, NULL);
|
2000-12-12 19:29:43 +00:00
|
|
|
|
2000-12-13 19:29:35 +00:00
|
|
|
new = g_new0 (GstPadTemplate, 1);
|
2000-12-12 19:29:43 +00:00
|
|
|
|
2000-12-13 19:29:35 +00:00
|
|
|
tag = (*factory)[i++];
|
|
|
|
g_return_val_if_fail (tag != NULL, new);
|
|
|
|
new->name_template = g_strdup ((gchar *)tag);
|
2000-12-12 19:29:43 +00:00
|
|
|
|
2000-12-13 19:29:35 +00:00
|
|
|
tag = (*factory)[i++];
|
|
|
|
new->direction = GPOINTER_TO_UINT (tag);
|
|
|
|
|
|
|
|
tag = (*factory)[i++];
|
|
|
|
new->presence = GPOINTER_TO_UINT (tag);
|
|
|
|
|
|
|
|
new->caps = gst_caps_register ((GstCapsFactory *)&(*factory)[i]);
|
|
|
|
|
|
|
|
return new;
|
2000-12-12 19:29:43 +00:00
|
|
|
}
|
|
|
|
|
2000-12-13 19:29:35 +00:00
|
|
|
GstPadTemplate*
|
|
|
|
gst_padtemplate_create (gchar *name_template,
|
|
|
|
GstPadDirection direction, GstPadPresence presence,
|
2000-12-15 01:57:34 +00:00
|
|
|
GstCaps *caps)
|
2000-12-12 19:29:43 +00:00
|
|
|
{
|
2000-12-14 17:21:29 +00:00
|
|
|
GstPadTemplate *new;
|
|
|
|
|
|
|
|
new = g_new0 (GstPadTemplate, 1);
|
|
|
|
|
|
|
|
new->name_template = name_template;
|
|
|
|
new->direction = direction;
|
|
|
|
new->presence = presence;
|
|
|
|
new->caps = caps;
|
|
|
|
|
|
|
|
return new;
|
2000-12-12 19:29:43 +00:00
|
|
|
}
|
|
|
|
|
2000-12-13 19:29:35 +00:00
|
|
|
|
2000-12-12 19:29:43 +00:00
|
|
|
xmlNodePtr
|
2000-12-13 19:29:35 +00:00
|
|
|
gst_padtemplate_save_thyself (GstPadTemplate *pad, xmlNodePtr parent)
|
2000-12-12 19:29:43 +00:00
|
|
|
{
|
|
|
|
xmlNodePtr subtree;
|
|
|
|
|
2000-12-13 19:29:35 +00:00
|
|
|
xmlNewChild(parent,NULL,"nametemplate", pad->name_template);
|
2000-12-12 19:29:43 +00:00
|
|
|
xmlNewChild(parent,NULL,"direction", (pad->direction == GST_PAD_SINK? "sink":"src"));
|
|
|
|
xmlNewChild(parent,NULL,"presence", (pad->presence == GST_PAD_ALWAYS? "always":"sometimes"));
|
|
|
|
subtree = xmlNewChild(parent,NULL,"caps", NULL);
|
|
|
|
|
2000-12-13 19:29:35 +00:00
|
|
|
gst_caps_save_thyself (pad->caps, subtree);
|
2000-12-12 19:29:43 +00:00
|
|
|
|
|
|
|
return parent;
|
|
|
|
}
|
|
|
|
|
2000-12-13 19:29:35 +00:00
|
|
|
GstPadTemplate*
|
|
|
|
gst_padtemplate_load_thyself (xmlNodePtr parent)
|
2000-12-12 19:29:43 +00:00
|
|
|
{
|
|
|
|
xmlNodePtr field = parent->childs;
|
2000-12-13 19:29:35 +00:00
|
|
|
GstPadTemplate *factory = g_new0 (GstPadTemplate, 1);
|
2000-12-12 19:29:43 +00:00
|
|
|
|
|
|
|
while (field) {
|
|
|
|
if (!strcmp(field->name, "nametemplate")) {
|
2000-12-13 19:29:35 +00:00
|
|
|
factory->name_template = g_strdup(xmlNodeGetContent(field));
|
2000-12-12 19:29:43 +00:00
|
|
|
}
|
|
|
|
if (!strcmp(field->name, "direction")) {
|
|
|
|
gchar *value = xmlNodeGetContent(field);
|
|
|
|
|
|
|
|
factory->direction = GST_PAD_UNKNOWN;
|
|
|
|
if (!strcmp(value, "sink")) {
|
|
|
|
factory->direction = GST_PAD_SINK;
|
|
|
|
}
|
|
|
|
else if (!strcmp(value, "src")) {
|
|
|
|
factory->direction = GST_PAD_SRC;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (!strcmp(field->name, "presence")) {
|
|
|
|
gchar *value = xmlNodeGetContent(field);
|
|
|
|
|
|
|
|
if (!strcmp(value, "always")) {
|
|
|
|
factory->presence = GST_PAD_ALWAYS;
|
|
|
|
}
|
|
|
|
else if (!strcmp(value, "sometimes")) {
|
|
|
|
factory->presence = GST_PAD_SOMETIMES;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else if (!strcmp(field->name, "caps")) {
|
2000-12-13 19:29:35 +00:00
|
|
|
factory->caps = gst_caps_load_thyself (field);
|
2000-12-12 19:29:43 +00:00
|
|
|
}
|
|
|
|
field = field->next;
|
|
|
|
}
|
|
|
|
return factory;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
WARNING: Don't grab this updated unless you're really, REALLY sure.
Original commit message from CVS:
WARNING: Don't grab this updated unless you're really, REALLY sure.
WARNING: Wait for the next one.
Whole lotta changes here, including a few random bits:
examples/*/Makefile: updated to use `libtool gcc`, not just `gcc`
gst/
gstbuffer.h: updated to new flag style
gst.c, gstdebug.h: added new debugging for function ptrs
gstpipeline.c: set type of parent_class to the class, not the object
gstthread.c: ditto
plugins/
cdparanoia/cdparanoia.c: added an argument type, updated some defaults
cobin/spindentity.c: updated to new do/while loopfunction style
mp3encode/lame/gstlame.c: argument types, whole lotta lame options
tests/: various changes
Now, for the big changes: Once again, the scheduling system has changed.
And once again, it broke a whole bunch of things. The gist of the change
is that there is now a function pointer for gst_pad_push and gst_pad_pull,
instead of a hard-wired function. Well, currently they are functions, but
that's for debugging purposes only, they just call the function pointer
after spewing lots of DEBUG().
This changed the GstPad structure a bit, and the GstPad API as well.
Where elements used to provide chain() and pull() functions, they provide
chain() and get() functions. gst_pad_set_pull[region]_function has been
changed to get_pad_set_get[region]_function. This means all the elements
out there that used to have pull functions need to be updated. The calls
to that function have been changed in the normal elements, but the names
of the functions passed is still _pull[region](), which is an aesthetic
issue more than anything.
As for what doesn't work yet, just about anything dealing with Connections
is hosed, meaning threaded stuff won't work. This will be fixed about 12
hours from now, after I've slept, etc. The simplefake.c test works in
both cothreaded and chained cases, but not much else will work due to the
Connection problem. Needless to say, don't grab this unless you *need*
these features *now*, else wait to update this stuff until tomorrow.
I'm going to sleep now.
2000-12-16 10:18:09 +00:00
|
|
|
#ifndef gst_pad_push
|
|
|
|
void gst_pad_push(GstPad *pad,GstBuffer *buf) {
|
|
|
|
DEBUG_ENTER("(%s:%s)",GST_DEBUG_PAD_NAME(pad));
|
|
|
|
if (pad->peer->pushfunc) {
|
|
|
|
DEBUG("calling pushfunc &%s of peer pad %s:%s\n",
|
|
|
|
GST_DEBUG_FUNCPTR_NAME(pad->peer->pushfunc),GST_DEBUG_PAD_NAME(pad->peer));
|
|
|
|
(pad->peer->pushfunc)(pad->peer,buf);
|
|
|
|
} else
|
|
|
|
DEBUG("no pushfunc\n");
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#ifndef gst_pad_pull
|
|
|
|
GstBuffer *gst_pad_pull(GstPad *pad) {
|
|
|
|
GstPad *peer = pad->peer;
|
|
|
|
DEBUG_ENTER("(%s:%s)",GST_DEBUG_PAD_NAME(pad));
|
|
|
|
if (peer->pullfunc) {
|
|
|
|
DEBUG("calling pullfunc &%s (@%p) of peer pad %s:%s\n",
|
|
|
|
GST_DEBUG_FUNCPTR_NAME(peer->pullfunc),&peer->pullfunc,GST_DEBUG_PAD_NAME(peer));
|
|
|
|
return (peer->pullfunc)(peer);
|
|
|
|
} else {
|
|
|
|
DEBUG("no pullfunc for peer pad %s:%s at %p\n",GST_DEBUG_PAD_NAME(peer),&peer->pullfunc);
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#ifndef gst_pad_pullregion
|
|
|
|
GstBuffer *gst_pad_pullregion(GstPad *pad,gulong offset,gulong size) {
|
|
|
|
DEBUG_ENTER("(%s:%s,%ld,%ld)",GST_DEBUG_PAD_NAME(pad),offset,size);
|
|
|
|
if (pad->peer->pullregionfunc) {
|
|
|
|
DEBUG("calling pullregionfunc &%s of peer pad %s:%s\n",
|
|
|
|
GST_DEBUG_FUNCPTR_NAME(pad->peer->pullregionfunc),GST_DEBUG_PAD_NAME(pad->peer));
|
|
|
|
return (pad->peer->pullregionfunc)(pad->peer,offset,size);
|
|
|
|
} else {
|
|
|
|
DEBUG("no pullregionfunc\n");
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
#endif
|