platform/upstream/gstreamer.git
12 years agocontroller: port the test to new api as well
Stefan Sauer [Tue, 6 Dec 2011 13:23:39 +0000 (14:23 +0100)]
controller: port the test to new api as well

12 years agocontroller: fix gpointer vs. gpointer* mess up
Stefan Sauer [Tue, 6 Dec 2011 13:23:12 +0000 (14:23 +0100)]
controller: fix gpointer vs. gpointer* mess up

12 years agopad: remove GST_FLOW_RESEND
Wim Taymans [Tue, 6 Dec 2011 13:24:15 +0000 (14:24 +0100)]
pad: remove GST_FLOW_RESEND

It is unused and undefined.

12 years agobasesrc: add async start option
Wim Taymans [Tue, 6 Dec 2011 13:01:50 +0000 (14:01 +0100)]
basesrc: add async start option

Add a method to enable async start behaviour. The subclass can then complete the
start operation from any other thread by caling gst_base_src_start_complete().
The base class can wait for the start to complete with
gst_base_src_start_wait().

12 years agofix compilation
Wim Taymans [Tue, 6 Dec 2011 12:58:00 +0000 (13:58 +0100)]
fix compilation

12 years agofilesrc: cleanup error path
Wim Taymans [Tue, 6 Dec 2011 12:47:29 +0000 (13:47 +0100)]
filesrc: cleanup error path

12 years agofix compilation
Wim Taymans [Tue, 6 Dec 2011 12:39:05 +0000 (13:39 +0100)]
fix compilation

12 years agocontroller: use _OBJECT logging variants more
Stefan Sauer [Tue, 6 Dec 2011 07:48:57 +0000 (08:48 +0100)]
controller: use _OBJECT logging variants more

12 years agocontroller: remove GstValueArray
Stefan Sauer [Tue, 6 Dec 2011 07:35:57 +0000 (08:35 +0100)]
controller: remove GstValueArray

Instead pass the values as arguments. This simplifies that code and helps
bindings.

12 years agocontroller: remove gst_object_get_value_arrays
Stefan Sauer [Tue, 6 Dec 2011 07:35:10 +0000 (08:35 +0100)]
controller: remove gst_object_get_value_arrays

One can easilly loop over the controlled properties manually. This is step 1 in
removing GstValueArray.

12 years agocorrect return value in gst_push_sticky
Matej Knopp [Sun, 4 Dec 2011 06:33:32 +0000 (07:33 +0100)]
correct return value in gst_push_sticky

12 years agomake some macros into inline functions
Wim Taymans [Mon, 5 Dec 2011 10:07:24 +0000 (11:07 +0100)]
make some macros into inline functions

12 years agomake some more macros as inline functions
Wim Taymans [Mon, 5 Dec 2011 09:24:34 +0000 (10:24 +0100)]
make some more macros as inline functions

Make some macros as inline functions for added type checking.
USe new gst_caps_take() in typefind

12 years agodocs: remove some old methods
Wim Taymans [Mon, 5 Dec 2011 09:23:04 +0000 (10:23 +0100)]
docs: remove some old methods

12 years agodocs: remove removed tag list functions from docs as well
Tim-Philipp Müller [Sun, 4 Dec 2011 15:42:07 +0000 (15:42 +0000)]
docs: remove removed tag list functions from docs as well

12 years agoMerge remote-tracking branch 'origin/master' into 0.11
Tim-Philipp Müller [Sun, 4 Dec 2011 15:38:09 +0000 (15:38 +0000)]
Merge remote-tracking branch 'origin/master' into 0.11

Conflicts:
gst/gstobject.h
libs/gst/check/gstcheck.h
libs/gst/controller/gstcontroller.c
plugins/elements/gstidentity.c
tools/gst-xmlinspect.c

12 years agoSuppress deprecation warnings in selected files, for g_static_rec_mutex_* mostly
Tim-Philipp Müller [Sun, 4 Dec 2011 14:38:26 +0000 (14:38 +0000)]
Suppress deprecation warnings in selected files, for g_static_rec_mutex_* mostly

GStaticRecMutex is part of our API/ABI, not much we can do here in 0.10.

12 years agoWork around deprecated thread API in glib master
Tim-Philipp Müller [Sun, 4 Dec 2011 13:35:38 +0000 (13:35 +0000)]
Work around deprecated thread API in glib master

Add private replacements for deprecated functions such as
g_mutex_new(), g_mutex_free(), g_cond_new() etc., mostly
to avoid the deprecation warnings. We can't change most of
these in 0.10 because they're part of our API and ABI.

12 years agobenchmarks: g_thread_create() is deprecated in GLib master, use g_thread_try_new...
Tim-Philipp Müller [Sun, 4 Dec 2011 13:09:53 +0000 (13:09 +0000)]
benchmarks: g_thread_create() is deprecated in GLib master, use g_thread_try_new() instead

12 years agonet: initialise GError variables to NULL
Tim-Philipp Müller [Sun, 4 Dec 2011 13:04:35 +0000 (13:04 +0000)]
net: initialise GError variables to NULL

12 years agowin32: Update defs files
Edward Hervey [Sun, 4 Dec 2011 10:43:10 +0000 (11:43 +0100)]
win32: Update defs files

12 years agogstquery: Fix unitialized variable
Edward Hervey [Sun, 4 Dec 2011 10:42:39 +0000 (11:42 +0100)]
gstquery: Fix unitialized variable

12 years agogstsegment: Initialize with proper type
Edward Hervey [Sun, 4 Dec 2011 10:32:57 +0000 (11:32 +0100)]
gstsegment: Initialize with proper type

12 years agog_thread_init() is deprecated in glib master
Tim-Philipp Müller [Sat, 3 Dec 2011 17:40:53 +0000 (17:40 +0000)]
g_thread_init() is deprecated in glib master

It's not needed any longer.

12 years agotagsetter: update for thread API deprecations in glib master
Tim-Philipp Müller [Sat, 3 Dec 2011 16:02:36 +0000 (16:02 +0000)]
tagsetter: update for thread API deprecations in glib master

12 years agotaglist: update for thread API deprecations in glib master
Tim-Philipp Müller [Sat, 3 Dec 2011 15:36:58 +0000 (15:36 +0000)]
taglist: update for thread API deprecations in glib master

12 years agotaglist: remove gst_tag_list_get_{char,uchar}
Tim-Philipp Müller [Sat, 3 Dec 2011 15:18:21 +0000 (15:18 +0000)]
taglist: remove gst_tag_list_get_{char,uchar}

Those are unused and should never be used anywhere anyway
really.

12 years agodocs: remove some macros that no longer exist
Tim-Philipp Müller [Sat, 3 Dec 2011 14:06:58 +0000 (14:06 +0000)]
docs: remove some macros that no longer exist

12 years agog_thread_create() is deprecated in GLib master, use g_thread_try_new() instead
Tim-Philipp Müller [Sat, 3 Dec 2011 13:58:51 +0000 (13:58 +0000)]
g_thread_create() is deprecated in GLib master, use g_thread_try_new() instead

12 years agobasesink: use dts and pts for sync
Wim Taymans [Sat, 3 Dec 2011 06:06:11 +0000 (07:06 +0100)]
basesink: use dts and pts for sync

First use DTS, then fall back to PTS for synchronization.

12 years agobasesink: small cleanups
Wim Taymans [Sat, 3 Dec 2011 06:01:45 +0000 (07:01 +0100)]
basesink: small cleanups

12 years agobasesink: merge preroll functions
Wim Taymans [Sat, 3 Dec 2011 05:45:26 +0000 (06:45 +0100)]
basesink: merge preroll functions

Inline a function that is only called from one place to make things a little
easier to follow.

12 years agobasesink: more cleanups
Wim Taymans [Sat, 3 Dec 2011 05:29:42 +0000 (06:29 +0100)]
basesink: more cleanups

Don't pass around the object type, we can find that very efficiently from the
object itself now.

12 years agobasesink: clean up method names
Wim Taymans [Fri, 2 Dec 2011 22:13:23 +0000 (23:13 +0100)]
basesink: clean up method names

12 years agobasesink: merge render_object into chain
Wim Taymans [Fri, 2 Dec 2011 21:50:55 +0000 (22:50 +0100)]
basesink: merge render_object into chain

Merge the render_object code with the chain method. It is only called from there
and there are quite a few variables that can be reused to makes things less
confusing.

12 years agobasesink: remove obsolete code
Wim Taymans [Fri, 2 Dec 2011 21:36:38 +0000 (22:36 +0100)]
basesink: remove obsolete code

Remove some more code now that the render_object ethod is only
called with buffers or bufferlsts.

12 years agobasesink: clean up event handling
Wim Taymans [Fri, 2 Dec 2011 21:20:08 +0000 (22:20 +0100)]
basesink: clean up event handling

Add new wait_eos vmethod to wait for the eos timeout before posting the EOS
message on the bus.
Add default event handler. Move the default event actions in there. Call the
event vmethod from the pad event handler. Subclasses are now supposed to chain
up to the parent event handler or unref the event and do their own thing.
Avoid passing unused parameters to functions.

12 years agoidentity: unlock clock wait when appropriate
Mark Nauwelaerts [Fri, 2 Dec 2011 12:19:38 +0000 (13:19 +0100)]
identity: unlock clock wait when appropriate

... notably FLUSH and state change to READY.

12 years agobasesink: small cleanup
Wim Taymans [Fri, 2 Dec 2011 12:35:58 +0000 (13:35 +0100)]
basesink: small cleanup

Avoid passing around the segment.

12 years agobasesink: remove clip_segment
Wim Taymans [Fri, 2 Dec 2011 12:28:18 +0000 (13:28 +0100)]
basesink: remove clip_segment

We only need one segment now that the preroll queue is gone.

12 years agobasesink: more cleanups
Wim Taymans [Fri, 2 Dec 2011 11:42:50 +0000 (12:42 +0100)]
basesink: more cleanups

Remove some unneeded functions, inline the code.
Remove the queue_object functions, we can proceed with the rendering
immediately.

12 years agobasesink: remove obsolete code
Wim Taymans [Fri, 2 Dec 2011 11:20:02 +0000 (12:20 +0100)]
basesink: remove obsolete code

Remove the preroll queue and proceed directly to the rendering of objects.

12 years agoMerge remote-tracking branch 'origin/master' into 0.11
Tim-Philipp Müller [Thu, 1 Dec 2011 23:35:26 +0000 (23:35 +0000)]
Merge remote-tracking branch 'origin/master' into 0.11

Conflicts:
gst/gstbus.c
gst/gstevent.c
libs/gst/base/gstbasetransform.c

12 years agotaglist: make some tags of type GstSample
Wim Taymans [Thu, 1 Dec 2011 17:50:32 +0000 (18:50 +0100)]
taglist: make some tags of type GstSample

Make the image and attachment tags of type GstSample so that we can include
extra caps and info along with the buffer data.

12 years agosample: remove const
Wim Taymans [Thu, 1 Dec 2011 17:49:34 +0000 (18:49 +0100)]
sample: remove const

The writability of the structure is ensured by the refcount of the parent
miniobject and we're fine if the parent is writable.

12 years agobasesink: last-buffer -> last-sample
Wim Taymans [Thu, 1 Dec 2011 15:46:06 +0000 (16:46 +0100)]
basesink: last-buffer -> last-sample

Rename the last-buffer property to last-sample and make it return the new
GstSample type so that we can include caps and timing info in one nice bundle.

12 years agosample: add new sample miniobject
Wim Taymans [Thu, 1 Dec 2011 15:37:46 +0000 (16:37 +0100)]
sample: add new sample miniobject

Add a new simple miniobject that is a combination of a GstBuffer, GstCaps,
GstSegment and other arbitrary info organized in a GstStructure. This object can
be used to exchange samples between an element and the application or for
storing album art in tags etc.

12 years agobus: use GST_MESSAGE_SOURCE_NAME() which also takes care of src=NULL.
Stefan Sauer [Thu, 1 Dec 2011 15:25:07 +0000 (16:25 +0100)]
bus: use GST_MESSAGE_SOURCE_NAME() which also takes care of src=NULL.

12 years agobasetrans: add some more debug
Wim Taymans [Thu, 1 Dec 2011 14:35:26 +0000 (15:35 +0100)]
basetrans: add some more debug

12 years agobasesrc: add some more debug
Wim Taymans [Thu, 1 Dec 2011 14:35:02 +0000 (15:35 +0100)]
basesrc: add some more debug

12 years agobuffer: add copy flag for meta
Wim Taymans [Thu, 1 Dec 2011 14:34:06 +0000 (15:34 +0100)]
buffer: add copy flag for meta

Add a flag to control if the meta should be copied or not instead of always
copying.

12 years agobasesink: remove old property
Wim Taymans [Tue, 29 Nov 2011 18:08:41 +0000 (19:08 +0100)]
basesink: remove old property

12 years agobasetransform: Always intersect the suggested sink caps with the peer caps
Sebastian Dröge [Wed, 30 Nov 2011 12:59:46 +0000 (13:59 +0100)]
basetransform: Always intersect the suggested sink caps with the peer caps

This makes sure that we get correct and complete caps. The suggested caps
could be incomplete, e.g. video/x-raw-rgb without any fields, and by
intersecting with the peer caps we get something usable.

Fixes bug #662199.

12 years agoinfo: move FIXME log level after WARNING
Tim-Philipp Müller [Wed, 30 Nov 2011 12:39:34 +0000 (12:39 +0000)]
info: move FIXME log level after WARNING

So it's now ERROR < WARNING < FIXME < INFO and *:5 becomes *:6.

12 years agoevent: warn and fail instead of creating newsegment events in GST_FORMAT_UNDEFINED
Tim-Philipp Müller [Wed, 30 Nov 2011 00:24:32 +0000 (00:24 +0000)]
event: warn and fail instead of creating newsegment events in GST_FORMAT_UNDEFINED

12 years agobin: keep the element flags up-to-date
Wim Taymans [Tue, 29 Nov 2011 14:53:01 +0000 (15:53 +0100)]
bin: keep the element flags up-to-date

Keep the require/provide_clock flags up to date.

12 years agoqueue: source and sink pads proxy caps
Edward Hervey [Tue, 29 Nov 2011 10:47:34 +0000 (11:47 +0100)]
queue: source and sink pads proxy caps

12 years agoMerge remote-tracking branch 'origin/master' into 0.11
Tim-Philipp Müller [Mon, 28 Nov 2011 21:15:31 +0000 (21:15 +0000)]
Merge remote-tracking branch 'origin/master' into 0.11

12 years agoupdate for indexable change
Wim Taymans [Mon, 28 Nov 2011 17:23:41 +0000 (18:23 +0100)]
update for indexable change

12 years agoelement: add indexable flag
Wim Taymans [Mon, 28 Nov 2011 17:12:34 +0000 (18:12 +0100)]
element: add indexable flag

Remove the is_indexable method check and use an element flag to check if the
element can use an index.

12 years agodefs: update
Wim Taymans [Mon, 28 Nov 2011 16:50:17 +0000 (17:50 +0100)]
defs: update

12 years agoelement: use flags for require/provide clock
Wim Taymans [Mon, 28 Nov 2011 16:22:44 +0000 (17:22 +0100)]
element: use flags for require/provide clock

Remove the _require/_provide_clock() methods and use element flags to mark
elements instead of looking at the implementation of the vmethod.

12 years agoelement: clean up element flags
Wim Taymans [Mon, 28 Nov 2011 15:54:55 +0000 (16:54 +0100)]
element: clean up element flags

Clean up the element flags

12 years agoevent: add sticky custom quark
Wim Taymans [Mon, 28 Nov 2011 14:35:21 +0000 (15:35 +0100)]
event: add sticky custom quark

12 years agocaps: _CAPS_FLAGS_ -> CAPS_FLAG_
Wim Taymans [Mon, 28 Nov 2011 13:24:16 +0000 (14:24 +0100)]
caps: _CAPS_FLAGS_ -> CAPS_FLAG_

12 years agogstelement: add gst_element_class_add_pad_template_from_static
Vincent Penquerc'h [Mon, 28 Nov 2011 12:30:15 +0000 (12:30 +0000)]
gstelement: add gst_element_class_add_pad_template_from_static

This function helps ensure the pad template is unreffed
without having to complicate the calling code.

https://bugzilla.gnome.org/show_bug.cgi?id=662664

12 years agopad: Handle sticky event errors
Wim Taymans [Mon, 28 Nov 2011 12:54:43 +0000 (13:54 +0100)]
pad: Handle sticky event errors

Use GstFlowReturn to internally pass events between pads.
When we sticky events cause an error, translate this error into a GstFlowReturn.
Caps events will, for example, generate a NOT_NEGOTIATED return when the event
function returns an error.
This allows us then to refuse sending buffers if one of the sticky events is
refused and generate a correct error return value.

12 years agotests: don't return FALSE from events
Wim Taymans [Mon, 28 Nov 2011 12:52:00 +0000 (13:52 +0100)]
tests: don't return FALSE from events

Returning FALSE from the event handler shuts down the sender.

12 years agotest: fix refcount error
Wim Taymans [Mon, 28 Nov 2011 12:51:11 +0000 (13:51 +0100)]
test: fix refcount error

12 years agogstmeta: Some more docs
Edward Hervey [Mon, 28 Nov 2011 10:15:27 +0000 (11:15 +0100)]
gstmeta: Some more docs

12 years agoevent: add custom downstream sticky event
Wim Taymans [Mon, 28 Nov 2011 09:55:28 +0000 (10:55 +0100)]
event: add custom downstream sticky event

12 years agoqueue2: fix up comment after merge from 0.10
Tim-Philipp Müller [Mon, 28 Nov 2011 01:12:48 +0000 (01:12 +0000)]
queue2: fix up comment after merge from 0.10

12 years agoMerge remote-tracking branch 'origin/master' into 0.11
Tim-Philipp Müller [Mon, 28 Nov 2011 01:11:47 +0000 (01:11 +0000)]
Merge remote-tracking branch 'origin/master' into 0.11

12 years agoqueue2: fix refactoring of draining-on-eos, munge flow return to FLOW_OK
Tim-Philipp Müller [Mon, 28 Nov 2011 01:10:20 +0000 (01:10 +0000)]
queue2: fix refactoring of draining-on-eos, munge flow return to FLOW_OK

12 years agoqueue2: fix up new bufferlist code for 0.11
Tim-Philipp Müller [Mon, 28 Nov 2011 01:00:28 +0000 (01:00 +0000)]
queue2: fix up new bufferlist code for 0.11

12 years agoMerge remote-tracking branch 'origin/master' into 0.11
Tim-Philipp Müller [Mon, 28 Nov 2011 00:40:55 +0000 (00:40 +0000)]
Merge remote-tracking branch 'origin/master' into 0.11

Conflicts:
plugins/elements/gstqueue2.c

12 years agoqueue2: add bufferlist support
Tim-Philipp Müller [Thu, 3 Nov 2011 10:34:49 +0000 (10:34 +0000)]
queue2: add bufferlist support

We want to maintain buffer lists if possible.

12 years agoqueue2: split out draining of queue on FLOW_UNEXPECTED into separate function
Tim-Philipp Müller [Thu, 3 Nov 2011 13:02:36 +0000 (13:02 +0000)]
queue2: split out draining of queue on FLOW_UNEXPECTED into separate function

12 years agoqueue2: pass item type enum to _enqueue instead of simple isbuffer boolean
Tim-Philipp Müller [Thu, 3 Nov 2011 08:55:20 +0000 (08:55 +0000)]
queue2: pass item type enum to _enqueue instead of simple isbuffer boolean

Avoids some unnecessary GST_IS_EVENT()

12 years agocaps: fix compilation warning
Matej Knopp [Sun, 27 Nov 2011 19:32:14 +0000 (20:32 +0100)]
caps: fix compilation warning

GST_STATIC_CAPS is missing initializer for GstMiniObject's n_weak_refs and
weak_refs resulting in compilation warning (llvm-gcc  -Wall)

https://bugzilla.gnome.org/show_bug.cgi?id=664927

12 years agoMerge remote-tracking branch 'origin/master' into 0.11
Tim-Philipp Müller [Sun, 27 Nov 2011 22:26:12 +0000 (22:26 +0000)]
Merge remote-tracking branch 'origin/master' into 0.11

12 years agowarnings: avoid set-but-unused warnings with load-save disabled
Stefan Sauer [Sun, 27 Nov 2011 19:36:31 +0000 (20:36 +0100)]
warnings: avoid set-but-unused warnings with load-save disabled

12 years agobaseparse: fix broken default caps query
Matej Knopp [Sat, 26 Nov 2011 16:34:12 +0000 (17:34 +0100)]
baseparse: fix broken default caps query

https://bugzilla.gnome.org/show_bug.cgi?id=664880

12 years agoMerge remote-tracking branch 'origin/master' into 0.11
Tim-Philipp Müller [Sat, 26 Nov 2011 19:51:37 +0000 (19:51 +0000)]
Merge remote-tracking branch 'origin/master' into 0.11

12 years agouri: fix wrong G_GNUC_MALLOC
Tim-Philipp Müller [Sat, 26 Nov 2011 19:45:48 +0000 (19:45 +0000)]
uri: fix wrong G_GNUC_MALLOC

_get_protocols() points to const memory in 0.10
despite the non-const return value.

12 years agoMerge remote-tracking branch 'origin/master' into 0.11
Tim-Philipp Müller [Sat, 26 Nov 2011 19:44:23 +0000 (19:44 +0000)]
Merge remote-tracking branch 'origin/master' into 0.11

Conflicts:
gst/gstbuffer.h
gst/gstbufferlist.h
gst/gstcaps.h
gst/gstdatetime.h
gst/gstelementfactory.h
gst/gstevent.h
gst/gstghostpad.h
gst/gstindexfactory.h
gst/gstiterator.h
gst/gstmessage.h
gst/gstminiobject.h
gst/gstpipeline.h
gst/gstquery.h
gst/gstsegment.h
gst/gststructure.h
gst/gsttaglist.h
gst/gsturi.h
gst/gstvalue.h
libs/gst/base/gstbitreader.h
libs/gst/base/gstbytereader.h
libs/gst/base/gstbytewriter.h

Note: can't use G_GNUC_MALLOC with GstCaps return
values in 0.11 because of the EMPTY+ANY singletons.

12 years agolibs: sprinkle some G_GNUC_MALLOC
Tim-Philipp Müller [Sat, 26 Nov 2011 18:58:38 +0000 (18:58 +0000)]
libs: sprinkle some G_GNUC_MALLOC

Maybe gcc can do something clever with that, or at least
warn us if we don't save the return value somewhere.

12 years agogst: sprinkle some G_GNUC_MALLOC
Tim-Philipp Müller [Sat, 26 Nov 2011 18:57:44 +0000 (18:57 +0000)]
gst: sprinkle some G_GNUC_MALLOC

Maybe gcc can do something clever with that, or at least
warn us if we don't save the return value somewhere.

12 years agoMerge remote-tracking branch 'origin/master' into 0.11
Tim-Philipp Müller [Fri, 25 Nov 2011 23:54:57 +0000 (23:54 +0000)]
Merge remote-tracking branch 'origin/master' into 0.11

12 years agodocs: update porting docs.
Stefan Sauer [Fri, 25 Nov 2011 22:15:23 +0000 (23:15 +0100)]
docs: update porting docs.

12 years agobuffer: add gst_buffer_{set,get}_qdata()
Tim-Philipp Müller [Wed, 16 Nov 2011 01:04:45 +0000 (01:04 +0000)]
buffer: add gst_buffer_{set,get}_qdata()

Allows people/us to attach arbitrary metadata to buffers.

https://bugzilla.gnome.org/show_bug.cgi?id=664720

API: gst_buffer_set_qdata()
API: get_buffer_get_qdata()

12 years agopad: fix blocking probe emission
Wim Taymans [Fri, 25 Nov 2011 06:11:24 +0000 (07:11 +0100)]
pad: fix blocking probe emission

If we are dealing with a blocking probe, only then check if one the
blocking flags of the hook matches.
Add some more debug.
Make the pad unit test less racy.

12 years agoAdd test for PAD_PROBE_TYPE_BLOCK and PAD_PROBE_TYPE_BLOCKING
Matej Knopp [Thu, 24 Nov 2011 16:47:09 +0000 (17:47 +0100)]
Add test for PAD_PROBE_TYPE_BLOCK and PAD_PROBE_TYPE_BLOCKING

12 years agoutils: fix debug of query result
Wim Taymans [Fri, 25 Nov 2011 04:54:09 +0000 (05:54 +0100)]
utils: fix debug of query result

12 years agoquery: fix typo in doc
René Stadler [Thu, 24 Nov 2011 21:52:19 +0000 (22:52 +0100)]
query: fix typo in doc

Causes a warning from the introspection scanner.

12 years agocheck: drop caps argument from gst_check_setup_{src,sink}_pad
René Stadler [Thu, 24 Nov 2011 20:36:12 +0000 (21:36 +0100)]
check: drop caps argument from gst_check_setup_{src,sink}_pad

Calling set_caps at that point is not useful in 0.10 (FIXME comment!), and in
0.11 it is totally pointless: the caps event doesn't stick to a flushing pad.

12 years agodefs: update for byte_writer_put_buffer
René Stadler [Thu, 24 Nov 2011 13:07:14 +0000 (14:07 +0100)]
defs: update for byte_writer_put_buffer

12 years agoMerge branch 'master' into 0.11
Sebastian Dröge [Thu, 24 Nov 2011 10:23:07 +0000 (11:23 +0100)]
Merge branch 'master' into 0.11

Conflicts:
gst/gstpad.c
libs/gst/base/gstbaseparse.c

12 years agopad: Remove g_warning() if pad accepted caps that are not a subset of the pad caps
Sebastian Dröge [Thu, 24 Nov 2011 10:15:29 +0000 (11:15 +0100)]
pad: Remove g_warning() if pad accepted caps that are not a subset of the pad caps

This check is correct but unfortunately it's impossible to implement
in a threadsafe way because the caps could have changed in the meantime.

Fixes bug #659606.

12 years agomultiqueue: check filled state of queues even if another one is empty
Vincent Penquerc'h [Mon, 3 Oct 2011 11:34:20 +0000 (12:34 +0100)]
multiqueue: check filled state of queues even if another one is empty

This will avoid a case where overrun is never signalled if some
stream never produces any data, causing playbin2 to not end preroll.

https://bugzilla.gnome.org/show_bug.cgi?id=660778