platform/upstream/gstreamer.git
11 years agopad: clarify docs
Wim Taymans [Thu, 4 Apr 2013 13:45:23 +0000 (15:45 +0200)]
pad: clarify docs

11 years agopad: improve debug
Wim Taymans [Thu, 4 Apr 2013 13:45:15 +0000 (15:45 +0200)]
pad: improve debug

11 years agosegment: don't WARN, just DEBUG
Wim Taymans [Thu, 4 Apr 2013 08:17:30 +0000 (10:17 +0200)]
segment: don't WARN, just DEBUG

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=696911

11 years agosegment: Fix seeking when position is slightly outside the segment
Nicolas Dufresne [Sat, 30 Mar 2013 11:06:59 +0000 (11:06 +0000)]
segment: Fix seeking when position is slightly outside the segment

Very often, when the end of a segment is detected by demuxer, the position
is slightly outside the segment boundaries. Currently, if that is the case
the base will be set to NONE instead of normal accumulation. This would
break non-flushing seeks in oggdemux and most likely other demuxers.

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=696899

11 years agodefs: update
Wim Taymans [Wed, 3 Apr 2013 15:29:51 +0000 (17:29 +0200)]
defs: update

11 years agogst: add stream flags
Wim Taymans [Wed, 3 Apr 2013 14:02:34 +0000 (16:02 +0200)]
gst: add stream flags

11 years agoporting-to-1.0.txt: subtitle text media types changed as well
Brendan Long [Wed, 3 Apr 2013 00:17:00 +0000 (18:17 -0600)]
porting-to-1.0.txt: subtitle text media types changed as well

https://bugzilla.gnome.org/show_bug.cgi?id=697153

11 years agodocs: more since markers and other docs fixes
Tim-Philipp Müller [Tue, 2 Apr 2013 22:51:06 +0000 (23:51 +0100)]
docs: more since markers and other docs fixes

11 years agodocs: add since markers to capsfeatures docs
Tim-Philipp Müller [Tue, 2 Apr 2013 22:21:39 +0000 (23:21 +0100)]
docs: add since markers to capsfeatures docs

11 years agoscripts: add some more modules to git-update.sh
Tim-Philipp Müller [Tue, 2 Apr 2013 22:18:42 +0000 (23:18 +0100)]
scripts: add some more modules to git-update.sh

https://bugzilla.gnome.org/show_bug.cgi?id=697058

11 years agogst-uninstalled: add uninstalled orc/orc-test/.libs to library paths
Tim-Philipp Müller [Tue, 2 Apr 2013 22:04:51 +0000 (23:04 +0100)]
gst-uninstalled: add uninstalled orc/orc-test/.libs to library paths

This is needed by the unit tests in gst-plugins-base, -good etc.

Spotted by Alex Kaye.

https://bugzilla.gnome.org/show_bug.cgi?id=697093

11 years agocapsfeatures: Add GST_CAPS_FEATURES_ANY
Sebastian Dröge [Tue, 2 Apr 2013 20:13:22 +0000 (22:13 +0200)]
capsfeatures: Add GST_CAPS_FEATURES_ANY

This is equal to any other caps features but results in unfixed caps. It
would be used by elements that only look at the buffer metadata or are
currently working in passthrough mode, and as such don't care about any
specific features.

11 years agoinfo: minor cosmetic changes
Tim-Philipp Müller [Mon, 1 Apr 2013 21:13:10 +0000 (22:13 +0100)]
info: minor cosmetic changes

Let's not use ugly leading underscores for
static functions.

11 years agoinfo: fix object printing of caps features in debug log
Tim-Philipp Müller [Mon, 1 Apr 2013 20:23:21 +0000 (21:23 +0100)]
info: fix object printing of caps features in debug log

11 years agodraft-tracing: update draft design with timer ideas
Stefan Sauer [Mon, 1 Apr 2013 14:38:43 +0000 (16:38 +0200)]
draft-tracing: update draft design with timer ideas

11 years agoinfo: refactor pretty printing objects
Stefan Sauer [Sat, 30 Mar 2013 16:03:44 +0000 (17:03 +0100)]
info: refactor pretty printing objects

Extract formatters into local functions. Change the structure filtering so that
it works for taglists too.

11 years agoelement: make post_message and query more alike
Stefan Sauer [Sat, 30 Mar 2013 15:39:38 +0000 (16:39 +0100)]
element: make post_message and query more alike

11 years agogst-inspect: Print caps features too
Sebastian Dröge [Mon, 1 Apr 2013 08:20:01 +0000 (10:20 +0200)]
gst-inspect: Print caps features too

11 years agocapsfeatures: For copying features it's not required to have no parent refcount
Sebastian Dröge [Mon, 1 Apr 2013 08:19:25 +0000 (10:19 +0200)]
capsfeatures: For copying features it's not required to have no parent refcount

11 years agocaps: Set features' parent refcount in gst_caps_set_features() too
Sebastian Dröge [Mon, 1 Apr 2013 08:19:01 +0000 (10:19 +0200)]
caps: Set features' parent refcount in gst_caps_set_features() too

11 years agocaps: Set sysmem features if explicitely requested
Sebastian Dröge [Mon, 1 Apr 2013 08:18:39 +0000 (10:18 +0200)]
caps: Set sysmem features if explicitely requested

11 years agodesign: Add the caps features and describe how the caps operations actually work
Sebastian Dröge [Sun, 31 Mar 2013 17:09:46 +0000 (19:09 +0200)]
design: Add the caps features and describe how the caps operations actually work

11 years agocaps: Add new data type for handling caps features to the caps
Sebastian Dröge [Sat, 30 Mar 2013 14:35:19 +0000 (15:35 +0100)]
caps: Add new data type for handling caps features to the caps

These are meant to specify features in caps that are required
for a specific structure, for example a specific memory type
or meta.

Semantically they could be though of as an extension of the media
type name of the structures and are handled exactly like that.

11 years agoevent: add SELECT and UNSELECT stream flags for stream-start event
Tim-Philipp Müller [Sun, 31 Mar 2013 14:30:19 +0000 (15:30 +0100)]
event: add SELECT and UNSELECT stream flags for stream-start event

So demuxers can signal which audio/video/subtitle streams should
be selected by default and which should not be selected
automatically.

API: GST_STREAM_FLAG_SELECT
API: GST_STREAM_FLAG_UNSELECT

https://bugzilla.gnome.org/show_bug.cgi?id=695968
https://bugzilla.gnome.org/show_bug.cgi?id=690911

11 years agoevent: add stream flags to stream-start event
Tim-Philipp Müller [Sun, 6 Jan 2013 20:27:54 +0000 (20:27 +0000)]
event: add stream flags to stream-start event

API: gst_event_set_stream_flags()
API: gst_event_parse_stream_flags()
API: GST_STREAM_FLAG_NONE
API: GST_STREAM_FLAG_SPARSE

https://bugzilla.gnome.org/show_bug.cgi?id=600648

11 years agoelement: Add API to get the last set context from an element
Sebastian Dröge [Sun, 31 Mar 2013 09:26:32 +0000 (11:26 +0200)]
element: Add API to get the last set context from an element

Elements should override GstElement::set_context() and also call
gst_element_set_context() to keep this context up-to-date with
the very latest context they internally use.

11 years agodesign: add initial tracing design doc
Stefan Sauer [Sat, 30 Mar 2013 13:04:28 +0000 (14:04 +0100)]
design: add initial tracing design doc

11 years agopad: Let gst_pad_get_allowed_caps() pass the caps own caps as filter
Sebastian Dröge [Sat, 30 Mar 2013 10:47:29 +0000 (11:47 +0100)]
pad: Let gst_pad_get_allowed_caps() pass the caps own caps as filter

This will reduce the number of caps created downstream and we don't
need to intersect in the very end with the pad's own caps.

11 years agobasesink: fixes compiler warning
Josep Torra [Sat, 30 Mar 2013 09:24:27 +0000 (10:24 +0100)]
basesink: fixes compiler warning

gstbasesink.c: In function 'gst_base_sink_chain_unlocked':
gstbasesink.c:3204: warning: 'ret' may be used uninitialized in this function

11 years agochildproxy: fix gir warning
Stefan Sauer [Fri, 29 Mar 2013 18:01:59 +0000 (19:01 +0100)]
childproxy: fix gir warning

11 years agoRevert "pwg: Fix example"
Edward Hervey [Fri, 29 Mar 2013 17:49:14 +0000 (18:49 +0100)]
Revert "pwg: Fix example"

This reverts commit 5d64f27d881274a40f0441bb8c5b3816fdfc5b9e.

*sigh*

11 years agopwg: Fix example
Edward Hervey [Tue, 6 Nov 2012 08:41:58 +0000 (09:41 +0100)]
pwg: Fix example

11 years agopad: Remove SEGMENT sticky events when flushing
Edward Hervey [Sun, 11 Nov 2012 12:52:25 +0000 (13:52 +0100)]
pad: Remove SEGMENT sticky events when flushing

When flushing, it is expected that upstream will send a SEGMENT
event afterwards.
This also avoids stray SEGMENT events from coming through after a
flush.

11 years agogst: Add new GstContext miniobject for sharing contexts in a pipeline
Sebastian Dröge [Thu, 28 Mar 2013 14:35:13 +0000 (15:35 +0100)]
gst: Add new GstContext miniobject for sharing contexts in a pipeline

11 years agoargb-controlbinding: fix messed up property setter
Stefan Sauer [Thu, 28 Mar 2013 15:42:50 +0000 (16:42 +0100)]
argb-controlbinding: fix messed up property setter

This was misassigning the control sources. It was also leaking the old values if
a control source would be replaced.

11 years agobaseparse: reset next_pts upon SEGMENT event
Mark Nauwelaerts [Wed, 27 Mar 2013 17:25:08 +0000 (18:25 +0100)]
baseparse: reset next_pts upon SEGMENT event

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=691481

11 years agobaseparse: more inter-timestamp tracking
Mark Nauwelaerts [Tue, 26 Feb 2013 18:58:49 +0000 (19:58 +0100)]
baseparse: more inter-timestamp tracking

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=691481

11 years agobuffer: Add annotations and pygi friendly extraction function
Olivier Crête [Tue, 26 Mar 2013 23:22:18 +0000 (19:22 -0400)]
buffer: Add annotations and pygi friendly extraction function

API: gst_buffer_extract_dup

11 years agoporting: hopefully clarify a little
Wim Taymans [Wed, 27 Mar 2013 16:08:51 +0000 (17:08 +0100)]
porting: hopefully clarify a little

11 years agoUpdate docs for gst_pad_create_stream_id_printf()
David Schleef [Tue, 26 Mar 2013 01:11:54 +0000 (18:11 -0700)]
Update docs for gst_pad_create_stream_id_printf()

To indicate that format strings should be alpha sortable.

11 years agocaps: Add unit test for GST_CAPS_{ANY,NONE} and GST_STATIC_CAPS_{ANY,NONE}
Sebastian Dröge [Mon, 25 Mar 2013 08:16:05 +0000 (09:16 +0100)]
caps: Add unit test for GST_CAPS_{ANY,NONE} and GST_STATIC_CAPS_{ANY,NONE}

https://bugzilla.gnome.org/show_bug.cgi?id=696435

11 years agocaps: Fix gst_static_caps_get(GST_STATIC_CAPS_NONE)
Sebastian Dröge [Mon, 25 Mar 2013 08:19:24 +0000 (09:19 +0100)]
caps: Fix gst_static_caps_get(GST_STATIC_CAPS_NONE)

https://bugzilla.gnome.org/show_bug.cgi?id=696435

11 years agobasesrc: Don't send error if negotiate fails because we are flushing
Nicolas Dufresne [Fri, 22 Mar 2013 01:00:54 +0000 (21:00 -0400)]
basesrc: Don't send error if negotiate fails because we are flushing

Negotiation may be aborted by a flush from another thread that need to
stop the task (i.e. seek). Check that case and silently pause the task.

https://bugzilla.gnome.org/show_bug.cgi?id=696357

11 years agotoc: some documentation updates
Tim-Philipp Müller [Sun, 24 Mar 2013 17:53:35 +0000 (17:53 +0000)]
toc: some documentation updates

11 years agodocs: fix typo in query function example in Plugin Writer's Guide
Tim-Philipp Müller [Fri, 22 Mar 2013 20:02:57 +0000 (20:02 +0000)]
docs: fix typo in query function example in Plugin Writer's Guide

https://bugzilla.gnome.org/show_bug.cgi?id=696142

11 years agotests: fix spurious failure in test_collect collectpads test
Jonas Holmberg [Thu, 7 Mar 2013 11:11:30 +0000 (12:11 +0100)]
tests: fix spurious failure in test_collect collectpads test

pop() in collected callback.

There were three threads in the test cases that hanged: the test thread and two
threads that push buffers. Each thread push one buffer on one pad. There are
two pads in the collectpads so the second buffer will trigger the
collect-callback.

This is what happens when the hang occurs:

The first thread pushes a buffer and initializes a cookie to the value of a
counter in the collectpads object and waits on a cond for the counter to change
and for someone to consume the buffer (i.e. _pop() it).

The second thread pushes a buffer and calls the collected callback, which
signals the cond that the test thread is waiting for.

The test thread pops both buffers (without holding any lock). Each call to
_pop() increases the counter broadcasts the condition that the first thread is
now waiting for. It then joins both threads (hangs).

The first thread wakes up and returns, since its buffer has been consumed.

The second thread starts executing again. When the callback, called by the
second thread, has returned it initializes a cookie to the value of a counter,
which has already prematurely been increased by the test thread when it popped
the buffers, and wait's on a cond for the counter to change and for someone to
consume the buffer (i.e. _pop() it). Since the buffer has already been poped
and the counter has already been increased it will be stuck forever.

https://bugzilla.gnome.org/show_bug.cgi?id=685555

11 years agogst-uninstalled: remove ffmpeg libs from dynamic linker paths
Tim-Philipp Müller [Sat, 16 Mar 2013 12:05:39 +0000 (12:05 +0000)]
gst-uninstalled: remove ffmpeg libs from dynamic linker paths

We link those libs into the plugin statically, or use external
system libs, but never the internal snapshot dynamically.

11 years agogst-uninstalled: add gst-editing-serves to GI_TYPELIB_PATH
Tim-Philipp Müller [Sat, 16 Mar 2013 12:00:55 +0000 (12:00 +0000)]
gst-uninstalled: add gst-editing-serves to GI_TYPELIB_PATH

So pygi can find it.

https://bugzilla.gnome.org/show_bug.cgi?id=695937

11 years agoutils: make sure slow GST_READ_UINT* variants don't have unexpected side effects
Tim-Philipp Müller [Tue, 12 Mar 2013 13:53:57 +0000 (13:53 +0000)]
utils: make sure slow GST_READ_UINT* variants don't have unexpected side effects

Fixes unit test on systems where unaligned memory access is not possible.

https://bugzilla.gnome.org/show_bug.cgi?id=695599

11 years agolaunch: don't exit the loop on buffering in paused
Wim Taymans [Tue, 5 Mar 2013 10:14:41 +0000 (11:14 +0100)]
launch: don't exit the loop on buffering in paused

When we receive a buffering message of 100% in the paused state, we exit
the event_loop and move to the PLAYING state. What should happen is that
we wait for both ASYNC-DONE and 100% buffering before continueing.

11 years agodesign: update controller design and add some thoughs for future stuff
Stefan Sauer [Fri, 8 Mar 2013 12:15:32 +0000 (13:15 +0100)]
design: update controller design and add some thoughs for future stuff

11 years agodocs: mention clipping of values in control-binding docs
Stefan Sauer [Fri, 8 Mar 2013 07:13:06 +0000 (08:13 +0100)]
docs: mention clipping of values in control-binding docs

11 years agocontroller: code cleanups
Stefan Sauer [Fri, 8 Mar 2013 07:10:20 +0000 (08:10 +0100)]
controller: code cleanups

Use a property for accessing the control-source on the binding. Drop base_init
on the test object.

11 years agoquery: gst_query_get_n_allocation_params() returns a new ref to the allocator
Sebastian Dröge [Thu, 7 Mar 2013 10:46:33 +0000 (11:46 +0100)]
query: gst_query_get_n_allocation_params() returns a new ref to the allocator

11 years agocontroller: remove a bogus test
Stefan Sauer [Thu, 7 Mar 2013 10:35:03 +0000 (11:35 +0100)]
controller: remove a bogus test

We previously forgot to initilize the amplitde property to the default and thus it was 0.0. Therefore a default lfo controlsource returned a series of 0.0 and the test was asserting on that.

11 years agocontrolbinding: relax the pspec for the control-source
Stefan Sauer [Thu, 7 Mar 2013 08:15:14 +0000 (09:15 +0100)]
controlbinding: relax the pspec for the control-source

We can change control sources on controlbindings.

11 years agolfo: set a sensible lower boundary for the frequency
Stefan Sauer [Thu, 7 Mar 2013 08:12:59 +0000 (09:12 +0100)]
lfo: set a sensible lower boundary for the frequency

Use DBL_MIN, which is a the smalles double greater than zero that is not in
denormal format. This exposes the limit better than the runtime check.

11 years agoAutomatic update of common submodule
Tim-Philipp Müller [Wed, 6 Mar 2013 23:59:28 +0000 (23:59 +0000)]
Automatic update of common submodule

From 2de221c to 04c7a1e

11 years agolfocontrolsource: init the amplitude to the default and update the docs
Stefan Sauer [Wed, 6 Mar 2013 15:40:27 +0000 (16:40 +0100)]
lfocontrolsource: init the amplitude to the default and update the docs

11 years agogst-uninstalled: Add all the base/bad libraries and remove obsolete ones
Sebastian Dröge [Tue, 5 Mar 2013 10:30:22 +0000 (11:30 +0100)]
gst-uninstalled: Add all the base/bad libraries and remove obsolete ones

11 years agobasesink: fix use of uninitialized variable
Tim-Philipp Müller [Sun, 3 Mar 2013 16:00:11 +0000 (16:00 +0000)]
basesink: fix use of uninitialized variable

Running suite(s): GstBaseSink
==22023== Conditional jump or move depends on uninitialised value(s)
==22023==    at 0x505FFCE: gst_base_sink_get_sync_times (gstbasesink.c:1936)
==22023==    by 0x5068C80: gst_base_sink_do_sync (gstbasesink.c:2379)
==22023==    by 0x506BCD2: gst_base_sink_default_wait_event (gstbasesink.c:2903)
==22023==    by 0x50633A4: gst_base_sink_default_event (gstbasesink.c:2918)
==22023==    by 0x6F5C216: gst_fake_sink_event (gstfakesink.c:383)
==22023==    by 0x505F164: gst_base_sink_event (gstbasesink.c:3108)
==22023==    by 0x52FA090: gst_pad_send_event_unchecked (gstpad.c:4822)
==22023==    by 0x5303756: gst_pad_send_event (gstpad.c:4984)
==22023==    by 0x40165B: basesink_test_gap (basesink.c:148)

11 years agobuffer: Fix memory copying logic in copy_into()
Sebastian Dröge [Sun, 3 Mar 2013 11:06:36 +0000 (12:06 +0100)]
buffer: Fix memory copying logic in copy_into()

https://bugzilla.gnome.org/show_bug.cgi?id=695035

11 years agoregistrychunks: Use correct print format specifiers to fix compiler warnings
Sebastian Dröge [Sun, 3 Mar 2013 10:28:32 +0000 (11:28 +0100)]
registrychunks: Use correct print format specifiers to fix compiler warnings

11 years agoFix misspellings of 'continuous'
David Schleef [Fri, 22 Feb 2013 22:22:01 +0000 (14:22 -0800)]
Fix misspellings of 'continuous'

11 years agocollectpads: take DTS into account
David Schleef [Fri, 22 Feb 2013 22:56:49 +0000 (14:56 -0800)]
collectpads: take DTS into account

Importantly, this patch converts DTS to running time.  Less importantly,
and possibly a problem for some muxers, is that it orders buffers by
DTS (if it is valid, otherwise PTS).  This is generally correct, but
might be somewhat surprising to muxers.

Also note that once converted to running time, DTS can end up negative.

11 years agomanual: improve the controller docs a little more
Stefan Sauer [Thu, 28 Feb 2013 21:59:43 +0000 (22:59 +0100)]
manual: improve the controller docs a little more

Reword some sections. Explain value mappings better.

11 years agogst-uninstalled: add gst-libav to pkg-config path
Tim-Philipp Müller [Thu, 28 Feb 2013 19:40:32 +0000 (19:40 +0000)]
gst-uninstalled: add gst-libav to pkg-config path

11 years agoseeking: add more logging for seeking
Stefan Sauer [Wed, 27 Feb 2013 21:15:48 +0000 (22:15 +0100)]
seeking: add more logging for seeking

Especially add logging to error code paths.

11 years agobuffer: Gracefully handle gst_memory_copy() returning NULL without crashing
Sebastian Dröge [Wed, 27 Feb 2013 09:09:56 +0000 (10:09 +0100)]
buffer: Gracefully handle gst_memory_copy() returning NULL without crashing

gst_buffer_copy_into() and gst_buffer_resize_range() can now fail.

11 years agoallocator: small internal cleanups
Wim Taymans [Tue, 26 Feb 2013 16:33:30 +0000 (17:33 +0100)]
allocator: small internal cleanups

Rename System memory allocator to GstAllocatorSysmem and the memory to
GstMemorySystem.

11 years agomemory: add method to check memory type
Wim Taymans [Tue, 26 Feb 2013 14:37:24 +0000 (15:37 +0100)]
memory: add method to check memory type

Add a method to check if a memory was allocated from an allocator of
a given type.

API: gst_memory_is_type()

11 years agodocs: improve docs a little
Wim Taymans [Tue, 26 Feb 2013 14:36:10 +0000 (15:36 +0100)]
docs: improve docs a little

11 years agouse GST_MESSAGE_DURATION_CHANGED in docs and code
Wim Taymans [Tue, 26 Feb 2013 14:32:25 +0000 (15:32 +0100)]
use GST_MESSAGE_DURATION_CHANGED in docs and code

11 years agomemory: don't unref allocator too soon
Wim Taymans [Tue, 26 Feb 2013 13:40:38 +0000 (14:40 +0100)]
memory: don't unref allocator too soon

Unref the allocator *after* we have freed the memory. We also need to keep
a ref to the allocator around because following the now freed memory would
lead to crashes.

11 years agobuffer: Fix inverted logic for deciding if memory should be shared or copied
Sebastian Dröge [Tue, 26 Feb 2013 08:08:49 +0000 (09:08 +0100)]
buffer: Fix inverted logic for deciding if memory should be shared or copied

https://bugzilla.gnome.org/show_bug.cgi?id=694717

11 years agoporting: mention segment accumulation in the porting guide
Stefan Sauer [Tue, 26 Feb 2013 06:50:13 +0000 (07:50 +0100)]
porting: mention segment accumulation in the porting guide

This needs more detail, but at least gives people a hint on the issue.

11 years agodocs: s/start/done/ copy'n'paste mistake
Stefan Sauer [Tue, 26 Feb 2013 06:48:35 +0000 (07:48 +0100)]
docs: s/start/done/ copy'n'paste mistake

11 years agoquery: Add new API to remove allocation params and pools from the allocation query
Sebastian Dröge [Mon, 25 Feb 2013 12:57:43 +0000 (13:57 +0100)]
query: Add new API to remove allocation params and pools from the allocation query

11 years agoquery: Document that the first allocator in the allocation query should allow mapping...
Sebastian Dröge [Mon, 25 Feb 2013 12:24:33 +0000 (13:24 +0100)]
query: Document that the first allocator in the allocation query should allow mapping to system memory

11 years agomemory: Keep a reference to the allocator
Sebastian Dröge [Sun, 24 Feb 2013 08:24:26 +0000 (09:24 +0100)]
memory: Keep a reference to the allocator

Otherwise the allocator might get freed while it's still used
by the memory

11 years agobuffer: If sharing a GstMemory fails, fall back to copying it
Sebastian Dröge [Sun, 24 Feb 2013 08:33:03 +0000 (09:33 +0100)]
buffer: If sharing a GstMemory fails, fall back to copying it

11 years agodocs: porting-to-1.0.txt: some element names have changed
Tim-Philipp Müller [Sat, 23 Feb 2013 18:36:15 +0000 (18:36 +0000)]
docs: porting-to-1.0.txt: some element names have changed

11 years agomemory: It's contiguous, not continous
Sebastian Dröge [Sat, 23 Feb 2013 07:19:48 +0000 (08:19 +0100)]
memory: It's contiguous, not continous

11 years agomemory: Add new memory flag to specify that memory can't be mapped
Sebastian Dröge [Fri, 22 Feb 2013 11:41:09 +0000 (12:41 +0100)]
memory: Add new memory flag to specify that memory can't be mapped

11 years agomemory: Add memory flag to mark physically continous memory
Sebastian Dröge [Fri, 22 Feb 2013 08:02:22 +0000 (09:02 +0100)]
memory: Add memory flag to mark physically continous memory

11 years agobasetransform: don't pass NULL outcaps to transform_size on shutdown
Tim-Philipp Müller [Sat, 16 Feb 2013 23:02:21 +0000 (23:02 +0000)]
basetransform: don't pass NULL outcaps to transform_size on shutdown

gst_pad_get_current_caps() on the source pad might yield NULL caps
if we're being shut down and the source pad has already been
deactivated by the other thread that's changing state. Just bail
out in that case, instead of passing NULL caps to the transform_size
function, which it might not expect.

Fixes spurious warnings in audioresample shutdown unit test.

https://bugzilla.gnome.org/show_bug.cgi?id=693996

11 years agogst-uninstalled: Add ORC
Sebastian Dröge [Thu, 21 Feb 2013 09:18:14 +0000 (10:18 +0100)]
gst-uninstalled: Add ORC

11 years agoutils: avoid unexpected side-effects of GST_WRITE_* macros
Tim-Philipp Müller [Tue, 19 Feb 2013 18:00:48 +0000 (18:00 +0000)]
utils: avoid unexpected side-effects of GST_WRITE_* macros

Make sure the data argument is only evaluated once.

11 years agocheck: add some more fail_unless_*() macros for convenience
Tim-Philipp Müller [Tue, 19 Feb 2013 17:36:50 +0000 (17:36 +0000)]
check: add some more fail_unless_*() macros for convenience

API: fail_unless_equals_int_hex
API: assert_equals_int_hex
API: fail_unless_equals_int64_hex
API: assert_equals_int64_hex
API: fail_unless_equals_uint64_hex
API: assert_equals_uint64_hex
API: fail_unless_equals_pointer
API: assert_equals_pointer

11 years agoscripts: add new -base allocators library to gst-uninstalled search paths
Tim-Philipp Müller [Tue, 19 Feb 2013 12:42:32 +0000 (12:42 +0000)]
scripts: add new -base allocators library to gst-uninstalled search paths

11 years agocollectpads: add two more tests using collectpads within an element
Stefan Sauer [Mon, 18 Feb 2013 19:47:04 +0000 (20:47 +0100)]
collectpads: add two more tests using collectpads within an element

Add a static plugin with a rudimentary element using collectpads and do some
pipeline based tests.

11 years agodocs: Fix some ambiguous wording
David Schleef [Sun, 17 Feb 2013 20:50:03 +0000 (12:50 -0800)]
docs: Fix some ambiguous wording

11 years agotriggercontrolsource: add missing end_iter check for sequence
Stefan Sauer [Sun, 17 Feb 2013 18:53:55 +0000 (19:53 +0100)]
triggercontrolsource: add missing end_iter check for sequence

Avoid accessing the end-iter, this is a marker without a data field.

11 years agodocs: link to the appropriate messages from gst_elements_set_state() docs
Stefan Sauer [Sun, 17 Feb 2013 12:20:20 +0000 (13:20 +0100)]
docs: link to the appropriate messages from gst_elements_set_state() docs

For an async state change return one would wait for ASYNC_DONE or STATE_CHANGED.

11 years agobuffer: add since marker for new COPY_DEEP buffer flag
Tim-Philipp Müller [Sat, 16 Feb 2013 14:20:06 +0000 (14:20 +0000)]
buffer: add since marker for new COPY_DEEP buffer flag

11 years agobuffer: add option to deep copy a buffer
Wim Taymans [Sat, 16 Feb 2013 13:59:02 +0000 (14:59 +0100)]
buffer: add option to deep copy a buffer

Add a buffer copy flag to force a memory copy in all cases.

11 years agoRevert "utils: Use gst_pad_get_pad_template() in gst_element_get_compatible_pad_templ...
Tim-Philipp Müller [Thu, 14 Feb 2013 14:09:13 +0000 (14:09 +0000)]
Revert "utils: Use gst_pad_get_pad_template() in gst_element_get_compatible_pad_template()"

This reverts commit 1a1a9e143fb0e155d7627aa8e489cd5d04bc093c.

This breaks the pipelines/tagschecking unit test for some reason
(fakesrc ! capsfilter ! qtmux linking fails now). It might be
a bug in the unit test of course, but someone will need to
investigate this. Reverting for now.

https://bugzilla.gnome.org/show_bug.cgi?id=692508

11 years agoallocator: improve fallback copy function
Wim Taymans [Fri, 15 Feb 2013 12:08:18 +0000 (13:08 +0100)]
allocator: improve fallback copy function

Only use the allocator of the copied memory when we can use the default
_alloc function on it. Otherwise we will have to use the default
allocator for the copy.

11 years agodocs: fix location in user's home directory where GStreamer looks for plugins
Tim-Philipp Müller [Thu, 14 Feb 2013 13:55:10 +0000 (13:55 +0000)]
docs: fix location in user's home directory where GStreamer looks for plugins

It's based on the xdg user data dir now in 1.0.

11 years agoutils: Use gst_pad_get_pad_template() in gst_element_get_compatible_pad_template()
Niv Sardi [Fri, 25 Jan 2013 09:50:27 +0000 (06:50 -0300)]
utils: Use gst_pad_get_pad_template() in gst_element_get_compatible_pad_template()

motivation comes from: /* FIXME: why not gst_pad_get_pad_template (pad); */
this code path is quite nicer, we now only revert to creating the template
if gst_pad_get_pad_template fails.

with this fork, we gain a non-allocation of GstCaps *templcaps

https://bugzilla.gnome.org/show_bug.cgi?id=692508