Tim-Philipp Müller [Fri, 23 Nov 2012 12:47:25 +0000 (12:47 +0000)]
pad: don't use parenthesis for ORed pad probe flag enums
glib-mkenum doesn't like them for some reason.
https://bugzilla.gnome.org/show_bug.cgi?id=688804
Olivier Crête [Fri, 23 Nov 2012 09:58:25 +0000 (10:58 +0100)]
pad: Put all of the probe types in the enum so they work with bindings
https://bugzilla.gnome.org/show_bug.cgi?id=688804
Evan Nemerson [Wed, 21 Nov 2012 07:13:33 +0000 (23:13 -0800)]
libs: Add missing single include headers and use them in GIRs
Wim Taymans [Tue, 20 Nov 2012 15:34:28 +0000 (16:34 +0100)]
tests: don't use deprecated thread API
Wim Taymans [Tue, 20 Nov 2012 15:19:52 +0000 (16:19 +0100)]
basesink: add some debug
Wim Taymans [Tue, 20 Nov 2012 15:19:16 +0000 (16:19 +0100)]
basesink: reset START_TIME when needed
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=685870
Sebastian Dröge [Tue, 20 Nov 2012 14:37:11 +0000 (15:37 +0100)]
value: Make G-I happy by hiding gst_g_thread_get_type()
Christian Fredrik Kalager Schaller [Tue, 20 Nov 2012 14:07:37 +0000 (15:07 +0100)]
Remove xfig from spec file
Sebastian Dröge [Tue, 20 Nov 2012 14:06:14 +0000 (15:06 +0100)]
value: Use the GLib GType for GThread if compiling against GLib 2.35.3 or newer
Sebastian Dröge [Tue, 20 Nov 2012 11:56:12 +0000 (12:56 +0100)]
gst: Don't set the locale in gst_init()
The function init_pre() in gstreamer/gst/gst.c calls setlocale(LC_ALL, ""),
which sets the locale to the values specified in the environment. This is
wrong for two reasons:
1. It is absolutely not the task of a library to decide on the correct locale
for a program. Some programs change the locale for various (good or bad)
reasons, and libraries should respect that. Programs where GStreamer's
overwriting of the locale causes bugs include Emacs [1, 2], Sublime Text [3],
and Lua [4].
[1] http://debbugs.gnu.org/cgi/bugreport.cgi?bug=12392
[2] http://bugzilla.novell.com/show_bug.cgi?id=779426
[3] http://www.sublimetext.com/forum/viewtopic.php?f=3&t=8543
[4] https://github.com/pavouk/lgi/issues/19
Note that setting the locale can cause problems for programs that are not even
linked against GStreamer. In the case of Emacs, for example, GStreamer seems
to be initialized through GTK via libcanberra.
2. Setting the locale is not thread-safe, and therefore should not be done in a
library.
https://bugzilla.gnome.org/show_bug.cgi?id=685650
Arnaud Vrac [Fri, 16 Nov 2012 18:41:48 +0000 (19:41 +0100)]
baseparse: forward stream-start event in push mode
Christian Fredrik Kalager Schaller [Mon, 19 Nov 2012 12:38:30 +0000 (13:38 +0100)]
Remove two last .fig files from build, they are now replaced with .svg files. That said I don't think either .fig file was still being used anywhere. With this
change and the one in common GStreamer no longer depends on xfig.
Christian Fredrik Kalager Schaller [Mon, 19 Nov 2012 12:16:48 +0000 (13:16 +0100)]
Merge branch 'master' of ssh://git.freedesktop.org/git/gstreamer/gstreamer
Tim-Philipp Müller [Mon, 19 Nov 2012 11:23:32 +0000 (11:23 +0000)]
Automatic update of common submodule
From b497c4f to a72faea
Christian Fredrik Kalager Schaller [Mon, 19 Nov 2012 10:45:07 +0000 (11:45 +0100)]
Add SVG versions of .fig file
Tim-Philipp Müller [Sat, 17 Nov 2012 10:27:11 +0000 (10:27 +0000)]
examples: don't compile testrtpool example if pthreads are not available like on win32
Based on patch by: italarab@gmail.com
https://bugzilla.gnome.org/show_bug.cgi?id=688511
Arnaud Vrac [Tue, 13 Nov 2012 20:13:00 +0000 (21:13 +0100)]
inputselector: fix clock leak in wait_running_time
https://bugzilla.gnome.org/show_bug.cgi?id=688477
Tim-Philipp Müller [Sat, 17 Nov 2012 00:13:14 +0000 (00:13 +0000)]
gstcompat.h: move more deprecated API into the deprecated section
https://bugzilla.gnome.org/show_bug.cgi?id=675598
Philippe Normand [Wed, 14 Nov 2012 11:20:54 +0000 (12:20 +0100)]
task: documentation update
GStaticRecMutex usage has been replaced by GRecMutex, reflect this
change in the documentation.
Tim-Philipp Müller [Wed, 14 Nov 2012 10:55:15 +0000 (10:55 +0000)]
testclock: port to new GLib threading API
Tim-Philipp Müller [Tue, 13 Nov 2012 23:11:34 +0000 (23:11 +0000)]
tests: gst_adapter_prev_timestamp -> gst_adapter_prev_pts
https://bugzilla.gnome.org/show_bug.cgi?id=675598
Tim-Philipp Müller [Tue, 13 Nov 2012 22:42:05 +0000 (22:42 +0000)]
testclock: minor cleanups, add since markers for gtk-doc
https://bugzilla.gnome.org/show_bug.cgi?id=683012
Sebastian Rasmussen [Wed, 29 Aug 2012 23:58:41 +0000 (01:58 +0200)]
check: allow GstTestClock to handle clock notifications
API: gst_test_clock_peek_id_count()
API: gst_test_clock_has_id()
API: gst_test_clock_peek_next_pending_id()
API: gst_test_clock_wait_for_next_pending_id()
API: gst_test_clock_wait_for_pending_id_count()
API: gst_test_clock_process_next_clock_id()
API: gst_test_clock_get_next_entry_time()
https://bugzilla.gnome.org/show_bug.cgi?id=683012
Tim-Philipp Müller [Tue, 13 Nov 2012 21:29:01 +0000 (21:29 +0000)]
check: add dependency on gstcheck header files for exports.sym
So exports.sym gets updated correctly, and our new symbols get
exported correctly, which makes g-ir-scanner much happier in
terms of linking.
https://bugzilla.gnome.org/show_bug.cgi?id=683012
Sebastian Rasmussen [Wed, 29 Aug 2012 14:11:10 +0000 (16:11 +0200)]
check: add GstTestClock as a deterministic clock for testing
API: GstTestClock
API: gst_test_clock_new()
API: gst_test_clock_new_with_start_time()
API: gst_test_clock_set_time()
API: gst_test_clock_advance_time()
https://bugzilla.gnome.org/show_bug.cgi?id=683012
Tim-Philipp Müller [Fri, 9 Nov 2012 21:10:42 +0000 (21:10 +0000)]
basesrc: fix debug message
Tim-Philipp Müller [Thu, 8 Nov 2012 20:22:19 +0000 (20:22 +0000)]
structure: re-indent header file
Tabs to spaces.
Sebastian Dröge [Mon, 12 Nov 2012 10:40:59 +0000 (11:40 +0100)]
value: API: Add boxed type for GThread
Sebastian Dröge [Mon, 12 Nov 2012 09:30:08 +0000 (10:30 +0100)]
gst-inspect: Fix indention for printing typefinder features
Sebastian Rasmussen [Mon, 12 Nov 2012 00:40:42 +0000 (01:40 +0100)]
info: fix compiler warning when debugging disabled
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=688130
Alessandro Decina [Sat, 10 Nov 2012 08:50:49 +0000 (09:50 +0100)]
queue: remove unused label. Fixes compiler warning.
Alessandro Decina [Mon, 29 Oct 2012 12:08:31 +0000 (12:08 +0000)]
queue: don't fail in _sink_event for sticky events
Implement the same behaviour as gst_pad_push_event when pushing sticky events
fails, that is don't fail immediately but fail when data flow resumes and upstream
can aggregate properly.
This fixes segment seeks with decodebin and unlinked audio or video branches.
Fixes: https://bugzilla.gnome.org/show_bug.cgi?id=687899
Wim Taymans [Fri, 9 Nov 2012 15:50:50 +0000 (16:50 +0100)]
basesink: add simple rate control
Add a max-bitrate property that will slightly delay rendering of buffers if it
would exceed the maximum defined bitrate. This can be used to do
rate control on network sinks, for example.
API: GstBaseSink::max-bitrate
API: gst_base_sink_set_max_bitrate()
API: gst_base_sink_get_max_bitrate()
Matthew Waters [Thu, 8 Nov 2012 04:33:01 +0000 (15:33 +1100)]
bufferpool: lock before unlock in _get_config
Fixes deadlock on Windows
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=687896
Tim-Philipp Müller [Wed, 7 Nov 2012 18:15:11 +0000 (18:15 +0000)]
configure: update courtesy of autoupdate
Tim-Philipp Müller [Wed, 7 Nov 2012 17:59:49 +0000 (17:59 +0000)]
common: update for AG_GST_PLUGIN_DOCS python checks
Olivier Crête [Tue, 6 Nov 2012 17:29:28 +0000 (18:29 +0100)]
Document GST_DEBUG_FILE
Tim-Philipp Müller [Tue, 6 Nov 2012 17:03:47 +0000 (17:03 +0000)]
gst-inspect: fix alignment of rank, etc.
Tim-Philipp Müller [Tue, 6 Nov 2012 16:58:04 +0000 (16:58 +0000)]
elements: fix leading space in plugin description string
Tim-Philipp Müller [Sat, 3 Nov 2012 20:44:48 +0000 (20:44 +0000)]
Fix FSF address
https://bugzilla.gnome.org/show_bug.cgi?id=687520
Tim-Philipp Müller [Wed, 31 Oct 2012 19:33:30 +0000 (19:33 +0000)]
queue: add "flush-on-eos" property
In flush-on-eos=true mode any data remaining in the queue is
discarded when an EOS event is received, and the EOS passed
downstream as soon as possible (instead of waiting for all
buffers in the queue to get processed by downstream first).
May or may not be useful in capture/encoding scenarios.
Tim-Philipp Müller [Wed, 31 Oct 2012 18:32:38 +0000 (18:32 +0000)]
common: update for python detection
Fixes docs build.
Tim-Philipp Müller [Wed, 31 Oct 2012 17:37:37 +0000 (17:37 +0000)]
configure: let AG_GST_PLUGIN_DOCS check for python
And update common for move from AS_PATH_PYTHON to AM_PATH_PYTHON,
which as a side-effect should pick up newer python versions as well.
https://bugzilla.gnome.org/show_bug.cgi?id=563903
Jan Schmidt [Mon, 29 Oct 2012 23:04:44 +0000 (10:04 +1100)]
collectpads: Clarify docs about the buffer handler callback.
Clarify that the callback owns a ref on a passed buffer.
Jan Schmidt [Mon, 29 Oct 2012 23:04:14 +0000 (10:04 +1100)]
multiqueue: Add EOS status to debug output about filled/unfilled
Jan Schmidt [Sun, 21 Oct 2012 13:31:09 +0000 (00:31 +1100)]
check: Add a simple test for the CollectPads buffer collect callback
Tim-Philipp Müller [Mon, 29 Oct 2012 13:26:10 +0000 (13:26 +0000)]
g-i: fix "can't resolve libraries to shared libraries: gstcheck-1.0" build error
Revert --library=libfoo-1.0.la -> --library=foo-1.0 change made
in previous commit. Turns out that was wrong, despite what the
man page says.
https://bugzilla.gnome.org/show_bug.cgi?id=603710
Tim-Philipp Müller [Mon, 29 Oct 2012 11:30:30 +0000 (11:30 +0000)]
pad: downgrade 'creating random stream-id' debug log message
No need for it to be a warning.
Vincent Penquerc'h [Wed, 13 Jun 2012 12:02:48 +0000 (13:02 +0100)]
baseparse: prevent excessively high memory usage with long streams
Large streams would index one frame every second, which can get quite
large with multi-hour streams, so add an additional byte-based
minimum distance as well, which will kick in for long streams
and make sure we never have more than a couple of thousand index
entries.
https://bugzilla.gnome.org/show_bug.cgi?id=666053
Tim-Philipp Müller [Sun, 28 Oct 2012 17:17:49 +0000 (17:17 +0000)]
libs: g-i: avoid multiple libraries in the shared-library tag
Using multiple libraries causes problems for the C# bindings and
will for similiar languages such as Java when there are bindings
for them.
Also change --library=libgstfoo-X.la to --library=gstfoo-X as
the man page suggests it should be done.
https://bugzilla.gnome.org/show_bug.cgi?id=679315
Tim-Philipp Müller [Sun, 28 Oct 2012 15:53:19 +0000 (15:53 +0000)]
pluginfeature: add gst_plugin_feature_get_plugin_name()
API: gst_plugin_feature_get_plugin_name()
https://bugzilla.gnome.org/show_bug.cgi?id=571832
Tim-Philipp Müller [Sat, 27 Oct 2012 13:40:14 +0000 (14:40 +0100)]
info: allow setting of GST_DEBUG levels by name
e.g. GST_DEBUG=*:INFO,*src:LOG
Thibault Saunier [Fri, 29 Jun 2012 16:38:52 +0000 (12:38 -0400)]
gst: make us of the new gst_debug_set_threshold_from_string function
https://bugzilla.gnome.org/show_bug.cgi?id=679152
Thibault Saunier [Fri, 29 Jun 2012 16:05:36 +0000 (12:05 -0400)]
info: add a function to set debug threshold from a GST_DEBUG-style string
Use the same format as with the GST_DEBUG environment variable.
API: gst_debug_set_threshold_from_string()
https://bugzilla.gnome.org/show_bug.cgi?id=679152
Sebastian Dröge [Thu, 25 Oct 2012 13:27:23 +0000 (15:27 +0200)]
queuearray: Fix unit test
Sebastian Dröge [Mon, 22 Oct 2012 08:13:20 +0000 (10:13 +0200)]
dataqueue/queuearray: Make public API again
These are actually used outside of coreelements nowadays.
Also hide lots of internals and add padding and documentation.
Tim-Philipp Müller [Thu, 25 Oct 2012 11:10:27 +0000 (12:10 +0100)]
Back to feature development
Tim-Philipp Müller [Wed, 24 Oct 2012 23:04:49 +0000 (00:04 +0100)]
Release 1.0.2
Руслан Ижбулатов [Wed, 24 Oct 2012 12:13:34 +0000 (16:13 +0400)]
examples: link testrtpool to pthreads
Fixes #686787
Sebastian Dröge [Wed, 24 Oct 2012 09:46:01 +0000 (11:46 +0200)]
event: Allow GST_CLOCK_TIME_NONE as duration for GAP events
Wim Taymans [Wed, 24 Oct 2012 09:16:12 +0000 (11:16 +0200)]
basesrc: use new GCond for async state change
Use a new GCond, protected with the object lock, to signal completion
of the async state change. We can't reuse the live lock because that
one can be locked when the create function blocks.
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=686723
Miguel Angel Cabrera Moya [Mon, 22 Oct 2012 18:25:43 +0000 (20:25 +0200)]
allocator: fix memory leak in _fallback_mem_copy
https://bugzilla.gnome.org/show_bug.cgi?id=686658
Miguel Angel Cabrera Moya [Mon, 22 Oct 2012 18:33:06 +0000 (20:33 +0200)]
preset: remove variable not read
https://bugzilla.gnome.org/show_bug.cgi?id=686659
Sebastian Dröge [Mon, 22 Oct 2012 13:04:26 +0000 (15:04 +0200)]
configure: Properly check for pthread
The old check failed on Android for example.
Sebastian Dröge [Mon, 22 Oct 2012 08:25:40 +0000 (10:25 +0200)]
info: Don't use GST_DEBUG() in gst_debug_add_log_function() and related functions unconditionally
If GStreamer was not initialized yet this will cause g_warnings().
Tim-Philipp Müller [Sat, 20 Oct 2012 18:44:43 +0000 (19:44 +0100)]
collectpads: fix g-i annotation for GstCollectPadsBufferFunction
We pass ownership of the buffer to the function.
Tim-Philipp Müller [Sat, 20 Oct 2012 11:54:06 +0000 (12:54 +0100)]
g_type_init() is no longer required and deprecated in glib >= 2.35.0
https://bugzilla.gnome.org/show_bug.cgi?id=686456
Michael Smith [Fri, 19 Oct 2012 20:36:33 +0000 (13:36 -0700)]
GstSample: fix typo in G-I annotations, allows creating GstSamples from bindings.
Sebastian Dröge [Thu, 18 Oct 2012 13:31:17 +0000 (15:31 +0200)]
poll: Fix compiler warning about constness
passing argument 1 of 'g_mutex_lock' discards 'const' qualifier from pointer target type
passing argument 1 of 'g_mutex_unlock' discards 'const' qualifier from pointer target type
Tim-Philipp Müller [Wed, 17 Oct 2012 15:49:19 +0000 (16:49 +0100)]
bin, pipeline: use gst_element_class_set_static_metadata()
So the strings aren't copied.
Sebastian Dröge [Tue, 16 Oct 2012 10:31:50 +0000 (12:31 +0200)]
element: API: Add GstElement::post_message() vfunc
Conflicts:
gst/gstelement.h
Wim Taymans [Tue, 16 Oct 2012 09:54:54 +0000 (11:54 +0200)]
pwg: link to caps and qos chapters
Wim Taymans [Tue, 16 Oct 2012 09:20:59 +0000 (11:20 +0200)]
pwg: add section about query function
Wim Taymans [Tue, 16 Oct 2012 09:12:38 +0000 (11:12 +0200)]
pwg: fix event function
Tim-Philipp Müller [Mon, 15 Oct 2012 18:56:36 +0000 (19:56 +0100)]
collectpads: minor docs fixes
Tim-Philipp Müller [Mon, 15 Oct 2012 18:55:42 +0000 (19:55 +0100)]
collectpads: fix buffer leak in clip_time
Mark Nauwelaerts [Mon, 15 Oct 2012 16:44:52 +0000 (18:44 +0200)]
collectpads: call clip function with user data
Wim Taymans [Mon, 15 Oct 2012 12:06:45 +0000 (14:06 +0200)]
pwg: reorder some chapters
Reorder some chapter so that they match the steps done in the
element.
Wim Taymans [Mon, 15 Oct 2012 11:59:50 +0000 (13:59 +0200)]
pwg: small tweaks to negotiation
Wim Taymans [Mon, 15 Oct 2012 11:44:51 +0000 (13:44 +0200)]
pwg: improve negotiation documentation some more
Wim Taymans [Mon, 15 Oct 2012 10:10:46 +0000 (12:10 +0200)]
pwg: update negotiation part
Wim Taymans [Mon, 15 Oct 2012 10:10:11 +0000 (12:10 +0200)]
docs: update synchronization docs
Wim Taymans [Fri, 12 Oct 2012 14:58:03 +0000 (16:58 +0200)]
pwg: work on rewriting caps negotiation docs
Wim Taymans [Fri, 12 Oct 2012 14:09:26 +0000 (16:09 +0200)]
design: rename passthrough negotiation
Rename passthrough negotiation to transform negotiation to avoid
confusion with passthrough operation.
Wim Taymans [Fri, 12 Oct 2012 11:15:48 +0000 (13:15 +0200)]
manual: no more new-decoded-pad
Wim Taymans [Fri, 12 Oct 2012 11:13:32 +0000 (13:13 +0200)]
manual: move embedding elements to separate chapter
Wim Taymans [Fri, 12 Oct 2012 11:01:28 +0000 (13:01 +0200)]
pwg: small example for throttle
Wim Taymans [Fri, 12 Oct 2012 10:55:57 +0000 (12:55 +0200)]
pwg: add info about QoS
Wim Taymans [Fri, 12 Oct 2012 10:55:25 +0000 (12:55 +0200)]
pwg: adds some more links
Wim Taymans [Fri, 12 Oct 2012 10:55:03 +0000 (12:55 +0200)]
qos: messages are posted, not dropped
Wim Taymans [Fri, 12 Oct 2012 08:35:43 +0000 (10:35 +0200)]
manual: update graphics
Wim Taymans [Thu, 11 Oct 2012 15:10:17 +0000 (17:10 +0200)]
manual: add example of no-rebuffer buffering strategy
Wim Taymans [Thu, 11 Oct 2012 15:10:04 +0000 (17:10 +0200)]
manual: small tweaks
Wim Taymans [Thu, 11 Oct 2012 15:09:29 +0000 (17:09 +0200)]
query: buffering time left is in milliseconds
Wim Taymans [Thu, 11 Oct 2012 15:07:39 +0000 (17:07 +0200)]
manual: add some text about bin state change order
Wim Taymans [Wed, 10 Oct 2012 14:43:58 +0000 (16:43 +0200)]
manual: talk about playsink
Talk about playsink and give an example of its usage.
Wim Taymans [Wed, 10 Oct 2012 11:08:31 +0000 (13:08 +0200)]
manual: add something about uridecodebin
Mark Nauwelaerts [Wed, 10 Oct 2012 09:35:01 +0000 (11:35 +0200)]
collectpads: ensure all timestamps are in same time domain
... by not only processing incoming buffers through a clip function,
but also other timestamps such as those coming from GAP event.
Mark Nauwelaerts [Wed, 10 Oct 2012 08:36:32 +0000 (10:36 +0200)]
docs: adjust some parameter mismatches
Sebastian Dröge [Wed, 10 Oct 2012 09:34:27 +0000 (11:34 +0200)]
pad: Downgrade GST_WARNING to GST_INFO
It's usually not a problem if a query fails if there's no peer,
especially as it will happen during pad linking (caps query)
quite often and spams the logs.