platform/upstream/gstreamer.git
12 years agodocs: fix some docs
Wim Taymans [Thu, 13 Sep 2012 09:35:41 +0000 (11:35 +0200)]
docs: fix some docs

from git grep for ffmpegcolorspace and x-raw-

12 years agoparse: add missing declaration
Wim Taymans [Thu, 13 Sep 2012 08:48:08 +0000 (10:48 +0200)]
parse: add missing declaration

12 years agobasesrc: indent fix
Wim Taymans [Thu, 13 Sep 2012 08:24:16 +0000 (10:24 +0200)]
basesrc: indent fix

12 years agobaseparse: Add a mode/flag for disabling PTS interpolation
Jan Schmidt [Thu, 13 Sep 2012 05:44:37 +0000 (22:44 -0700)]
baseparse: Add a mode/flag for disabling PTS interpolation

To be used by sub-classes implementing video formats with reordering
such as MPEG.

12 years agobaseparse: Handle GAP and still-frame events.
Jan Schmidt [Tue, 11 Sep 2012 01:38:57 +0000 (18:38 -0700)]
baseparse: Handle GAP and still-frame events.

Hacky, because the still-frame code all lives in -base, where we
can't use it - so this is a hacky duplication of -base code. Not
sure which way to fix this: Move baseparse to -base, or move still-frame
events to core?

12 years agobaseparse: Restructure event handling
Jan Schmidt [Wed, 5 Sep 2012 02:38:26 +0000 (19:38 -0700)]
baseparse: Restructure event handling

Make the event handling more like what videodecoder does,
to ensure that all events are passed to child classes before being
placed on the pending queue or pushed onward.

12 years agobaseparse: Store incoming cached events in reverse order
Jan Schmidt [Mon, 3 Sep 2012 17:30:08 +0000 (10:30 -0700)]
baseparse: Store incoming cached events in reverse order

Reverse the list just before sending. Prepending is more efficient
than appending, so this saves some cycles.

12 years agobaseparse: First attempt at handling both DTS and PTS
Jan Schmidt [Mon, 3 Sep 2012 06:32:50 +0000 (23:32 -0700)]
baseparse: First attempt at handling both DTS and PTS

12 years agotaglist: add warning when we get something else than a sample for a sample tag
Tim-Philipp Müller [Wed, 12 Sep 2012 23:38:21 +0000 (00:38 +0100)]
taglist: add warning when we get something else than a sample for a sample tag

Facilitate GstBuffer -> GstSample transition for some tags,
could be hard to catch otherwise when creating tags, since
it'll only be apparent later when someone tries to read the
tags.

12 years agoelementfactory: don't crash if no element klass has been set
Andreas Frisch [Wed, 12 Sep 2012 12:14:31 +0000 (14:14 +0200)]
elementfactory: don't crash if no element klass has been set

https://bugzilla.gnome.org/show_bug.cgi?id=683865

12 years agocollectpads: fix a misplaced ')'
Stefan Sauer [Wed, 12 Sep 2012 21:12:14 +0000 (23:12 +0200)]
collectpads: fix a misplaced ')'

12 years agoerror: don't tell people to file a bug for negotiation errors
Tim-Philipp Müller [Wed, 12 Sep 2012 20:20:46 +0000 (21:20 +0100)]
error: don't tell people to file a bug for negotiation errors

12 years agocollectpads: remove gst_collect_pads_add_pad_full
Stefan Sauer [Wed, 12 Sep 2012 18:54:50 +0000 (20:54 +0200)]
collectpads: remove gst_collect_pads_add_pad_full

Rename gst_collect_pads_add_pad_full() to gst_collect_pads_add_pad() and fix all
invocations.

12 years agofilesink: fix build on Cygwin
Mark Nauwelaerts [Wed, 12 Sep 2012 15:16:41 +0000 (17:16 +0200)]
filesink: fix build on Cygwin

... where __fbufsize is not available

12 years agoRevert "tests: fix buffer leak in queue2 unit test"
Tim-Philipp Müller [Wed, 12 Sep 2012 12:00:15 +0000 (13:00 +0100)]
Revert "tests: fix buffer leak in queue2 unit test"

This reverts commit 232fd2953eb00f694b667e7796704f5974cea452.

This was already fixed.

12 years agoqueue2: fix possible data corruption in ring buffer mode when seeking
Tim-Philipp Müller [Thu, 24 May 2012 12:08:16 +0000 (13:08 +0100)]
queue2: fix possible data corruption in ring buffer mode when seeking

Fix race that could cause data corruption when seeking in ring buffer
mode.

In perform_seek_to_offset(), called from the demuxer's pull_range
request, we drop the lock, tell upstream (usually a http source)
to seek to a different offset, then re-acquire the lock before we
do things to the ranges. However, between us sending the seek event
and re-acquiring the lock, the source thread might already have pushed
some data and moved along the range's writing_pos beyond the seek
offset. In that case we don't want to set the writing position back
to the requested seek position, as it would cause data to be written
to the wrong offset in the file or ring buffer.

Reproducible doing seek-emulated fast-forward/backward on 006653.

Conflicts:
plugins/elements/gstqueue2.c

12 years agotests: fix buffer leak in queue2 unit test
Tim-Philipp Müller [Thu, 24 May 2012 12:06:47 +0000 (13:06 +0100)]
tests: fix buffer leak in queue2 unit test

12 years agocheck: remove glib deprecation compatibility trickery
Mark Nauwelaerts [Wed, 12 Sep 2012 10:23:22 +0000 (12:23 +0200)]
check: remove glib deprecation compatibility trickery

12 years agocheck: port to the new GLib thread API
Mark Nauwelaerts [Wed, 12 Sep 2012 10:22:53 +0000 (12:22 +0200)]
check: port to the new GLib thread API

12 years agotests: port to new GLib thread API
Mark Nauwelaerts [Wed, 12 Sep 2012 09:52:25 +0000 (11:52 +0200)]
tests: port to new GLib thread API

12 years agotests: benchmarks: align error message with code
Mark Nauwelaerts [Wed, 12 Sep 2012 09:49:55 +0000 (11:49 +0200)]
tests: benchmarks: align error message with code

12 years agopad: expose gst_pad_mode_get_name() and use it in baseparse
Tim-Philipp Müller [Tue, 11 Sep 2012 18:49:58 +0000 (19:49 +0100)]
pad: expose gst_pad_mode_get_name() and use it in baseparse

12 years agoscripts: update for gst-ffmpeg -> gst-libav
Tim-Philipp Müller [Tue, 11 Sep 2012 12:22:57 +0000 (13:22 +0100)]
scripts: update for gst-ffmpeg -> gst-libav

Now that we have a gst-libav git repository (symlinked to gst-ffmpeg).

12 years agoquery: adjust test logic for scheduling mode with flagS
Mark Nauwelaerts [Tue, 11 Sep 2012 15:27:03 +0000 (17:27 +0200)]
query: adjust test logic for scheduling mode with flagS

12 years agoquery: add convenience API to query for scheduling mode and flags
Mark Nauwelaerts [Tue, 11 Sep 2012 14:39:34 +0000 (16:39 +0200)]
query: add convenience API to query for scheduling mode and flags

12 years agoevents: remove STREAM_CONFIG
Wim Taymans [Tue, 11 Sep 2012 14:29:12 +0000 (16:29 +0200)]
events: remove STREAM_CONFIG

We won't be able to implement this so it's better to move it out of the way.

12 years agocollectpads: clean up header indentation
Wim Taymans [Tue, 11 Sep 2012 14:09:37 +0000 (16:09 +0200)]
collectpads: clean up header indentation

12 years agoutils: allow NULL stream_id also when 0 srcpads
Wim Taymans [Tue, 11 Sep 2012 09:34:11 +0000 (11:34 +0200)]
utils: allow NULL stream_id also when 0 srcpads

We usually first create the stream_id for the stream_start event and then add
the pad to the element. This means that this functions should work when there
are no pads on the element yet.

12 years agobaseparse, typefind: only activate in pull mode if upstream is seekable
Tim-Philipp Müller [Mon, 10 Sep 2012 20:39:32 +0000 (21:39 +0100)]
baseparse, typefind: only activate in pull mode if upstream is seekable

Upstream might support pull mode, but only sequential pulls,
which isn't gonna do much for us.

https://bugzilla.gnome.org/show_bug.cgi?id=634927

12 years agoporting-to-0.11.txt: some minor fixes
Tim-Philipp Müller [Mon, 10 Sep 2012 19:30:32 +0000 (20:30 +0100)]
porting-to-0.11.txt: some minor fixes

12 years agosample: free info structure with sample if there is one and fix copy with NULL info...
Tim-Philipp Müller [Mon, 10 Sep 2012 15:52:46 +0000 (16:52 +0100)]
sample: free info structure with sample if there is one and fix copy with NULL info structure

12 years agomemory: add padding to GstMapInfo
Mark Nauwelaerts [Mon, 10 Sep 2012 10:20:15 +0000 (12:20 +0200)]
memory: add padding to GstMapInfo

12 years agolibs: adjust comment style
Mark Nauwelaerts [Mon, 10 Sep 2012 10:12:02 +0000 (12:12 +0200)]
libs: adjust comment style

12 years agogst: remove some defunct commented code
Mark Nauwelaerts [Mon, 10 Sep 2012 10:11:23 +0000 (12:11 +0200)]
gst: remove some defunct commented code

12 years agodocs: improve porting doc
Wim Taymans [Mon, 10 Sep 2012 10:00:22 +0000 (12:00 +0200)]
docs: improve porting doc

12 years agotests: disable deprecation warnings
Wim Taymans [Mon, 10 Sep 2012 08:08:09 +0000 (10:08 +0200)]
tests: disable deprecation warnings

define GLIB_DISABLE_DEPRECATION_WARNINGS earlier so that it is defined before
the glib headers are loaded or else we trip over the GValueArray deprecations in
gst-inspect.c.

12 years agocontroller: fix direct control binding double -> int conversion
Tim-Philipp Müller [Fri, 7 Sep 2012 00:02:10 +0000 (01:02 +0100)]
controller: fix direct control binding double -> int conversion

Round properly to nearest integer. Fixes controller
unit test on PowerPC G4.

12 years agoexamples: fix bus/fd leak in hello world example
Tim-Philipp Müller [Thu, 6 Sep 2012 14:06:08 +0000 (15:06 +0100)]
examples: fix bus/fd leak in hello world example

https://bugzilla.gnome.org/show_bug.cgi?id=683470

12 years agogst-element-check.m4: fix action-if-found and not-found invocation
Tim-Philipp Müller [Wed, 5 Sep 2012 18:55:10 +0000 (19:55 +0100)]
gst-element-check.m4: fix action-if-found and not-found invocation

Arguments got shifted back by one.

12 years agocollectpads: handle GAP event
Mark Nauwelaerts [Wed, 5 Sep 2012 13:37:13 +0000 (15:37 +0200)]
collectpads: handle GAP event

12 years agobasesink: wait_eos -> wait_event
Wim Taymans [Tue, 4 Sep 2012 10:13:11 +0000 (12:13 +0200)]
basesink: wait_eos -> wait_event

Fix a FIXME. Now we can also pass the GAP event to the subclass.

12 years agoexamples: update Makefile.am android bits in controller example
Tim-Philipp Müller [Mon, 3 Sep 2012 17:45:03 +0000 (18:45 +0100)]
examples: update Makefile.am android bits in controller example

Should fix build failure reported on IRC.

12 years agopad: check sticky events also after pad block
Wim Taymans [Thu, 30 Aug 2012 17:15:42 +0000 (19:15 +0200)]
pad: check sticky events also after pad block

Recheck for sticky events after doing a pad block because the pad block could
have caused a relink and then we need to resend the events to the newly linked
pad.

Fixes things like switching of visualisations.

12 years agobaseparse: update for gst_message_new_duration -> _duration_changed()
Tim-Philipp Müller [Sun, 2 Sep 2012 01:04:14 +0000 (02:04 +0100)]
baseparse: update for gst_message_new_duration -> _duration_changed()

12 years agomessage: rename GST_MESSAGE_DURATION -> GST_MESSAGE_DURATION_CHANGED
Tim-Philipp Müller [Sun, 2 Sep 2012 00:17:44 +0000 (01:17 +0100)]
message: rename GST_MESSAGE_DURATION -> GST_MESSAGE_DURATION_CHANGED

The duration should be re-queried via a query using the
normal path, we don't want applications to use the value
from the message itself, since it might no match what a
duration query done from the sink upstream might yield.

Also disables duration caching in GstBin. It should be
added back again at some point.

12 years agoconfigure: add reminder to remove GST_UNSTABLE_API stuff before 1.0.0
Tim-Philipp Müller [Sat, 1 Sep 2012 22:54:23 +0000 (23:54 +0100)]
configure: add reminder to remove GST_UNSTABLE_API stuff before 1.0.0

12 years agogst-element-check.m4: rename AM_GST_ELEMENT_CHECK to GST_ELEMENT_CHECK
Tim-Philipp Müller [Sat, 1 Sep 2012 17:06:58 +0000 (18:06 +0100)]
gst-element-check.m4: rename AM_GST_ELEMENT_CHECK to GST_ELEMENT_CHECK

And allow passing of a minimum version (if not needed, pass 1.0).

https://bugzilla.gnome.org/show_bug.cgi?id=682968

12 years agotests: add check for gst-inspect --exists functionality
Tim-Philipp Müller [Sat, 1 Sep 2012 16:50:14 +0000 (17:50 +0100)]
tests: add check for gst-inspect --exists functionality

12 years agotools: add --exists and --atleast-version option to gst-inspect
Tim-Philipp Müller [Sat, 1 Sep 2012 16:47:58 +0000 (17:47 +0100)]
tools: add --exists and --atleast-version option to gst-inspect

For checking if an element exists with a given minimum version.
Will use that in our new GST_ELEMENT_CHECK m4 macro.

https://bugzilla.gnome.org/show_bug.cgi?id=682968

12 years agopluginfeature: disable version mangling for post-1.0.0 release
Tim-Philipp Müller [Sat, 1 Sep 2012 16:32:04 +0000 (17:32 +0100)]
pluginfeature: disable version mangling for post-1.0.0 release

Just in case we don't grep for FIXME 1.0 before the release.

12 years agobasesink: Make GAP events actually trigger preroll
Jan Schmidt [Fri, 31 Aug 2012 18:31:45 +0000 (11:31 -0700)]
basesink: Make GAP events actually trigger preroll

Slightly hacky approach needing refinement

12 years agogstpad: make some debug statements more verbose
Jan Schmidt [Fri, 31 Aug 2012 13:25:22 +0000 (06:25 -0700)]
gstpad: make some debug statements more verbose

12 years agoghostpad: Make some debugging more verbose
Jan Schmidt [Fri, 31 Aug 2012 13:23:53 +0000 (06:23 -0700)]
ghostpad: Make some debugging more verbose

Also, remove an unnecessary #include in input-selector

12 years agoGstSegment: Fix doc description string last_stop->position
Jan Schmidt [Tue, 28 Aug 2012 22:44:48 +0000 (15:44 -0700)]
GstSegment: Fix doc description string last_stop->position

12 years agoinputselector: fix clock leak
Arnaud Vrac [Thu, 30 Aug 2012 18:47:57 +0000 (19:47 +0100)]
inputselector: fix clock leak

https://bugzilla.gnome.org/show_bug.cgi?id=682997

12 years agotools: output gst-inspect errors to stderr
Tim-Philipp Müller [Wed, 29 Aug 2012 21:57:02 +0000 (22:57 +0100)]
tools: output gst-inspect errors to stderr

12 years agovalue: fix crash serialising a 0 flags value when there's no name for it
Alban Browaeys [Tue, 28 Aug 2012 05:39:50 +0000 (07:39 +0200)]
value: fix crash serialising a 0 flags value when there's no name for it

Fixes segfault when doing gst-launch-1.0 -v -m camerabin
(encodebin notifies a 0 value for its "flag" property).

https://bugzilla.gnome.org/show_bug.cgi?id=682958

12 years agogst: log performance warning debug message if glib emulates atomic ops
Tim-Philipp Müller [Fri, 24 Aug 2012 22:14:57 +0000 (23:14 +0100)]
gst: log performance warning debug message if glib emulates atomic ops

12 years agogst: use configure-detected or externally provided glib-mkenums
Lionel Landwerlin [Thu, 23 Aug 2012 12:51:27 +0000 (13:51 +0100)]
gst: use configure-detected or externally provided glib-mkenums

To ease cross-compilation.

https://bugzilla.gnome.org/show_bug.cgi?id=677620

12 years agoAutomatic update of common submodule
Stefan Sauer [Wed, 22 Aug 2012 11:29:34 +0000 (13:29 +0200)]
Automatic update of common submodule

From 668acee to 4f962f7

12 years agoconfigure: bump gtk-doc req to 1.12 (mar-2009)
Stefan Sauer [Wed, 22 Aug 2012 11:14:56 +0000 (13:14 +0200)]
configure: bump gtk-doc req to 1.12 (mar-2009)

This allows us to e.g. unconditionally use gtkdoc-rebase.

12 years agomemory: add _make_writable
Wim Taymans [Tue, 21 Aug 2012 11:30:52 +0000 (13:30 +0200)]
memory: add _make_writable

12 years agodocs: mention some media type changes in porting-to-0.11.txt doc
Tim-Philipp Müller [Mon, 20 Aug 2012 23:03:37 +0000 (00:03 +0100)]
docs: mention some media type changes in porting-to-0.11.txt doc

12 years agodocs: minor update to porting doc for child proxy lookup method
Tim-Philipp Müller [Mon, 20 Aug 2012 12:51:05 +0000 (13:51 +0100)]
docs: minor update to porting doc for child proxy lookup method

And a typo fix.

12 years agoallocator: make a copy with the same alignment
Wim Taymans [Mon, 20 Aug 2012 09:31:51 +0000 (11:31 +0200)]
allocator: make a copy with the same alignment

When making a copy of the memory allocated from the default memory allocator,
make sure the new copy has the same alignment as the original memory.

See https://bugzilla.gnome.org/show_bug.cgi?id=680796

12 years agobaseparse: make seeking in DEFAULT format work if the subclass can convert for us
Tim-Philipp Müller [Sun, 19 Aug 2012 16:51:00 +0000 (17:51 +0100)]
baseparse: make seeking in DEFAULT format work if the subclass can convert for us

We only deal in TIME format ourselves, but if the subclass can handle
converting other formats into TIME format, we can support that too.

Fixes seeking in DEFAULT (sample) format with flacparse,
and the flacdec unit test.

12 years agotools: minor fixes to gst-launch man page
Tim-Philipp Müller [Sat, 18 Aug 2012 20:42:23 +0000 (21:42 +0100)]
tools: minor fixes to gst-launch man page

12 years agopreset: implement child_proxy support
Stefan Sauer [Fri, 17 Aug 2012 10:23:50 +0000 (12:23 +0200)]
preset: implement child_proxy support

Elements such as the GstIirEqualizerNBands would so far not store the properties
of their children. Now we also grab the properties of child elements and try to
restore them.

12 years agoinputselector: Wait for other streams to advance on unselected pads
Edward Hervey [Tue, 14 Aug 2012 16:44:38 +0000 (18:44 +0200)]
inputselector: Wait for other streams to advance on unselected pads

Otherwise we end up dropping a lot of data in the case where data starts
arriving on the non-selected pad, resulting in big gaps in stream switching

12 years agoinputselector: More debug statements
Edward Hervey [Tue, 14 Aug 2012 16:43:54 +0000 (18:43 +0200)]
inputselector: More debug statements

12 years agoinputselector: Don't forward stream-start sticky events
Edward Hervey [Tue, 14 Aug 2012 16:42:31 +0000 (18:42 +0200)]
inputselector: Don't forward stream-start sticky events

Only one STREAM_START event should be let through, else it will
confuse downstream elements that think a new stream is starting
whereas in fact we are just switching to a different input.

In the future we might want to let them through but with the same
sequence number.

12 years agodocs: Add new basesrc/basetransform API to the docs
Sebastian Dröge [Tue, 14 Aug 2012 13:46:35 +0000 (15:46 +0200)]
docs: Add new basesrc/basetransform API to the docs

12 years agobasetransform: getters for pool and allocator
Víctor Manuel Jáquez Leal [Tue, 7 Aug 2012 15:38:53 +0000 (17:38 +0200)]
basetransform: getters for pool and allocator

Sometimes a transform filter would need the buffer pool or the memory
allocator negotiated by the base class, for example, for querying different
parameters, such as a bigger number of buffers to allocate by the buffer pool.

This patch expose a two getters accessors: one for the buffer pool and the
other for the memory allocator.

12 years agobasesrc: getters for pool and allocator
Víctor Manuel Jáquez Leal [Tue, 7 Aug 2012 15:35:48 +0000 (17:35 +0200)]
basesrc: getters for pool and allocator

Sometimes the sources would use the buffer pool or the memory allocator for
something else than just allocating output buffers; for example, querying for
different parameters, such as a bigger number of buffers to allocate by the
pool.

This patch expose a two getters accessors: one for the buffer pool and the
other for the memory allocator.

12 years agoregistry: remove some unused and in their current form pointless API
Tim-Philipp Müller [Mon, 13 Aug 2012 23:39:18 +0000 (00:39 +0100)]
registry: remove some unused and in their current form pointless API

Not so useful: just adds/reads stuff from an internal GList without
actually doing anything with those paths, so remove for now:

 gst_registry_add_path
 gst_registry_get_path_list

https://bugzilla.gnome.org/show_bug.cgi?id=608841

12 years agoparse: fix up for gst_child_proxy_lookup() only working on child proxy interfaces
Tim-Philipp Müller [Sun, 12 Aug 2012 12:27:06 +0000 (13:27 +0100)]
parse: fix up for gst_child_proxy_lookup() only working on child proxy interfaces

https://bugzilla.gnome.org/show_bug.cgi?id=681681

12 years agochildproxy: make gst_child_proxy_lookup() a proper GstChildProxy method
Tim-Philipp Müller [Sun, 12 Aug 2012 12:24:18 +0000 (13:24 +0100)]
childproxy: make gst_child_proxy_lookup() a proper GstChildProxy method

No longer accept any old GObjects. This makes things nicer for
bindings. If a utility function that handles both nicely
is deemed worthwhile, we can still add one to gstutils.

https://bugzilla.gnome.org/show_bug.cgi?id=681681

12 years agovalue: when serialising arrays or lists, handle types we can't serialise more gracefully
Tim-Philipp Müller [Sun, 12 Aug 2012 23:01:16 +0000 (00:01 +0100)]
value: when serialising arrays or lists, handle types we can't serialise more gracefully

https://bugzilla.gnome.org/show_bug.cgi?id=681322

12 years agoconsistencychecker: add some more details to failure messages
Tim-Philipp Müller [Sun, 12 Aug 2012 18:39:46 +0000 (19:39 +0100)]
consistencychecker: add some more details to failure messages

Mention pad where the problem occured, and the event name.

12 years agotests: fix collectpads test
Tim-Philipp Müller [Sun, 12 Aug 2012 17:36:09 +0000 (18:36 +0100)]
tests: fix collectpads test

After an EOS we must send a FLUSH_STOP event if
we want to send data again.

12 years agoevent: fix leak in gst_event_parse_stream_start()
Tim-Philipp Müller [Sun, 12 Aug 2012 17:31:13 +0000 (18:31 +0100)]
event: fix leak in gst_event_parse_stream_start()

gst_structure_id_get() will make a copy of the string
extracted, but we're assigning it to a const gchar *.

12 years agotests: make pipeline test valgrind clean
Tim-Philipp Müller [Sun, 12 Aug 2012 15:40:03 +0000 (16:40 +0100)]
tests: make pipeline test valgrind clean

12 years agotests: fix pipeline unit test
Tim-Philipp Müller [Sun, 12 Aug 2012 15:37:02 +0000 (16:37 +0100)]
tests: fix pipeline unit test

Which was disabled because it failed.

12 years agoscripts: fix unterminated quoted string in create-uninstalled-setup.sh
Tim-Philipp Müller [Sun, 12 Aug 2012 14:48:20 +0000 (15:48 +0100)]
scripts: fix unterminated quoted string in create-uninstalled-setup.sh

12 years agodocs: mention gst_video_format_parse_caps() in porting guide
Tim-Philipp Müller [Sat, 11 Aug 2012 23:12:56 +0000 (00:12 +0100)]
docs: mention gst_video_format_parse_caps() in porting guide

12 years agodocs: fix up docs a bit
Tim-Philipp Müller [Sat, 11 Aug 2012 21:19:32 +0000 (22:19 +0100)]
docs: fix up docs a bit

12 years agochildproxy: fix up g-i annotation for _lookup() paramspec return value
Tim-Philipp Müller [Sat, 11 Aug 2012 21:18:13 +0000 (22:18 +0100)]
childproxy: fix up g-i annotation for _lookup() paramspec return value

No ref is returned here.

12 years agowin32: update .def file for new buffer functions
Tim-Philipp Müller [Sat, 11 Aug 2012 21:17:35 +0000 (22:17 +0100)]
win32: update .def file for new buffer functions

12 years agobaseparse: fix reverse playback with upstream demuxers that support it
Tim-Philipp Müller [Fri, 10 Aug 2012 21:58:56 +0000 (22:58 +0100)]
baseparse: fix reverse playback with upstream demuxers that support it

Don't just return FALSE for seek events with negative rates when
operating in push mode. An upstream demuxer may support this just
fine, so if we're not operating in pull mode always check upstream
first if it can handle the seek event. This fixes reverse playback
where the upstream demuxer supports it (e.g. with qtdemux). The
same code would work fine in 0.10, because baseparse will just
call the default pad event handler if FALSE was returned from the
baseparse event handler, and the pad event handler will just
forward it upstream. In 0.11 the baseclass or subclass is
responsible for chaining up to the parent class or forwarding the
event upstream in any case.

Disable reverse playback in pull mode for now, there seems to
be something going wrong with the segment configuration in that
case.

12 years agobasetransform: do not error on not-negotiated
Thiago Santos [Sat, 4 Aug 2012 14:48:52 +0000 (11:48 -0300)]
basetransform: do not error on not-negotiated

Don't error out too early and let upstream decide if it can
workaround a not-negotiated problem

https://bugzilla.gnome.org/show_bug.cgi?id=681198

12 years agobasesrc: retry on not-negotiate if a reconfigure is pending
Thiago Santos [Sat, 4 Aug 2012 14:48:13 +0000 (11:48 -0300)]
basesrc: retry on not-negotiate if a reconfigure is pending

Before erroring out on not-negotiated returns, check if the pad
has the reconfigure flag set and retry.

https://bugzilla.gnome.org/show_bug.cgi?id=681198

12 years agopad: add gst_pad_needs_reconfigure
Thiago Santos [Sat, 4 Aug 2012 14:42:05 +0000 (11:42 -0300)]
pad: add gst_pad_needs_reconfigure

Add an alternative version of gst_pad_check_reconfigure that doesn't
clear the reconfigure flag.

Useful for increasing error resilience without duplicating the
reconfigure code in pad task functions.

API: gst_pad_needs_reconfigure

https://bugzilla.gnome.org/show_bug.cgi?id=681198

12 years agopad: add GST_PAD_LINK_CHECK_DEFAULT to GstPadLinkCheck
Evan Nemerson [Sun, 29 Jul 2012 22:44:45 +0000 (15:44 -0700)]
pad: add GST_PAD_LINK_CHECK_DEFAULT to GstPadLinkCheck

This allows introspection-based bindings to access
Gst.PadLinkCheck.DEFAULT instead of
Gst.PAD_LINK_CHECK_DEFAULT.

https://bugzilla.gnome.org/show_bug.cgi?id=678301

12 years agobuffer: mark gst_buffer_wrapped* data as array
Evan Nemerson [Sun, 29 Jul 2012 21:57:41 +0000 (14:57 -0700)]
buffer: mark gst_buffer_wrapped* data as array

https://bugzilla.gnome.org/show_bug.cgi?id=678301

12 years agointrospection: fix some warnings generated by g-ir-scanner.
Evan Nemerson [Tue, 24 Jul 2012 20:26:00 +0000 (13:26 -0700)]
introspection: fix some warnings generated by g-ir-scanner.

https://bugzilla.gnome.org/show_bug.cgi?id=678301

12 years agobuffer: convert gst_buffer_* macros to functions
Evan Nemerson [Tue, 31 Jul 2012 04:46:18 +0000 (21:46 -0700)]
buffer: convert gst_buffer_* macros to functions

GObject Introspection does not support macros.
This is needed for bindings. We can still add back
macros or inline functions again later if we think
it's worth it.

https://bugzilla.gnome.org/show_bug.cgi?id=678301

12 years agonetclientclock: fix printf format in debug message
Tim-Philipp Müller [Fri, 10 Aug 2012 12:50:41 +0000 (13:50 +0100)]
netclientclock: fix printf format in debug message

12 years agobufferpool: fix max_buffers handling
Wim Taymans [Fri, 10 Aug 2012 10:23:03 +0000 (12:23 +0200)]
bufferpool: fix max_buffers handling

When max_buffers > 0 and the pool is empty, actually try to allocate more
buffers up to the max_buffers limit.
We need to add a counter for this to count how many buffers we allocated and
check this against the max_buffers limit.
Reorganise and clean up some code.

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=681153

12 years agonetclientclock: simplify by using g_socket_condition_timed_wait()
Tim-Philipp Müller [Fri, 10 Aug 2012 08:19:25 +0000 (09:19 +0100)]
netclientclock: simplify by using g_socket_condition_timed_wait()

No need to use a custom main context and custom timeout sources,
just use g_socket_condition_timed_wait() instead, which was added
for exactly this case.

Also seems to help with the unit test deadlocking with glib 2.33.x

https://bugzilla.gnome.org/show_bug.cgi?id=681575

12 years agogstobject: fix double string escaping in gst_object_default_deep_notify()
Tim-Philipp Müller [Thu, 9 Aug 2012 18:15:29 +0000 (19:15 +0100)]
gstobject: fix double string escaping in gst_object_default_deep_notify()

Make output of gst-launch -v readable again.

last-message = "event\ \ \ \*\*\*\*\*\*\*\ \(fakesink0:sink\)\ E\ \(type:\ tag\ \(20510\)\,\ GstTagList-stream\,\ taglist\=\(taglist\)\"taglist\\\,\\\ video-codec\\\=\\\(string\\\)H264\\\,\\\
 minimum-bitrate\\\=\\\(uint\\\)636611\\\,\\\ bitrate\\\=\\\(uint\\\)980729\\\,\\\ maximum-bitrate\\\=\\\(uint\\\)1116707\\\;\"\;\)\ 0x15bc760"

vs.

last-message = event   ******* (fakesink0:sink) E (type: tag (20510), GstTagList-stream, taglist=(taglist)"taglist\,\ video-codec\=\(string\)H264\,\ minimum-bitrate\=\(uint\)856039\,\ bitrate
\=\(uint\)1019748\,\ maximum-bitrate\=\(uint\)1116707\;";) 0x11149e0