platform/upstream/gstreamer.git
14 years agocheck: Attempt to fix type-punning warning in the gstiterator test
Jan Schmidt [Wed, 7 Oct 2009 15:36:31 +0000 (16:36 +0100)]
check: Attempt to fix type-punning warning in the gstiterator test

14 years agocheck: Make sure to init the dataprotocol lib.
Jan Schmidt [Wed, 7 Oct 2009 15:00:12 +0000 (16:00 +0100)]
check: Make sure to init the dataprotocol lib.

Call the gst_dp_init() function to ensure that the debug
category is initialised, to avoid g_criticals when running with
GST_DEBUG=5

14 years agocheck: Use GST_DEBUG instead of g_message in the gdp test
Jan Schmidt [Wed, 7 Oct 2009 14:47:45 +0000 (15:47 +0100)]
check: Use GST_DEBUG instead of g_message in the gdp test

14 years agocheck: Add GST_PLUGIN_SCANNER env var to the check environment
Jan Schmidt [Wed, 7 Oct 2009 14:14:46 +0000 (15:14 +0100)]
check: Add GST_PLUGIN_SCANNER env var to the check environment

14 years agoregistry: improve plugin loader failure message for uninstalled setups
Tim-Philipp Müller [Wed, 7 Oct 2009 13:34:17 +0000 (14:34 +0100)]
registry: improve plugin loader failure message for uninstalled setups

Everyone running an uninstalled git setup is going to wonder about
this failure next time they update, so let's mention the solution
in the error message.

14 years agoconfigure: Beef up the test for __uint128_t on GCC
Jan Schmidt [Wed, 7 Oct 2009 12:59:47 +0000 (13:59 +0100)]
configure: Beef up the test for __uint128_t on GCC

GCC 3.4.3 on the SPARC buildbot crashes when actually
using __uint128_t. Beef up the configure test to detect that the
type is actually usable.

14 years agowin32: add new API to .def file
Tim-Philipp Müller [Wed, 7 Oct 2009 08:56:42 +0000 (09:56 +0100)]
win32: add new API to .def file

And add API: marker in commit message that was omitted in the original
commit:

API: gst_data_queue_new_full()

14 years agobenchmarks: add bufferstress binary to .gitignore
Tim-Philipp Müller [Wed, 7 Oct 2009 08:55:15 +0000 (09:55 +0100)]
benchmarks: add bufferstress binary to .gitignore

14 years agoconfigure: bump GLib requirement to 2.18
Tim-Philipp Müller [Wed, 7 Oct 2009 08:42:55 +0000 (09:42 +0100)]
configure: bump GLib requirement to 2.18

Bump GLib requirement as per the release planning docs.

14 years agomessage: whitespace fixes
Wim Taymans [Wed, 7 Oct 2009 08:37:39 +0000 (10:37 +0200)]
message: whitespace fixes

14 years agopad: flip the G_UNLIKELY
Stefan Kost [Wed, 7 Oct 2009 08:12:57 +0000 (11:12 +0300)]
pad: flip the G_UNLIKELY

Its likely that we have caps and unlikely (error) otherwise.

14 years agodocs: add new queue api to the docs to fix the build
Stefan Kost [Wed, 7 Oct 2009 08:04:56 +0000 (11:04 +0300)]
docs: add new queue api to the docs to fix the build

14 years agoplugins/multiqueue: Avoid instance check
Edward Hervey [Mon, 28 Sep 2009 13:25:22 +0000 (15:25 +0200)]
plugins/multiqueue: Avoid instance check

We know earlier on in the code whether we're handling an event or a buffer,
just pass that information through.

This commit and the previous commit reduce instruction fetch:
* when pushing buffer (_chain) by 10%
* when popping buffer (_loop) by 3%

14 years agoplugins/multiqueue: Cache input/output time, avoid expensive calls.
Edward Hervey [Mon, 28 Sep 2009 13:24:02 +0000 (15:24 +0200)]
plugins/multiqueue: Cache input/output time, avoid expensive calls.

* Cache the input/output time
* Only recalculate it when needed.

Avoids 50% calls to gst_segment_to_running_time

14 years agobuild: sprintf, sscanf need stdio.h
Stefan Kost [Wed, 7 Oct 2009 07:00:05 +0000 (10:00 +0300)]
build: sprintf, sscanf need stdio.h

14 years agochildproxy: initialize gvalue in _valist function. Fixes #595602
Stefan Kost [Mon, 5 Oct 2009 08:46:34 +0000 (11:46 +0300)]
childproxy: initialize gvalue in _valist function. Fixes #595602

Reflow the code to move error handling to the end of the functions. Initialize
gvalue like we do in the setter. Add a unit-test module with two simple tests
the catche this bug.

14 years agopad: don't intersect with any in proxy_pad_get_caps
Stefan Kost [Thu, 1 Oct 2009 14:39:45 +0000 (17:39 +0300)]
pad: don't intersect with any in proxy_pad_get_caps

We initialize the caps with any and if a pad has NULL caps, just skip it instead
of intersecting with any. Also add branch prediction here.

14 years agodocs: rename aggregator to adder in the docs.
Stefan Kost [Wed, 30 Sep 2009 13:41:07 +0000 (16:41 +0300)]
docs: rename aggregator to adder in the docs.

14 years agoman: fix copy and past mistake for -q option
Stefan Kost [Wed, 30 Sep 2009 06:47:23 +0000 (09:47 +0300)]
man: fix copy and past mistake for -q option

14 years agogst-uninstalled: Extend environment variables to allow using an uninstalled gstreamer...
Sebastian Dröge [Wed, 7 Oct 2009 07:54:41 +0000 (09:54 +0200)]
gst-uninstalled: Extend environment variables to allow using an uninstalled gstreamer-sharp

14 years agoplugins/multiqueue: Use new GstDataQueue constructor
Edward Hervey [Mon, 28 Sep 2009 13:19:44 +0000 (15:19 +0200)]
plugins/multiqueue: Use new GstDataQueue constructor

14 years agogstdataqueue: new constructor which takes callbacks.
Edward Hervey [Mon, 28 Sep 2009 13:18:37 +0000 (15:18 +0200)]
gstdataqueue: new constructor which takes callbacks.

This allows us to avoid going through glib's signalling system

14 years agoplugins/multiqueue: Use cached value instead of expensive object get.
Edward Hervey [Mon, 28 Sep 2009 11:19:10 +0000 (13:19 +0200)]
plugins/multiqueue: Use cached value instead of expensive object get.

The task will always exist as long as its owner (i.e. the pad) and that
owner's owner (i.e. multiqueue) exist.

Reduces the number of instruction fetches by 36%.

14 years agoplugins/queue: Use previous knowledge of data type to avoid typecheck.
Edward Hervey [Mon, 28 Sep 2009 13:41:52 +0000 (15:41 +0200)]
plugins/queue: Use previous knowledge of data type to avoid typecheck.

We know whether we have a buffer or an event, use that instead of going
trough the expensive GLib typecheck.

The overall instruction fetch reduction introduced by this commit and the
2 previous commits:
* receiving a buffer (_chain) by 20%
* popping a buffer (_loop) by 14%

Numbers acquired through callgrind passing 100000 buffers through queue.

14 years agoplugins/queue: Avoid useless segment_to_running_time() calculations.
Edward Hervey [Mon, 28 Sep 2009 13:20:06 +0000 (15:20 +0200)]
plugins/queue: Avoid useless segment_to_running_time() calculations.

* Cache src and sink time
* Use a boolean to know whether src/sink time need to be recalculated

Avoids 50% calls to gst_segment_to_running_time()

14 years agoplugins/queue: Just cast to the object parent instead of typechecking.
Edward Hervey [Mon, 28 Sep 2009 11:21:07 +0000 (13:21 +0200)]
plugins/queue: Just cast to the object parent instead of typechecking.

14 years agobenchmark: New benchmark for testing contention when creating buffers
Edward Hervey [Wed, 23 Sep 2009 14:19:32 +0000 (16:19 +0200)]
benchmark: New benchmark for testing contention when creating buffers

14 years agogst-launch: Don't activate tracing if not requested.
Edward Hervey [Wed, 23 Sep 2009 14:17:09 +0000 (16:17 +0200)]
gst-launch: Don't activate tracing if not requested.

14 years agotests: init more variables to avoid compiler warning on osx
Edward Hervey [Wed, 7 Oct 2009 06:37:05 +0000 (08:37 +0200)]
tests: init more variables to avoid compiler warning on osx

Init variable to avoid compiler warning and make the build bot happy
(the compiler most likely complains about this because it doesn't know
here that fail_unless will abort/exit in the path where it fails).

14 years agomultiqueue: Improve iterate internal links function
Sebastian Dröge [Sat, 26 Sep 2009 09:43:06 +0000 (11:43 +0200)]
multiqueue: Improve iterate internal links function

Pads have their GstSingleQueue stored as element private data
so there's no need to iterate over the list of single queues
every time. Also every pad only has a single internal link so
use a single iterator instead of a complex custom iterator.

Set the element private data of the pad to NULL when freeing the
single queue.

14 years agointrospection: Add annotations for gst_element_query_{duration,position}
Johan Bilien [Thu, 17 Sep 2009 20:30:43 +0000 (16:30 -0400)]
introspection: Add annotations for gst_element_query_{duration,position}

Fixes bug #595511.

14 years agobytereader: add inline version of gst_byte_reader_skip
Tim-Philipp Müller [Sun, 4 Oct 2009 23:11:20 +0000 (00:11 +0100)]
bytereader: add inline version of gst_byte_reader_skip

14 years agopo: update translation files for new and changed strings
Tim-Philipp Müller [Tue, 6 Oct 2009 23:47:59 +0000 (00:47 +0100)]
po: update translation files for new and changed strings

14 years agoghostpad: take locks around smaller section
Wim Taymans [Mon, 28 Sep 2009 20:43:51 +0000 (22:43 +0200)]
ghostpad: take locks around smaller section

We don't need the hold the proxy mutex locked for getting the internal pad and
for linking the new target pad when we retarget. So take the lock a little later
and release it earlier.

Fixes #596366

14 years agotests: init variable to avoid compiler warning on osx
Tim-Philipp Müller [Sun, 4 Oct 2009 18:51:40 +0000 (19:51 +0100)]
tests: init variable to avoid compiler warning on osx

Init variable to avoid compiler warning and make the build bot happy
(the compiler most likely complains about this because it doesn't know
here that fail_unless will abort/exit in the path where it fails).

14 years agogst: remove more unnecessary cast when using g_signal_*()
Tim-Philipp Müller [Sat, 3 Oct 2009 20:08:54 +0000 (21:08 +0100)]
gst: remove more unnecessary cast when using g_signal_*()

14 years agodataqueue, elements: avoid unnecessary runtime type checks
Tim-Philipp Müller [Sat, 3 Oct 2009 19:49:54 +0000 (20:49 +0100)]
dataqueue, elements: avoid unnecessary runtime type checks

14 years agodocs: Update the release script
Jan Schmidt [Mon, 5 Oct 2009 15:41:50 +0000 (16:41 +0100)]
docs: Update the release script

Remove old cruft from the release script, and change some CVS
references to equivalent git commands

14 years agopluginloader: Add a magic number and maximum size limit.
Jan Schmidt [Sun, 4 Oct 2009 13:30:34 +0000 (14:30 +0100)]
pluginloader: Add a magic number and maximum size limit.

Guard against a hostile child process that sends bogus data
due to memory corruption by adding a magic number to each packet,
and limit the maximum size of any message to 32MB

14 years agoregistry: Also check the binary registry chunk version of the child.
Jan Schmidt [Mon, 9 Feb 2009 13:33:07 +0000 (13:33 +0000)]
registry: Also check the binary registry chunk version of the child.

When trying to find a function plugin-scanner, include a check on the
version of the binary registry chunks it sends, to make sure it's
what we understand.

14 years agoregistry: Support installed/uninstalled plugin-scanner helper
Jan Schmidt [Fri, 6 Feb 2009 09:49:34 +0000 (09:49 +0000)]
registry: Support installed/uninstalled plugin-scanner helper

Add a simple version check when starting the plugin-scanner so we can
verify we're talking to one that talks the same language.
First try a plugin-scanner in the installed path, then try one via the
GST_PLUGIN_SCANNER env var if that doesn't work.
Update the uninstalled script.
Install the plugin-scanner to the libexec dir

14 years agoRemove checking for and mentions of fork where possible.
Jan Schmidt [Fri, 30 Jan 2009 14:18:13 +0000 (14:18 +0000)]
Remove checking for and mentions of fork where possible.

We no longer use fork() directly, instead using glib's spawn
functionality, so don't check for it, and don't use it in the
documentation notes.

14 years agoRe-enable and fix disabled bit of the registry test
Jan Schmidt [Fri, 30 Jan 2009 13:06:13 +0000 (13:06 +0000)]
Re-enable and fix disabled bit of the registry test

14 years agoOnly load the registry cache once per process.
Jan Schmidt [Fri, 30 Jan 2009 13:04:52 +0000 (13:04 +0000)]
Only load the registry cache once per process.

When updating the registry, we don't need to re-read the registry cache
and waste time replacing all our existing, hopefully identical, plugins
and features that we're about to re-scan anyway.

14 years agoAdd some more debug the registry.
Jan Schmidt [Thu, 29 Jan 2009 13:22:14 +0000 (13:22 +0000)]
Add some more debug the registry.

Add the full set of debug about why it's decided that a given plugin is
stale or not, and include the plugin name when finalizing it.

14 years agoAdd restarting of the plugin loader and blacklisting of broken files
Jan Schmidt [Fri, 23 Jan 2009 21:15:43 +0000 (21:15 +0000)]
Add restarting of the plugin loader and blacklisting of broken files

14 years agoPlugin loader phase 2
Jan Schmidt [Fri, 23 Jan 2009 15:47:08 +0000 (15:47 +0000)]
Plugin loader phase 2

phase 2 - make the plugin loader receive the list of plugins to load and
send back the results asynchronously, so we don't context switch back
and forth so much.

14 years agoregistry: Add registry helper phase 1
Jan Schmidt [Sat, 14 Mar 2009 23:07:40 +0000 (23:07 +0000)]
registry: Add registry helper phase 1

Phase 1 of adding the registry scan helper

14 years agoregistry: Rearrange some things.
Jan Schmidt [Mon, 14 Sep 2009 22:31:10 +0000 (23:31 +0100)]
registry: Rearrange some things.

Prepare to land the external plugin helper process

14 years agoBack to development -> 0.10.25.1
Jan Schmidt [Tue, 6 Oct 2009 18:41:38 +0000 (19:41 +0100)]
Back to development -> 0.10.25.1

14 years agoRelease 0.10.25 RELEASE-0.10.25
Jan Schmidt [Mon, 5 Oct 2009 11:57:03 +0000 (12:57 +0100)]
Release 0.10.25

14 years agoUpdate .po files
Jan Schmidt [Mon, 5 Oct 2009 11:41:42 +0000 (12:41 +0100)]
Update .po files

14 years ago0.10.24.4 pre-release
Jan Schmidt [Thu, 1 Oct 2009 15:24:52 +0000 (16:24 +0100)]
0.10.24.4 pre-release

14 years agolibgstcheck: Don't use character classes in sed expressions
Jan Schmidt [Wed, 30 Sep 2009 14:52:33 +0000 (15:52 +0100)]
libgstcheck: Don't use character classes in sed expressions

Apparently the sed that ships on Solaris 10 doesn't support character
classes like [:alnum:], so don't use them. We don't need them for the
symbol names that are being extracted anyway.

Also, use $(SED) instead of 'sed'

Fixes: #596877

14 years ago0.10.24.3 pre-release
Jan Schmidt [Thu, 17 Sep 2009 00:20:03 +0000 (01:20 +0100)]
0.10.24.3 pre-release

14 years agoutils: Fix GMP scaling unit test
Sebastian Dröge [Tue, 15 Sep 2009 07:41:28 +0000 (09:41 +0200)]
utils: Fix GMP scaling unit test

GMP only uses "unsigned long int", which is 32 bit
on 32 bit architectures and can't hold a guint64.
This resulted in false unit test failures on 32 bit architectures.

Fixes bug #595133.

14 years agoFix out-of-tree build
David Schleef [Mon, 14 Sep 2009 19:47:26 +0000 (12:47 -0700)]
Fix out-of-tree build

14 years agodocs: GST_MESSAGE_STREAM_STATUS is implemented nowadays.
Stefan Kost [Mon, 14 Sep 2009 11:07:55 +0000 (14:07 +0300)]
docs: GST_MESSAGE_STREAM_STATUS is implemented nowadays.

Docs were still mention it as "not yet implemented".

14 years agointrospection: Build pkgconfig before all libraries and set PKG_CONFIG_PATH
Sebastian Dröge [Sat, 12 Sep 2009 11:52:00 +0000 (13:52 +0200)]
introspection: Build pkgconfig before all libraries and set PKG_CONFIG_PATH

This way g-ir-scanner can find the gstreamer-0.10 pkg-config file.

14 years agointrospection: Don't typedef GstTagList to GstStructure for gobject-introspection
Sebastian Dröge [Sat, 12 Sep 2009 11:51:29 +0000 (13:51 +0200)]
introspection: Don't typedef GstTagList to GstStructure for gobject-introspection

14 years ago0.10.24.2 pre-release
Jan Schmidt [Fri, 11 Sep 2009 22:21:30 +0000 (23:21 +0100)]
0.10.24.2 pre-release

14 years agoDon't use C++ style comments
Jan Schmidt [Fri, 11 Sep 2009 21:42:51 +0000 (22:42 +0100)]
Don't use C++ style comments

14 years agomessage: Disable restriction that structure changes are sink pads
Jan Schmidt [Fri, 11 Sep 2009 21:22:34 +0000 (22:22 +0100)]
message: Disable restriction that structure changes are sink pads

The structure_change message was originally emitted on source pads and
then recently changed to be sink pads. This causes a failure in the
gst-python testsuite. Disable the restriction so that the published
behaviour is still allowed.

14 years agocheck: Fix version check tests
Jan Schmidt [Fri, 11 Sep 2009 17:24:18 +0000 (18:24 +0100)]
check: Fix version check tests

Accomodate the slightly changed semantics in the plugin version check
where a CVS version just before a release is acceptable.

14 years agobinaryregistry: don't crash in cleaning up on error.
Stefan Kost [Fri, 11 Sep 2009 18:20:57 +0000 (21:20 +0300)]
binaryregistry: don't crash in cleaning up on error.

Don't dereference NULL pointers.

14 years agodebug: use dummy code to avoid spurious semicolons
David Schleef [Mon, 20 Jul 2009 19:54:00 +0000 (12:54 -0700)]
debug: use dummy code to avoid spurious semicolons

Fixes bug #589173.

14 years agowhitespace fixes
Wim Taymans [Thu, 10 Sep 2009 09:53:09 +0000 (11:53 +0200)]
whitespace fixes

14 years agopluginfeature: improve version check
Wim Taymans [Thu, 10 Sep 2009 09:41:56 +0000 (11:41 +0200)]
pluginfeature: improve version check

Also parse the nano of the version and assume that X.Y.Z-1.1 >= X.Y.Z
With this change we can also check development versions against the version of
the upcomming release.

14 years agotaglist: Add FIXME for 0.11 to not typedef GstTagList to be a GstStructure
Sebastian Dröge [Thu, 10 Sep 2009 08:05:38 +0000 (10:05 +0200)]
taglist: Add FIXME for 0.11 to not typedef GstTagList to be a GstStructure

See bug #518934.

14 years agoFix typo in inline documentation
David Schleef [Wed, 9 Sep 2009 23:29:10 +0000 (16:29 -0700)]
Fix typo in inline documentation

14 years agoUpdate common
Tim-Philipp Müller [Wed, 9 Sep 2009 16:57:54 +0000 (17:57 +0100)]
Update common

14 years agoutils: Add a comment to the scaling functions to explain why the rounding is correct
Sebastian Dröge [Wed, 9 Sep 2009 16:38:29 +0000 (18:38 +0200)]
utils: Add a comment to the scaling functions to explain why the rounding is correct

14 years agoghostpad: don't unref NULL caps
Wim Taymans [Wed, 9 Sep 2009 14:45:17 +0000 (16:45 +0200)]
ghostpad: don't unref NULL caps

Caps can be NULL so don't call unref on it unconditionally, instead use an
existing exit pad for the function.

14 years agoutils: Use gcc's __uint128_t for 64bit unsigned integer scaling
Sebastian Dröge [Wed, 9 Sep 2009 12:53:16 +0000 (14:53 +0200)]
utils: Use gcc's __uint128_t for 64bit unsigned integer scaling

This is available in newer gcc releases and it should only exist
on platforms that provide some native 128bit integer arithmetic
instructions.

The x86-64 assembly for this is still kept for non-gcc compilers
that don't provide __uint128_t magic.

14 years agodesign: add ideas for buffer management
Stefan Kost [Wed, 9 Sep 2009 06:38:54 +0000 (09:38 +0300)]
design: add ideas for buffer management

Right now we're operating suboptimal when talking to kernel interfaces. Write
doesn some ideas.

14 years agofix whitespace
Wim Taymans [Mon, 7 Sep 2009 16:27:08 +0000 (18:27 +0200)]
fix whitespace

14 years agoadapter: fix whitespace
Wim Taymans [Thu, 3 Sep 2009 17:06:28 +0000 (19:06 +0200)]
adapter: fix whitespace

14 years agodocs: Fix typo in gst_value_union()
Benjamin Otte [Mon, 7 Sep 2009 14:14:57 +0000 (16:14 +0200)]
docs: Fix typo in gst_value_union()

14 years agobitreader, bytereader: add some FIXME 0.11 comments and fix indenting
Tim-Philipp Müller [Sun, 6 Sep 2009 18:43:47 +0000 (19:43 +0100)]
bitreader, bytereader: add some FIXME 0.11 comments and fix indenting

14 years agobytereader: add unchecked and inline versions of the float getters/peekers
Tim-Philipp Müller [Fri, 4 Sep 2009 16:15:35 +0000 (17:15 +0100)]
bytereader: add unchecked and inline versions of the float getters/peekers

API: gst_byte_reader_get_float*_unchecked()

14 years agobytereader: add inline versions of the most common getters and setters
Tim-Philipp Müller [Fri, 4 Sep 2009 15:52:06 +0000 (16:52 +0100)]
bytereader: add inline versions of the most common getters and setters

14 years agobytereader: add inlined _unchecked() variants for some functions
Tim-Philipp Müller [Wed, 2 Sep 2009 10:20:04 +0000 (11:20 +0100)]
bytereader: add inlined _unchecked() variants for some functions

API: gst_byte_reader_skip_unchecked()
API: gst_byte_reader_peek_*_unchecked()
API: gst_byte_reader_get_*_unchecked()
API: gst_byte_reader_{peek,get,dup}_data_unchecked()

14 years agointrospection: Strip Gst prefix from all types/functions
Sebastian Dröge [Sat, 5 Sep 2009 10:30:07 +0000 (12:30 +0200)]
introspection: Strip Gst prefix from all types/functions

14 years agointrospection: Fix for out-of-tree builds
Sebastian Dröge [Sat, 5 Sep 2009 10:22:37 +0000 (12:22 +0200)]
introspection: Fix for out-of-tree builds

14 years agointrospection: Fix out-of-tree build
Sebastian Dröge [Sat, 5 Sep 2009 10:04:59 +0000 (12:04 +0200)]
introspection: Fix out-of-tree build

14 years agointrospection: Fix build if gir-repository is not installed
Sebastian Dröge [Sat, 5 Sep 2009 09:51:55 +0000 (11:51 +0200)]
introspection: Fix build if gir-repository is not installed

14 years agonet: Add gobject-introspection support
Sebastian Dröge [Sat, 5 Sep 2009 07:36:21 +0000 (09:36 +0200)]
net: Add gobject-introspection support

14 years agodataprotocol: Add gobject-introspection support
Sebastian Dröge [Sat, 5 Sep 2009 07:34:30 +0000 (09:34 +0200)]
dataprotocol: Add gobject-introspection support

Because of a bug in gobject-introspection this is disabled for now.

14 years agocontroller: Add gobject-introspection support
Sebastian Dröge [Sat, 5 Sep 2009 07:28:48 +0000 (09:28 +0200)]
controller: Add gobject-introspection support

14 years agocheck: Add gobject-introspection support
Sebastian Dröge [Sat, 5 Sep 2009 07:27:04 +0000 (09:27 +0200)]
check: Add gobject-introspection support

14 years agogstbase: Add gobject-introspection support
Sebastian Dröge [Sat, 5 Sep 2009 07:23:19 +0000 (09:23 +0200)]
gstbase: Add gobject-introspection support

14 years agogst: Add gobject-introspection support
Sebastian Dröge [Fri, 4 Sep 2009 18:56:43 +0000 (20:56 +0200)]
gst: Add gobject-introspection support

Partially fixes bug #550616.

14 years agoAutomatic update of common submodule
Sebastian Dröge [Sat, 5 Sep 2009 08:19:23 +0000 (10:19 +0200)]
Automatic update of common submodule

From 94f95e3 to 19fa4f3

14 years agodocs: fix docs for gst_byte_reader_{get|peek}_float*()
Tim-Philipp Müller [Fri, 4 Sep 2009 18:37:15 +0000 (19:37 +0100)]
docs: fix docs for gst_byte_reader_{get|peek}_float*()

14 years agoevent: whitespace fixes
Wim Taymans [Fri, 4 Sep 2009 09:35:59 +0000 (11:35 +0200)]
event: whitespace fixes

14 years agobin: Only unref EOS message after it is not used anymore
Aurelien Grimaud [Fri, 4 Sep 2009 07:51:26 +0000 (09:51 +0200)]
bin: Only unref EOS message after it is not used anymore

Fixes bug #594107.

14 years agostates: post structure change on sinkpads
Wim Taymans [Wed, 2 Sep 2009 16:54:06 +0000 (18:54 +0200)]
states: post structure change on sinkpads

Post the structure change messages on the sinkpads of the elements. This allows
us to catch unlinked pads earlier without ending up with inconsistent element
degrees.

14 years agobin: avoid false 'loop detected' warnings
Wim Taymans [Wed, 2 Sep 2009 16:13:22 +0000 (18:13 +0200)]
bin: avoid false 'loop detected' warnings

When we detect a pad unlink in progress, we will not be updating the degree of
the parent element. This can cause false loop detected warnings because the
degree counter is invalid. Handle this case by marking the iterator as 'dirty'
when we detect a pad unlink and avoid emiting the warning in this case. We have
to continue our state change as good as we can, we will eventually resync when
the pad unlink completed.

14 years agobasesrc: whitespace fixes
Wim Taymans [Tue, 1 Sep 2009 14:49:56 +0000 (16:49 +0200)]
basesrc: whitespace fixes

14 years agobuffer: whitespace fixes
Wim Taymans [Tue, 1 Sep 2009 14:49:22 +0000 (16:49 +0200)]
buffer: whitespace fixes