Andoni Morales Alastruey [Wed, 17 Feb 2010 00:27:22 +0000 (01:27 +0100)]
Fix compilation of fdsink and fdsrc with MSVC
Tim-Philipp Müller [Thu, 18 Feb 2010 14:58:52 +0000 (14:58 +0000)]
0.10.26.2 pre-release
Tim-Philipp Müller [Thu, 18 Feb 2010 13:12:39 +0000 (13:12 +0000)]
.gitignore: ignore some more temporary docs cruft
Tim-Philipp Müller [Thu, 18 Feb 2010 13:09:17 +0000 (13:09 +0000)]
build: fix indenting in win32-update target
No idea why we need to run gst-indent twice on that file, but it
only seems to settle on a final format with minimal diff to the
one in git after two runs.
Tim-Philipp Müller [Thu, 18 Feb 2010 13:08:32 +0000 (13:08 +0000)]
gstcheck: more debug logging for gst_check_element_push_buffer_list()
Thiago Santos [Thu, 18 Feb 2010 14:52:28 +0000 (11:52 -0300)]
collectpads: Improve docs about 'data' attribute
Adds a reminder to 'data' attribute doc
Fixes #610366
Mark Nauwelaerts [Thu, 18 Feb 2010 16:15:35 +0000 (17:15 +0100)]
memindex: avoid busy loop when doing EXACT lookup
Fixes #610367.
Johan Bilien [Tue, 22 Dec 2009 03:09:10 +0000 (11:09 +0800)]
introspection: add annotation for gst_element_get_state
state and pending are "out" arguments.
Fixes #605189.
Edward Hervey [Wed, 17 Feb 2010 11:16:37 +0000 (12:16 +0100)]
filesrc: Don't use expensive cast checks in _create
_create() is a pad function set by ourselves, therefore we're sure basesrc
is a GstFileSrc.
Speeds up _create() by 17% and the total call by 8% (instruction calls measurements
done with valgrind).
Fixes #610246
Edward Hervey [Wed, 17 Feb 2010 11:14:09 +0000 (12:14 +0100)]
basesrc: Don't use expensive cast checks in get_range.
_get_range() is a pad function set by ourselves, therefore we're certain that
the parent is a GstBaseSrc.
Speeds up _get_range by 38%, and the total call by 30%. (valgrind instruction
calls measurements).
Fixes #610246
Stefan Kost [Wed, 17 Feb 2010 09:31:07 +0000 (11:31 +0200)]
fdsrc: cleanup parameter initialisation and add comemnt+logging
Initialize new_fd with DEFAULT_FD and fd with -1. Setting the property will set
new_fd and in _update_fd() we cehck fd against -1. Also add a coment about the
warning we get in the log from gst_poll_remove_fd(). We could get rid of the
warning if we want by tracking if fd has been added to fdset.
Stefan Kost [Wed, 17 Feb 2010 07:55:52 +0000 (09:55 +0200)]
design: write about the current state of tag-handling
Document the taghandling in gstreamer. List gaps and propose new mechanisms to
deal with them.
Stefan Kost [Tue, 16 Feb 2010 08:27:18 +0000 (10:27 +0200)]
taglist: remove blank lines in variable declarations
Tim-Philipp Müller [Tue, 16 Feb 2010 11:30:35 +0000 (11:30 +0000)]
tools: call g_set_prgname() before doing the option parsing
g_setprgname is implicitly called by g_option_context_new() with a check
to see if it's been set already, so set it before g_option_context_new()
Move version printing back until after the options have been parsed,
otherwise it won't work, since it evaluates a flag set by the
option parser.
Tim-Philipp Müller [Tue, 16 Feb 2010 11:24:33 +0000 (11:24 +0000)]
Revert "tools: Move gst_tools_print_version call to avoid warning from new GLib."
This reverts commit
93dd95f02ef3fa530f54ce81e8ffba96f3b679cb.
This commit made --version not work any longer. The g_setprgname()
warning is fixed in recent GLib versions.
Tim-Philipp Müller [Tue, 16 Feb 2010 08:26:59 +0000 (08:26 +0000)]
build: make sure gst-plugin-scanner gets installed where we expect it
Add check to make sure gst-plugin-scanner really gets installed where
we will look for it later, ie. paths and prefixes are set at configure
time and not specified via make.
Fixes #609941.
Stefan Kost [Mon, 15 Feb 2010 21:02:59 +0000 (23:02 +0200)]
docs: prefer short desc from GstElementDetails
Tim-Philipp Müller [Mon, 15 Feb 2010 01:24:25 +0000 (01:24 +0000)]
docs: fix gtk-doc chunk for gst_check_element_push_buffer_list()
Tim-Philipp Müller [Mon, 15 Feb 2010 00:31:16 +0000 (00:31 +0000)]
po: update po files for new comments
Tim-Philipp Müller [Mon, 15 Feb 2010 00:29:37 +0000 (00:29 +0000)]
tags: wrap long string constants
And fix indenting issue
Tim-Philipp Müller [Mon, 15 Feb 2010 00:21:43 +0000 (00:21 +0000)]
tags: add some comments for translators so tag mnemonics get translated correctly
We want 'preview image' translated as a noun, not as 'preview [the] image'.
Tim-Philipp Müller [Thu, 4 Feb 2010 17:43:32 +0000 (17:43 +0000)]
pad: don't print WARN debug statements for normal things like EOS
Sebastian Dröge [Sun, 14 Feb 2010 22:15:45 +0000 (23:15 +0100)]
Automatic update of common submodule
From 96dc793 to 44ecce7
Edward Hervey [Sat, 13 Feb 2010 14:18:05 +0000 (15:18 +0100)]
typefind: Reset the working mode when going to READY/NULL
This allows properly re-using typefind (else it would think it's
already done the typefinding when being re-used with another
stream).
Thiago Santos [Fri, 22 Jan 2010 14:38:59 +0000 (11:38 -0300)]
bytewriter: Adds a test for _fill
Thiago Santos [Fri, 22 Jan 2010 12:19:31 +0000 (09:19 -0300)]
bytewriter: add _fill function
Adds a new function to GstByteWriter that writes
a constant value to a memory area (aka memset).
Useful for adding padding to buffers.
Also updates .def file and docs.
API: gst_byte_writer_fill()
Thiago Santos [Thu, 28 Jan 2010 14:57:33 +0000 (11:57 -0300)]
typefind: Avoid messing pads activation
Typefind might mess up pads modes (pull/push) if a
downstream element is plugged and its pads activated
in 'step 2' of typefind pads activation.
This happens because the following steps don't check
if we already emitted typefound due to upstream setting
caps on buffers being pulled in the typefind helpers.
Avoid that by checking if typefound is already emmited.
Fixes #608036
Sebastian Dröge [Fri, 12 Feb 2010 13:49:52 +0000 (14:49 +0100)]
basesrc: Make locking of the segment a bit more strict and update documentation
Updating the segment values must only be done while holding the
STREAM_LOCK and OBJECT_LOCK. This means, reading can be done as
long as one of them is held, not both, which removes some lock-unlock
blocks from performance critical code paths.
Also document, that gst_base_src_set_format() *must* be called in
states <= READY and add an assertion for this. Changing the format
later will completely mess up the segment information.
Arun Raghavan [Mon, 8 Feb 2010 03:42:01 +0000 (09:12 +0530)]
pwg: several typo fixes
Fixes #609286.
Sebastian Dröge [Tue, 9 Feb 2010 16:52:13 +0000 (17:52 +0100)]
basesrc: Protect segment values from concurrent access from different threads
This could happen easily in the query functions or when the size is set
on appsrc from some non-streaming thread.
Sebastian Dröge [Thu, 4 Feb 2010 20:11:25 +0000 (21:11 +0100)]
typefindelement: Protect internal fields from concurrent changes from different threads
Fixes bug #608877.
Tim-Philipp Müller [Thu, 11 Feb 2010 20:14:59 +0000 (20:14 +0000)]
gst-launch: don't leak timeout GSource
Tim-Philipp Müller [Thu, 11 Feb 2010 00:18:39 +0000 (00:18 +0000)]
docs: flesh out release doc some more
Tim-Philipp Müller [Thu, 11 Feb 2010 01:10:38 +0000 (01:10 +0000)]
Update MAINTAINERS, add myself
Tim-Philipp Müller [Thu, 11 Feb 2010 19:49:00 +0000 (19:49 +0000)]
configure: back to development
Slushy freeze remains in effect.
Tim-Philipp Müller [Wed, 10 Feb 2010 19:17:28 +0000 (19:17 +0000)]
Release 0.10.26
Tim-Philipp Müller [Wed, 10 Feb 2010 15:32:06 +0000 (15:32 +0000)]
Update .po files
Tim-Philipp Müller [Tue, 9 Feb 2010 15:52:23 +0000 (15:52 +0000)]
configure: define GST_PLUGIN_SCANNER_INSTALLED in win32 config.h
Even if it's not used, it still needs to be defined for things to
compile.
Tim-Philipp Müller [Tue, 9 Feb 2010 10:19:04 +0000 (10:19 +0000)]
gst_private: MSVC doesn't seem to like #warning
Visual Studio complains about "invalid preprocessor command 'warning'"
even if the ifdef doesn't trigger, so just remove this again.
Edward Hervey [Wed, 10 Feb 2010 13:40:17 +0000 (14:40 +0100)]
tests: Fix multiqueue test for latest commits.
The problem lies in the fact that multiqueue will now operate somewhat
similarly to the flow aggregation logic of demuxers and therefore
will stopp whenever all downstream pads return NOT_LINKED and/or
UNEXPECTED and there's no more buffers to push.
The latest commits should not affect any regular use-case, but the bug
report will be kept open so the previous behaviour can be re-established
if needed.
Fixes #609486
Edward Hervey [Tue, 9 Feb 2010 14:51:18 +0000 (15:51 +0100)]
multiqueue: Don't stop threads on UNEXPECTED and forward flow returns.
When a downstream element returns GST_FLOW_UNEXPECTED we want to:
* let the dataqueue task running
* forward the flow return upstream.
This allows upstream elements to push EOS, and have that EOS event come
downstream.
Fixes #609274
Edward Hervey [Tue, 9 Feb 2010 12:35:08 +0000 (13:35 +0100)]
Revert "multiqueue: handle UNEXPECTED flowreturn better"
This reverts commit
fbdf4dcedad8692f1e3d8838551188987e462e74.
Partly fixes #609274
Robert Swain [Thu, 28 Jan 2010 06:27:49 +0000 (07:27 +0100)]
git-update.sh: Fix issues
Tim-Philipp Müller [Sun, 7 Feb 2010 08:59:25 +0000 (09:59 +0100)]
Revert "docs: fix ASCII art so that iterators are aligned property to the diagram"
This reverts commit
ae60d06e9e401d1ed4de5ef25b5c283db0696a31 (fixes: #609166)
Tim-Philipp Müller [Thu, 4 Feb 2010 18:30:56 +0000 (18:30 +0000)]
0.10.25.3 pre-release
Tim-Philipp Müller [Thu, 4 Feb 2010 17:45:46 +0000 (17:45 +0000)]
po: translation updates
Tim-Philipp Müller [Mon, 1 Feb 2010 12:50:40 +0000 (12:50 +0000)]
docs: fix ASCII art so that iterators are aligned property to the diagram
Thiago Santos [Mon, 1 Feb 2010 20:40:08 +0000 (17:40 -0300)]
gstbytewriter: Fix different function names in .h and .c
gst_byte_writer_reset_and_get_buffer wasn't declared
in .h, instead there was _reset_and_get_data_as_buffer.
Replace it with the real function name, that is smaller
and matches gst_byte_writer_free_and_get_buffer
https://bugzilla.gnome.org/show_bug.cgi?id=608726
Tim-Philipp Müller [Sun, 31 Jan 2010 17:30:54 +0000 (17:30 +0000)]
docs: add some more Since: markers to buffer list docs
Tim-Philipp Müller [Sat, 30 Jan 2010 18:57:44 +0000 (18:57 +0000)]
filesrc: fix typo in warning message
Spotted by bsreerenj@gmail.com.
Fixes #608442.
Tim-Philipp Müller [Sat, 30 Jan 2010 15:17:48 +0000 (15:17 +0000)]
Automatic update of common submodule
From 15d47a6 to 96dc793
Tim-Philipp Müller [Sat, 30 Jan 2010 13:45:58 +0000 (13:45 +0000)]
init: don't spew warning about late g_thread_init()s if GLib >= 2.23.2
Late g_thread_init() is fine with newer GLib versions and done automatically
from g_type_init() there, so don't warn if the application hasn't called
g_thread_init() yet when gst_init() is called with new GLib versions.
Fixes #608398.
Tim-Philipp Müller [Fri, 29 Jan 2010 09:41:30 +0000 (09:41 +0000)]
pkgconfig: don't put -DG_THREADS_MANDATORY into our pkg-config CFLAGS
If we force -DG_THREADS_MANDATORY onto apps, then g_thread_supported()
will always evaluate to TRUE, so the typical thread initialisation
boilerplate code if (!g_thread_supported()) g_thread_init(NULL); will
no longer work, and the threading system not be initialised and us
printing a warning in gst_init. This may be fine in most cases, since
late initialisation is allowed and automatically done in g_type_init()
since GLib 2.23.2, but let's be cautious and only use this define when
compiling GStreamer itself.
See #608398.
Wim Taymans [Thu, 28 Jan 2010 14:55:27 +0000 (15:55 +0100)]
pipeline: Take start_time after chaining up too
Refactor the code to take the current start_time when going to PAUSED.
Make sure we also call the start_time update code after we chained up to the
parent bin.
Fixes #607842
Tim-Philipp Müller [Tue, 26 Jan 2010 18:59:50 +0000 (18:59 +0000)]
0.10.25.2 pre-release
Tim-Philipp Müller [Wed, 27 Jan 2010 00:23:07 +0000 (00:23 +0000)]
checks: fix spurious ghost pad check failure
Tim-Philipp Müller [Tue, 26 Jan 2010 19:35:52 +0000 (19:35 +0000)]
win32: update windows headers to latest version
Tim-Philipp Müller [Tue, 26 Jan 2010 19:32:48 +0000 (19:32 +0000)]
docs: minor update to release notes
Tim-Philipp Müller [Tue, 26 Jan 2010 18:45:38 +0000 (18:45 +0000)]
po: update translation files
Tim-Philipp Müller [Tue, 26 Jan 2010 18:39:45 +0000 (18:39 +0000)]
tests: fix warning in rtpool-test
The stream status message object may be of a non-GObject type, e.g.
G_TYPE_POINTER (see GstAudioSrc), so print that properly instead
of assuming the value holds an object.
Wim Taymans [Tue, 26 Jan 2010 11:43:09 +0000 (12:43 +0100)]
multiqueue: handle UNEXPECTED flowreturn better
When we receive an UNEXPECTED flowreturn from downstream, we must not shutdown
the pushing thread because upstream will at some point push an EOS that we still
need to push further downstream.
To achieve this, convert the UNEXPECTED return value to OK. Add a fixme so that
we implement the right logic to propagate the flowreturn upstream at some point.
Also clean up the unit test a little.
Fixes #608136
Thiago Santos [Tue, 26 Jan 2010 11:52:16 +0000 (08:52 -0300)]
docs: Fix basics-bus docs
Fix wrong information about bus watch functions in the
application development manual.
Fixes #608127
Stefan Kost [Sun, 24 Jan 2010 21:12:22 +0000 (23:12 +0200)]
bin: also remove private function from def file
Stefan Kost [Sun, 24 Jan 2010 21:04:27 +0000 (23:04 +0200)]
bin: make a interface vmethod implementation static
This should not cause any troubles - the methods wasn't in any header.
Stefan Kost [Sun, 24 Jan 2010 20:22:07 +0000 (22:22 +0200)]
childproxy: remove ; after }
Wim Taymans [Fri, 22 Jan 2010 17:00:53 +0000 (18:00 +0100)]
queue2: add some docs to mark new property
Wim Taymans [Fri, 22 Jan 2010 16:55:39 +0000 (17:55 +0100)]
queue2: add option to remove the temp-file
Add an option to automatically remove the temp file (TRUE by default). This
should make it possible for the application to keep the temp file by other means
than hardlinking or holding an fd open.
Fixes #607739
Tim-Philipp Müller [Fri, 22 Jan 2010 02:02:01 +0000 (02:02 +0000)]
typefind: don't leak uri string
Wim Taymans [Thu, 21 Jan 2010 15:19:44 +0000 (16:19 +0100)]
gstindex: retab .h file
Benjamin Otte [Wed, 20 Jan 2010 13:13:11 +0000 (14:13 +0100)]
tools: Run g_thread_init() unconditionally
Since we define G_THREADS_MANDATORY, g_thread_supported() evaluates to
TRUE unconditionally, so calling g_thread_init() never happened.
Tim-Philipp Müller [Wed, 20 Jan 2010 10:58:29 +0000 (10:58 +0000)]
pluginloader: fix compiler warning on win32
Move variable that's only used on unix into the unix block so that
the compiler doesn't complain about the unused variable on win32
(see #597662).
Tim-Philipp Müller [Wed, 20 Jan 2010 09:45:06 +0000 (09:45 +0000)]
pluginloader: try scanner set via env var before using the installed one
If the GST_PLUGIN_SCANNER environment variable is set, we should try
the scanner specified there first, to make sure the right scanner binary
is used for uninstalled setups and builds from source when there's
already an installed version.
Sebastian Dröge [Wed, 20 Jan 2010 05:58:28 +0000 (06:58 +0100)]
build: Define G_THREADS_MANDATORY everywhere
We require threads to be supported in any case and defining this
will simplify the mutex, condition variable, etc. macros from gthread
to not always check if threads are really supported.
Fixes bug #607481.
Andoni Morales Alastruey [Fri, 8 Jan 2010 19:56:18 +0000 (20:56 +0100)]
pluginloader: disable external plugin loader on Windows until it is ported properly
See #597662.
Tim-Philipp Müller [Wed, 20 Jan 2010 01:09:55 +0000 (01:09 +0000)]
gst_private.h: make sure gst_private.h is included before glib.h
For the reason outlined at the beginning of gst_private.h (inline
functions in glib may need the g_log_domain variable). Also include
gst_private.h before using any G_OS_* defines, esp. in plugin loader.
Tim-Philipp Müller [Wed, 20 Jan 2010 01:33:38 +0000 (01:33 +0000)]
docs: minor gtk-doc markup fixes
Tim-Philipp Müller [Wed, 20 Jan 2010 00:53:01 +0000 (00:53 +0000)]
Automatic update of common submodule
From 14cec89 to 15d47a6
Wim Taymans [Tue, 19 Jan 2010 15:39:38 +0000 (16:39 +0100)]
docs: small docs updates
Thijs Vermeir [Tue, 19 Jan 2010 13:07:23 +0000 (14:07 +0100)]
gstpad: directly set the caps when pushing buffer with different caps.
This check is not necesarry as we are not negotiating anymore. And it can
be wrong if upstream can't produce this caps anymore, but downstream can
process them fine.
Sebastian Dröge [Mon, 18 Jan 2010 12:57:29 +0000 (13:57 +0100)]
miniobject: The GValue collection function can not assume that the destination is initialized
...and it will usually be either filled by zeroes or random values.
Fixes bug #607283.
Sebastian Dröge [Sat, 16 Jan 2010 20:52:06 +0000 (21:52 +0100)]
basetransform: Only use suggested caps in buffer allocation if a size was suggested too
Tim-Philipp Müller [Sat, 16 Jan 2010 19:41:29 +0000 (19:41 +0000)]
structure: remove superfluous guard against NULL
All callers of this static function already check for NULL-ness
themselves, so no need to do it again (and if we do it, we should
probably do so before dereferencing the pointer for the first time).
Tim-Philipp Müller [Thu, 17 Dec 2009 19:45:31 +0000 (19:45 +0000)]
structure: micro-optimise some getters
Avoid checking the GType of the value twice (once on our side and
once in g_value_get_*()) by by-passing g_value_get() and accessing
the GValue structure directly.
Wim Taymans [Fri, 15 Jan 2010 17:36:07 +0000 (18:36 +0100)]
message: update docs a little
Tim-Philipp Müller [Fri, 15 Jan 2010 00:46:23 +0000 (00:46 +0000)]
docs: minor release docs update
Sebastian Dröge [Thu, 14 Jan 2010 19:19:05 +0000 (20:19 +0100)]
basetransform: Handle buffers with NULL caps correctly
This means that the caps didn't change so don't try to handle
the NULL caps as the new caps.
Sebastian Dröge [Thu, 14 Jan 2010 09:44:16 +0000 (10:44 +0100)]
docs: Move field specific Since markers at the same line
Fixes gobject-introspection warnings about Since being defined multiple times.
Tim-Philipp Müller [Wed, 13 Jan 2010 10:17:03 +0000 (10:17 +0000)]
faq: remove revision history that no one updates or cares about anyway
Tim-Philipp Müller [Wed, 13 Jan 2010 09:32:59 +0000 (09:32 +0000)]
faq: fix link to gst-uninstalled on cgit
Fix link to gst-uninstalled now that it's been moved, and fix a typo
while we're at it. Also add a new section to 'Building GStreamer from
git' that points to the 'How do I develop against an uninstalled copy
of GStreamer' section.
Stefan Kost [Wed, 13 Jan 2010 08:32:46 +0000 (10:32 +0200)]
docs: we're in git since a while
Stefan Kost [Wed, 13 Jan 2010 08:31:26 +0000 (10:31 +0200)]
scripts: move gst-uninstalled from docs/faq to scripts
Don't include the long gst-uninstalled script in verbatim in the faq anymore
(there is a link to cgit). Dist the script under its new location.
Tim-Philipp Müller [Tue, 12 Jan 2010 21:34:03 +0000 (21:34 +0000)]
registry: avoid some more unnecessary malloc/frees
Tim-Philipp Müller [Tue, 12 Jan 2010 20:21:32 +0000 (20:21 +0000)]
registry: avoid some unnecessary strdup/free when reading the binary registry
Strings in the binary registry are NUL-terminated, so we can just use them
directly if we only need them temporarily, and avoid unnecessary mallocs
and frees.
Wim Taymans [Tue, 12 Jan 2010 16:38:32 +0000 (17:38 +0100)]
typefindelement: use new typefind function
Refactor a little.
Use the new typefind helper function that uses the extension to speed up
typefinding.
Wim Taymans [Tue, 12 Jan 2010 16:34:39 +0000 (17:34 +0100)]
typefind: add a new method that also uses the file extension
Add a method to perform get_range typefinding that also uses the
uri/location extension as an extra hint. It will first try to call the
typefind functions of the factories that handle the given extension. The result
is that in the common case, we only call one typefind function, which speeds up
the typefinding a lot.
Wim Taymans [Mon, 11 Jan 2010 13:58:11 +0000 (14:58 +0100)]
docs: update QoS documeent
Add some ideas about a new QoS message.
See also #322947
Håvard Graff [Mon, 11 Jan 2010 10:38:32 +0000 (11:38 +0100)]
tee: make release_pad threadsafe
Protect the ->removed field with the object lock as well. Take the DYN lock
earlier so that we can mark the pad removed and avoid a race in pad_alloc.
Fixes #606435
Wim Taymans [Fri, 11 Dec 2009 16:46:42 +0000 (17:46 +0100)]
bus: whitespace fixes
Stefan Kost [Sun, 10 Jan 2010 19:49:25 +0000 (21:49 +0200)]
utils: defer getting the classes until we actualy need them
This function has a lot of early returns. Give them soem more benefit.
Stefan Kost [Sun, 10 Jan 2010 19:40:24 +0000 (21:40 +0200)]
utils: avoid extra hop in gst_element_link
No need to call gst_element_link_pads_filtered with filter=NULL, which would
call gst_element_link_pads() in that way. Call it directly to save a call and
expensive gobject type checks.