Vincent Penquerc'h [Tue, 20 Jan 2015 15:08:24 +0000 (15:08 +0000)]
dcodebin2: fix lock/unlock mismatch on multiqueue overrun
Jan Alexander Steffens (heftig) [Tue, 13 Jan 2015 15:07:06 +0000 (16:07 +0100)]
audioresample: Try to prevent endless looping
Speex may decide not to consume any samples because it can't write any. I've
seen a hang during draining caused by the resample loop never terminating.
In that case, resampling happened as normal until olen was 0 but ilen was
still 1. _process_native then reduced ichunk to 0, so ilen never decreased
below 1 and the loop never terminated.
Instead of reverting 684cf44 ({audioresample: don't skip input samples),
break only if all output samples have been produced and speex refuses
to consume any more input samples.
https://bugzilla.gnome.org/show_bug.cgi?id=732908
Sebastian Dröge [Mon, 19 Jan 2015 10:17:18 +0000 (11:17 +0100)]
videorate: Add $(GST_PLUGINS_BASE_CFLAGS) to be able to find gst/video/video.h
Nicolas Dufresne [Sun, 18 Jan 2015 19:58:36 +0000 (14:58 -0500)]
videorate: Implement allocation query
The videorate element keeps 1 buffer internally. This buffer need
to be requested during allocation query otherwise the pipeline may
stall.
https://bugzilla.gnome.org/show_bug.cgi?id=738302
Nicolas Dufresne [Sun, 18 Jan 2015 19:17:07 +0000 (14:17 -0500)]
Revert "videorate: Implement allocation query"
This reverts commit
3c04db4a307048db70ee1d08c1d62e26ad9569d8.
Nicolas Dufresne [Sun, 18 Jan 2015 16:02:00 +0000 (11:02 -0500)]
videorate: Implement allocation query
VideRate keeps 1 buffer in order to duplicate base on closest buffer
relative to targeted time. This extra buffer need to be request
otherwise the pipeline may stall when fixed size buffer pool is used.
https://bugzilla.gnome.org/show_bug.cgi?id=738302
Sebastian Dröge [Sat, 17 Jan 2015 13:51:48 +0000 (14:51 +0100)]
decodebin: Fix compilation
Branislav Katreniak [Mon, 12 Jan 2015 13:38:09 +0000 (14:38 +0100)]
decodebin: do call set_queue_size in no_more_pads_cb
Consider pipeline: gst-launch-1.0 playbin uri=http://example.com/a.ogg
Consider 128kbit audio stream.
As soon as uridecodebin detects the bitrate, it configures its input
queue2 max-size to 32000 bytes.
The 2MB buffer in multiqueue is nearly 2 orders of magnitude bigger.
This non-deterministically drives queue2 buffer anywhere from
100% to 0% until multiqueue is filled.
This patch sets multiqueue size to 5 buffers early in no_more_pads_cb.
Partly reverts commit
db771185ed750627a6a1824c42b651d739e1b4a4.
https://bugzilla.gnome.org/show_bug.cgi?id=740689
Vincent Penquerc'h [Fri, 16 Jan 2015 15:21:14 +0000 (15:21 +0000)]
decodebin: free old groups when switching groups
Old groups are freed with one switch's delay when switching groups.
They're freed in a scratch thread to avoid delaying the switch.
Vincent Penquerc'h [Fri, 12 Dec 2014 17:02:35 +0000 (17:02 +0000)]
oggmux: fix clipped duration determination for non 0 based segments
https://bugzilla.gnome.org/show_bug.cgi?id=740422
Sebastian Dröge [Thu, 15 Jan 2015 09:51:37 +0000 (10:51 +0100)]
audio: Keep caps features when building the downstream filter
Based on
5fd4e3e0b6cc4f30d7b1489a105db946b43f1a9f for video
by Alessandro Decina.
Alessandro Decina [Thu, 15 Jan 2015 02:54:14 +0000 (13:54 +1100)]
videoutils: keep caps features in account when building the downstream filter
See 00c2ce6 and https://bugzilla.gnome.org/show_bug.cgi?id=741263 for reference.
Thiago Santos [Wed, 14 Jan 2015 13:35:34 +0000 (10:35 -0300)]
examples: playback: add labels with supported seek range
Add the supported seeking range in the advanced seek area.
Also implement seeking querying the pipeline to retrieve those
values and show to the user. It is done in a smaller frequency
compared to the position/duration querying.
Thiago Santos [Tue, 13 Jan 2015 22:25:52 +0000 (19:25 -0300)]
decodebin: disable pad link checks as it has already been done
Decodebin has already added the element to the bin and should only
select caps compatible pads. It should disable the pad link checks
to avoid doing those again.
https://bugzilla.gnome.org/show_bug.cgi?id=742885
Luis de Bethencourt [Tue, 13 Jan 2015 16:58:34 +0000 (16:58 +0000)]
visual: cleanup
Shameful fix to a silly mistake in the previous commit. Above email address for
any mockery
Luis de Bethencourt [Tue, 13 Jan 2015 16:36:09 +0000 (16:36 +0000)]
visual: handle the return of the setup function
Make the baseclass future proof by handling the gboolean return of the setup
function. So if/when a child class uses this the base class is ready.
Luis de Bethencourt [Tue, 13 Jan 2015 16:09:49 +0000 (16:09 +0000)]
Revert "visual: remove unnecessary variable"
This reverts commit
a91d521a3602f33083405467db9454d422b9da1b.
Being a base class it is better to check the value instead of ignoring it since
a child class could be created that returns valuable information.
Luis de Bethencourt [Tue, 13 Jan 2015 15:07:56 +0000 (15:07 +0000)]
visual: remove unnecessary variable
klass->setup (scope) will always return TRUE since all children of this class
do so, no need to store the return. Besides, the value is overwritten a few
lines down before it is ever used. Save the unnecessary memory and instructions.
CID #1226467
Luis de Bethencourt [Mon, 12 Jan 2015 15:27:18 +0000 (15:27 +0000)]
visual: use unused value
ret is assigned but not used and in the next cycle of the loop it is overwritten
with default_prepare_output_buffer (). If there is a flow error the function
should return instead.
CID #1226475
Stefan Sauer [Mon, 12 Jan 2015 14:56:06 +0000 (15:56 +0100)]
Automatic update of common submodule
From f2c6b95 to bc76a8b
Mark Nauwelaerts [Thu, 8 Jan 2015 20:20:14 +0000 (21:20 +0100)]
audioringbuffer: start ringbuffer if needed upon commit
... to provide for a running clock.
Mark Nauwelaerts [Fri, 2 Jan 2015 13:34:41 +0000 (14:34 +0100)]
videoencoder: fix comment typo
Luis de Bethencourt [Fri, 9 Jan 2015 15:38:09 +0000 (15:38 +0000)]
video-dither: remove check for below zero for unsigned value
CLAMP checks both if value is '< 0' and '> max'. Value will never be a negative
number since it is an unsigned integer. Removing that check and only checking if
it is bigger than max and setting it appropriately.
CID 1256559
Luis de Bethencourt [Fri, 9 Jan 2015 15:28:06 +0000 (15:28 +0000)]
video-resampler: remove check for below zero for unsigned value
CLAMP checks both if n_taps is '< 0' and '> max_taps'. n_taps will never be a
negative number because it is an unsigned integer. Removing that check and only
making sure it isn't set bigger than max.
CID 1256558
Sebastian Dröge [Thu, 8 Jan 2015 09:45:46 +0000 (10:45 +0100)]
video: Add support for BT2020 colorspace (UHD)
Wim Taymans [Wed, 7 Jan 2015 14:54:58 +0000 (15:54 +0100)]
video-scaler: remove useless debug
Wim Taymans [Wed, 7 Jan 2015 14:52:57 +0000 (15:52 +0100)]
video-converter: add options to control chroma resampling
Add an option to disable chroma resampling.
Improve the matrix option values so that you can choose to use the input
or output matrix or disable conversion.
Tim-Philipp Müller [Fri, 2 Jan 2015 15:27:23 +0000 (15:27 +0000)]
oggmux: remove unused enum
Tim-Philipp Müller [Wed, 31 Dec 2014 19:40:20 +0000 (19:40 +0000)]
oggmux: fix silly GQueue iteration code
Sam Thursfield [Fri, 26 Dec 2014 20:48:55 +0000 (20:48 +0000)]
Fix documentation that incorrectly says a return value should be freed
The gst_discoverer_info_get_missing_elements_installer_details()
documentation and annotation says that the return value should be freed
with g_strfreev(), but actually it's owned by the GstDiscovereInfo
object and should definitely not get freed by the caller as well.
https://bugzilla.gnome.org/show_bug.cgi?id=742006
Nirbheek Chauhan [Sat, 27 Dec 2014 09:14:51 +0000 (14:44 +0530)]
audiobasesrc: Explicitly document that buffer-time and latency-time may be ignored
Thiago Santos [Fri, 26 Dec 2014 21:55:08 +0000 (18:55 -0300)]
oggmux: only clip by duration if end of buffer is ahead of segment
It might happen that the timestamp is before the segment and the
check would succeed. In this case reducing the duration makes no
sense and would lead to broken results.
Sebastian Dröge [Mon, 22 Dec 2014 21:04:41 +0000 (22:04 +0100)]
videotestsrc: Report our latency properly in live mode
While we have no latency at all in theory, any other live source has the
duration of one buffer as minimum latency. Do the same in videotestsrc.
https://bugzilla.gnome.org/show_bug.cgi?id=741879
Sebastian Dröge [Mon, 22 Dec 2014 21:00:26 +0000 (22:00 +0100)]
audiotestsrc: Report our latency properly in live mode
While we have no latency at all in theory, any other live source has the
duration of one buffer as minimum latency. Do the same in audiotestsrc.
https://bugzilla.gnome.org/show_bug.cgi?id=741879
Song Bing [Mon, 22 Dec 2014 14:25:04 +0000 (09:25 -0500)]
videopool: update video alignment after applying
Video buffer pool will update video alignment to respect stride alignment
requirement. But haven't updated it to video alignment in configure.
Which will cause user get wrong video alignment.
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=741501
Thiago Santos [Fri, 28 Nov 2014 17:36:23 +0000 (14:36 -0300)]
audiobasesink: get the internal time before the clock reset
Otherwise calls to get the clock time might change its internal state
and the internal/external time for calibration get unbalanced leading to
a clock jump
https://bugzilla.gnome.org/show_bug.cgi?id=740834
Sebastian Dröge [Mon, 22 Dec 2014 10:45:53 +0000 (11:45 +0100)]
MAINTAINERS: Update my mail address
Sebastian Dröge [Mon, 22 Dec 2014 10:38:20 +0000 (11:38 +0100)]
video{en,de}coder: Call reset() before the start() vfunc
This makes sure that the element is in the same state before start() is called
the very first time and every future call after the element was used already.
Also it ensure that we always have a clean state before start(), cleaned the
same way in every case.
Sebastian Dröge [Mon, 22 Dec 2014 10:36:58 +0000 (11:36 +0100)]
audioencoder: Call reset() before the start() vfunc to guarantee a clean state
The same was done already in the decoder, and we cleaned some state just above
manually that would also be taken care of by reset().
This makes sure that the element is in the same state before start() is called
the very first time and every future call after the element was used already.
Sebastian Dröge [Mon, 22 Dec 2014 10:33:14 +0000 (11:33 +0100)]
video{en,de}coder: Reset the codec after calling the stop() vfunc
The stop() vfunc might mess with some of our fields we have just
reset, which could cause memory leaks or invalid state taken over
to later.
Also the stop() vfunc, or anything called until it from another thread,
might want to be able to use the fields that were just resetted and
become confused because of that.
In the decoder we already had a workaround for things like this happening,
this workaround is not needed anymore.
Sebastian Dröge [Mon, 22 Dec 2014 09:45:37 +0000 (10:45 +0100)]
audiobase{sink,src}: Don't hold the object lock while calling create_ringbuffer() vfunc
The implementation of that vfunc might want to use the object lock for
something too. It's generally not a good idea to keep the object lock while
calling any function implemented elsewhere.
Also the ringbuffer can only be NULL at this point, remove a useless if block.
And in the sink actually hold the object lock while setting the ringbuffer on
the instance. Code accessing this is expected to use the object lock, so do it
here ourselves too.
Sebastian Dröge [Thu, 18 Dec 2014 12:24:22 +0000 (13:24 +0100)]
riff-media: Error out early if we observe an invalid audio format
Sebastian Dröge [Thu, 18 Dec 2014 12:22:17 +0000 (13:22 +0100)]
riff: Also handle invalid block aligns for raw audio
Fixes audio playback of
http://demo.archermind.com/Test%20Sample/Video/MPEG%204/Divx3/Low-Motion/576-320.avi
Audio and video together is still broken because of other issues.
Edward Hervey [Thu, 18 Dec 2014 09:57:13 +0000 (10:57 +0100)]
audio: Fix private header include/dist
We want to dist it, but we don't want to install it.
Fixes make dist/distcheck
Sebastian Dröge [Thu, 18 Dec 2014 09:53:20 +0000 (10:53 +0100)]
Automatic update of common submodule
From ef1ffdc to f2c6b95
Thiago Santos [Wed, 17 Dec 2014 22:14:38 +0000 (19:14 -0300)]
video: audio: fix GI annotations for proxy caps function
Add the annotations to parameters that can be null and also for stating
the ownership of the returned caps
Thiago Santos [Wed, 17 Dec 2014 18:21:48 +0000 (15:21 -0300)]
tests: audiodecoder: tests for caps query implementation
Copied from videodecoder tests and updated to audio features
Thiago Santos [Wed, 17 Dec 2014 18:21:16 +0000 (15:21 -0300)]
audiodecoder: expose getcaps virtual function
Allows subclasses to do custom caps query replies.
Also exposes the standard caps query handler so subclasses can just
extend on top of it instead of reimplementing the caps query proxying.
Thiago Santos [Tue, 16 Dec 2014 21:36:57 +0000 (18:36 -0300)]
audiodecoder: implement caps and accept-caps queries
Allows decoders to proxy downstream restrictions on caps.
Also implements accept-caps query to prevent regressions caused by the
new fields on the return of a caps query that would cause the accept-caps
to fail as it uses subset caps comparisons
Thiago Santos [Tue, 16 Dec 2014 14:13:40 +0000 (11:13 -0300)]
audioencoder: refactor getcaps proxy function to be reusable
Makes the audioencoder's getcaps function that proxies downstream
restriction available to other elements in the audio module to use it
Thiago Santos [Wed, 17 Dec 2014 17:18:03 +0000 (14:18 -0300)]
videodecoder: expose getcaps virtual function
Allows subclasses to do custom caps query replies.
Also exposes the standard caps query handler so subclasses can just
extend on top of it instead of reimplementing the caps query proxying.
https://bugzilla.gnome.org/show_bug.cgi?id=741263
Thiago Santos [Mon, 15 Dec 2014 21:46:21 +0000 (18:46 -0300)]
videodecoder: accept-caps should only require fields from the template
With the new caps query results the caps returned might have extra fields
that are not required by the decoder (framerate for image decoders) and it
causes a regression making, for example, jpegdec reject caps that don't
have framerates.
The accept-caps implementation will do 2 checks:
1) Do subset check with the template caps, making sure all the required
fields that are present on the template are present on the received caps.
2) Do a intersection check with the result of a caps query, making sure
that downstream can accept the fields in the received caps.
https://bugzilla.gnome.org/show_bug.cgi?id=741263
Thiago Santos [Tue, 9 Dec 2014 19:08:12 +0000 (16:08 -0300)]
videoutils: proxy filter when doing a caps query downstream
Allows downstream to use the filter and possibly reduce caps complexity
to speed up negotiation
https://bugzilla.gnome.org/show_bug.cgi?id=741263
Thiago Santos [Tue, 9 Dec 2014 19:05:27 +0000 (16:05 -0300)]
videoutils: return empty if the element has no possible allowed caps
Instead of returning the template caps and having a failure happen
later because there are no possible caps
https://bugzilla.gnome.org/show_bug.cgi?id=741263
Thiago Santos [Mon, 8 Dec 2014 19:33:33 +0000 (16:33 -0300)]
videodecoder: implement caps query
Refactor the encoder's caps query proxying function to a common place
and use it in the videodecoder to proxy downstream restrictions.
The new function is private to the gstvideo lib.
https://bugzilla.gnome.org/show_bug.cgi?id=741263
Tim-Philipp Müller [Wed, 17 Dec 2014 12:01:19 +0000 (12:01 +0000)]
configure: require release version of orc now that there is one
Wim Taymans [Tue, 16 Dec 2014 11:57:55 +0000 (12:57 +0100)]
ximagesink: clear src and dest rectangles
Now that the center function also takes into account the x and y
coordinates of the dest rectangle, better clear all the fields before
using them.
Song Bing [Tue, 16 Dec 2014 11:10:53 +0000 (12:10 +0100)]
videopool: update buffer size after video alignment
Update the new buffer size after alignment in the pool configuration
before calling the parent set_config. This ensures that the parent knows
about the buffer size that we will allocate and makes the size check
work in the release_buffer method.
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=741420
Sebastian Dröge [Mon, 15 Dec 2014 19:57:14 +0000 (20:57 +0100)]
audiobasesrc/sink: Add _CAST macros
Edward Hervey [Mon, 15 Dec 2014 13:10:17 +0000 (14:10 +0100)]
video: Fix non-default usage of gst_video_sink_center_rect
Make sure we take into account non-0 x/y destination rectangles
Tim-Philipp Müller [Mon, 15 Dec 2014 12:12:44 +0000 (12:12 +0000)]
examples: improve playback-test help text a little
And allow pipeline type to be specified as string.
Sebastian Dröge [Mon, 15 Dec 2014 09:35:35 +0000 (10:35 +0100)]
pango: Add license/copyright header to header file
Sebastian Dröge [Mon, 15 Dec 2014 08:45:43 +0000 (09:45 +0100)]
Revert "decodebin: Only emit the drain signal for the main decode chain, not any subchains"
This reverts commit
a391dfe17f1a325f60e1d51a6d40c1a68eb196de.
It breaks gapless playback: https://bugzilla.gnome.org/show_bug.cgi?id=740045
Matej Knopp [Tue, 9 Dec 2014 02:18:37 +0000 (03:18 +0100)]
audiorate: Fill gap events
https://bugzilla.gnome.org/show_bug.cgi?id=741281
Sanjay NM [Wed, 10 Dec 2014 10:40:58 +0000 (16:10 +0530)]
audio: Add error handling to gst_audio_decoder_drain()
https://bugzilla.gnome.org/show_bug.cgi?id=740686
Sebastian Dröge [Sat, 13 Dec 2014 15:14:49 +0000 (16:14 +0100)]
audioclock: Fix redundant definitions compiler warning
gstaudioclock.c:51:31: error: redundant redeclaration of 'gst_audio_clock_init' [-Werror=redundant-decls]
G_DEFINE_TYPE (GstAudioClock, gst_audio_clock, GST_TYPE_SYSTEM_CLOCK);
gstaudioclock.c:51:31: error: redundant redeclaration of 'gst_audio_clock_class_init' [-Werror=redundant-decls]
G_DEFINE_TYPE (GstAudioClock, gst_audio_clock, GST_TYPE_SYSTEM_CLOCK);
Sebastian Dröge [Sat, 13 Dec 2014 15:04:40 +0000 (16:04 +0100)]
audioclock: No need to get the parent class in class_init, G_DEFINE_TYPE does that for us
Sebastian Dröge [Sat, 13 Dec 2014 15:01:44 +0000 (16:01 +0100)]
audioclock: Use G_DEFINE_TYPE instead of a custom get_type() function
Zaheer Abbas Merali [Fri, 12 Dec 2014 16:32:15 +0000 (08:32 -0800)]
rtcpbuffer: fix spelling of word in comment
Tim-Philipp Müller [Fri, 12 Dec 2014 14:59:49 +0000 (14:59 +0000)]
tests: rtpbasepayload: fix indentation
Tim-Philipp Müller [Fri, 12 Dec 2014 14:59:03 +0000 (14:59 +0000)]
tests: audiodecoder: fix indentation
Tim-Philipp Müller [Fri, 12 Dec 2014 14:56:36 +0000 (14:56 +0000)]
tests: audiodecoder: fix broken refcounting in unit test
The set_format vfunc does not pass ownership of the caps
to the decoder, so we mustn't unref the caps there.
gst_event_new_caps() does not take ownership of the caps
passed, so we must unref the caps afterwards.
Fixes leaks when running test in valgrind in 1.4 branch.
Sebastian Dröge [Fri, 12 Dec 2014 09:02:43 +0000 (10:02 +0100)]
video: Update disted orc source files
Sebastian Dröge [Fri, 12 Dec 2014 09:01:36 +0000 (10:01 +0100)]
Revert "video-converter: Fix compiler warning because of missing prototype of non-static function"
This reverts commit
406f32a9468c837a4d71f988de10dc2198a8edc9.
The problem was apparently that my video-orc.h was not updated and did not
include the prototype for that function. Only a "make clean" caused it to
be regenerated.
Sebastian Dröge [Fri, 12 Dec 2014 08:51:05 +0000 (09:51 +0100)]
video-converter: Fix compiler warning because of missing prototype of non-static function
video-converter.c:838:1: error: no previous prototype for function
'_custom_video_orc_matrix8' [-Werror,-Wmissing-prototypes]
Thiago Santos [Wed, 10 Dec 2014 01:47:31 +0000 (22:47 -0300)]
audiodecoder: do not use fixed caps on source pad
decoders can change the caps on their source pads, so they don't
use fixed caps. Having fixed caps can cause renegotiation issues.
Thiago Santos [Wed, 10 Dec 2014 01:46:42 +0000 (22:46 -0300)]
videodecoder: do not use fixed caps on source pad
decoders can change the caps on their source pads, so they don't
use fixed caps. Having fixed caps can cause renegotiation issues.
Thibault Saunier [Thu, 11 Dec 2014 12:45:38 +0000 (13:45 +0100)]
playbin: Do not mix up stream type when getting stream combiner element
We were always returning the video stream combiner whatever stream type
combiner was wanted.
Thiago Santos [Wed, 10 Dec 2014 16:23:23 +0000 (13:23 -0300)]
playbin2: always unref the combiner sinkpad when removing the srcpad
Create a function to do the pad cleanup of the GstSourceCombine struct
and use it to not forget to also cleanup the sink pad and fix a memory
leak.
https://bugzilla.gnome.org/show_bug.cgi?id=741198
Wim Taymans [Wed, 10 Dec 2014 15:42:12 +0000 (16:42 +0100)]
video-orc: make RGB pack/unpack faster
Avoid all the merging and splitting and use a pair of shifts and or
Jan Schmidt [Wed, 10 Dec 2014 14:53:15 +0000 (01:53 +1100)]
videodecoder: Add GST_VIDEO_DECODER_CAST macro
It's used in some macros already, so let's make it exist.
Göran Jönsson [Tue, 25 Nov 2014 12:31:48 +0000 (13:31 +0100)]
rtspconnection: No remove child if destroyed.
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=740730
Jan Schmidt [Mon, 8 Dec 2014 07:53:35 +0000 (18:53 +1100)]
reverse-play: fix seek to end when starting reverse
Start reverse playback by actually seeking to the end of
the file.
Wim Taymans [Sat, 6 Dec 2014 20:02:37 +0000 (21:02 +0100)]
video-converter: set bits and format after conversion
Update the current format, bits and pstride.
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=741187
Thiago Santos [Sat, 6 Dec 2014 01:09:45 +0000 (22:09 -0300)]
video-converter: free dither_lines
Avoid a memory leak
Nicolas Dufresne [Fri, 5 Dec 2014 23:16:53 +0000 (18:16 -0500)]
Bump ORC requirement to 4.22.1
We now depend on git commit f1cfa5, "orcc: allow setting custom
backup function"
Wim Taymans [Fri, 5 Dec 2014 13:51:28 +0000 (14:51 +0100)]
video-converter: use custom backup function
Use the new orc feature to set a custom backup function.
Wim Taymans [Fri, 5 Dec 2014 11:18:42 +0000 (12:18 +0100)]
video-converter: improve matrix8 function
Avoid using a constant.
Avoid doing saturated adds, results are not supposed to overflow here.
Rework the C backup function a little in preparation for custom backup
functions in ORC.
See https://bugzilla.gnome.org/show_bug.cgi?id=741015
Mathieu Duponchelle [Fri, 28 Nov 2014 14:06:27 +0000 (15:06 +0100)]
audiodecoder: Push pending events before sending EOS.
Segments are added to the pending events, and pushing a segment
is mandatory before sending EOS.
+ Adds a test.
https://bugzilla.gnome.org/show_bug.cgi?id=740853
Mathieu Duponchelle [Thu, 27 Nov 2014 04:53:20 +0000 (05:53 +0100)]
oggdemux: Fix seeking before the first frame.
The previous code was setting keytarget to target
to make sure the keyframe found for each pad was
indeed before the target.
Then if target == keytarget, it assumed a keyframe had been
found, which was not the case if target was before the first frame
in the file.
This patch checks that a keyframe was indeed found, and if not
seeks to 0, without bisecting again.
Assuming default gst qa assets in $HOME/gst-validate
seek_before_first_frame.scenario:
description, seek=true, handles-states=true
pause, playback-time=0.0
seek, playback-time=0.0, start=0.0, flags=accurate+flush
seek, playback-time=0.0, start=0.01, flags=accurate+flush
seek, playback-time=0.0, start=0.1, flags=accurate+flush
GST_DEBUG=*theoradec*:2 gst-validate-1.0 playbin \
uri=file://$HOME/gst-validate/gst-qa-assets/medias/ogg/vorbis_theora.0.ogg \
--set-scenario seek_before_first_frame.scenario
https://bugzilla.gnome.org/show_bug.cgi?id=741097
Edward Hervey [Wed, 8 Oct 2014 06:54:57 +0000 (08:54 +0200)]
playbin: Only check sinks which are in >= GST_STATE_READY
Otherwise we endup with bogus caps intersection (from the pad template
caps and not from what the actual hardware/device supports)
https://bugzilla.gnome.org/show_bug.cgi?id=738131
Wim Taymans [Wed, 3 Dec 2014 09:15:18 +0000 (10:15 +0100)]
video-converter: fix chroma resampling check
Decide if we need chroma resampling by checking if we have a progressive
or interlaced chroma resampler.
Wim Taymans [Wed, 3 Dec 2014 09:14:34 +0000 (10:14 +0100)]
video-converter: only do dithering when needed
Only do dithering when one of the quantizers is > 1.
Chad [Tue, 2 Dec 2014 20:58:00 +0000 (15:58 -0500)]
audiorate: Use gst_util_uint64_scale_int_round()
Using gst_util_uint64_scale_int() causes slight drift
which accumulates over time.
https://bugzilla.gnome.org/show_bug.cgi?id=741045
Wim Taymans [Tue, 2 Dec 2014 12:39:52 +0000 (13:39 +0100)]
defs: update defs file
Wim Taymans [Tue, 2 Dec 2014 10:51:19 +0000 (11:51 +0100)]
videoconvert: add dither-bits option
Fix the dither option.
Add a new option to set the quantizer
Wim Taymans [Tue, 2 Dec 2014 10:48:11 +0000 (11:48 +0100)]
video-scaler: add where orc functions could go
Add the disabled orc functions in #if 0 lines for when we can enable
them.
Wim Taymans [Tue, 2 Dec 2014 10:40:59 +0000 (11:40 +0100)]
video-converter: add dithering
Use the new dither object to perform dithering.
Add option to select dithering method.
Add option to quantize to a specific value
Wim Taymans [Tue, 2 Dec 2014 10:39:42 +0000 (11:39 +0100)]
video-converter: add palette when needed
Wim Taymans [Tue, 2 Dec 2014 10:32:28 +0000 (11:32 +0100)]
video-dither: add video dither helper object
Add a new object that implements various dithering methods.