platform/upstream/gstreamer.git
7 years agosouphttpsrc: fix svace issue
Eunhae Choi [Mon, 19 Dec 2016 04:28:12 +0000 (13:28 +0900)]
souphttpsrc: fix svace issue

Change-Id: Id303ad49d9b536a4b7a348a3969208bac9e33601
Signed-off-by: Eunhae Choi <eunhae1.choi@samsung.com>
7 years agosouphttpsrc: add cookie jar
Eunhae Choi [Mon, 12 Dec 2016 12:06:21 +0000 (21:06 +0900)]
souphttpsrc: add cookie jar

Change-Id: I1912c1fa3903723f3487d0ec65daa57269b89519

7 years agoMerge "rtsp: to support player api spec for rtsp kind of buffering" into tizen
Eunhae Choi [Mon, 12 Dec 2016 08:37:56 +0000 (00:37 -0800)]
Merge "rtsp: to support player api spec for rtsp kind of buffering" into tizen

7 years agortsp: to support player api spec for rtsp kind of buffering
Eunhae Choi [Fri, 9 Dec 2016 13:09:05 +0000 (22:09 +0900)]
rtsp: to support player api spec for rtsp kind of buffering

Change-Id: I33b42bd9da47e1041c429e62e1ac4665181756ac
Signed-off-by: Eunhae Choi <eunhae1.choi@samsung.com>
7 years ago[pulsesink/monoscope] fix memory leak
Eunhae Choi [Fri, 9 Dec 2016 04:48:53 +0000 (13:48 +0900)]
[pulsesink/monoscope] fix memory leak

Change-Id: Ib3491bf57df811967ce8d0e17375bb142f6854e2

7 years agoMerge branch 'tizen_3.0' into tizen
Eunhae Choi [Mon, 5 Dec 2016 12:50:37 +0000 (21:50 +0900)]
Merge branch 'tizen_3.0' into tizen

Change-Id: If2394fb67f6a2819df532ebaee14ba6f7840a09f

7 years agoaacparse: Fix svace issue(DIVISION_BY_ZERO)
Gilbok Lee [Tue, 29 Nov 2016 04:39:52 +0000 (13:39 +0900)]
aacparse: Fix svace issue(DIVISION_BY_ZERO)

Change-Id: Iac08af5804aa15c1b65f3122028392ace2b89e64

7 years agoMerge branch 'tizen' into tizen_3.0
Jeongmo Yang [Thu, 17 Nov 2016 06:39:25 +0000 (15:39 +0900)]
Merge branch 'tizen' into tizen_3.0

Change-Id: Iaaedd232419bc8343918081ccf6c72cc2c143fba

7 years agoUpdate expected-trailer-size related function
Jeongmo Yang [Thu, 17 Nov 2016 02:42:49 +0000 (11:42 +0900)]
Update expected-trailer-size related function

[Version] 1.6.1-14
[Profile] Common
[Issue Type] Update
[Dependency module] N/A
[Test] [M(T) - Boot=(OK), sdb=(OK), Home=(OK), Touch=(OK), Version=tizen-3.0-mobile_20161115.1]

Change-Id: Iff0a1633fc636297a12325cf83ab3d9b395b1ab4
Signed-off-by: Jeongmo Yang <jm80.yang@samsung.com>
8 years agowaveformsink: fix mem leak of debug msg
Eunhae Choi [Fri, 21 Oct 2016 08:58:51 +0000 (17:58 +0900)]
waveformsink: fix mem leak of debug msg

Change-Id: I76a5e82b8813ecbf79e6930c124e0d7e9c3cb0c1
Signed-off-by: Eunhae Choi <eunhae1.choi@samsung.com>
8 years agowaveformsink: fix mem leak of debug msg
Eunhae Choi [Fri, 21 Oct 2016 08:58:51 +0000 (17:58 +0900)]
waveformsink: fix mem leak of debug msg

Change-Id: I76a5e82b8813ecbf79e6930c124e0d7e9c3cb0c1
Signed-off-by: Eunhae Choi <eunhae1.choi@samsung.com>
8 years agov4l2src: remove libgudev-1.0-dev build dependency
Gilbok Lee [Mon, 10 Oct 2016 07:03:40 +0000 (16:03 +0900)]
v4l2src: remove libgudev-1.0-dev build dependency

Change-Id: I380e0959c0f50a4856c51f8da9cffcd6225191a8

8 years agoapply tizen build option rule
Eunhae Choi [Mon, 5 Sep 2016 12:50:29 +0000 (21:50 +0900)]
apply tizen build option rule

Change-Id: Id68ecfd92557efb73073a7419611a16edf9f378a

8 years agoqtdemux: handling zero segment-duration edit list
Gilbok Lee [Thu, 11 Aug 2016 08:42:10 +0000 (17:42 +0900)]
qtdemux: handling zero segment-duration edit list

    Based on document ISO_IEC_14496-12, edit list box can have
    segment duration as zero. It does not imply that media_start equals to
    media_stop. But, it just indicates a sample which should be presented
    at the first. This patch derives segment duration using media_time
    and duration of file. And set derived duration to segment-duration.

    https://bugzilla.gnome.org/show_bug.cgi?id=760781

Change-Id: I965e79f23970153bcc7c3cd935a29a09cc963246

8 years agoMerge "Fix SVACE issue - 103419(NULL_AFTER_DEREF)" into tizen
Jeongmo Yang [Mon, 8 Aug 2016 10:47:18 +0000 (03:47 -0700)]
Merge "Fix SVACE issue - 103419(NULL_AFTER_DEREF)" into tizen

8 years agoFix Svace issue(Missed return statement)
Gilbok Lee [Mon, 8 Aug 2016 10:08:36 +0000 (19:08 +0900)]
Fix Svace issue(Missed return statement)

Change-Id: I2d779221857fc4d2be43d7a1eaeaa066eb7f2e1a

8 years agoFix SVACE issue - 103419(NULL_AFTER_DEREF)
Haesu Gwon [Mon, 8 Aug 2016 10:06:51 +0000 (19:06 +0900)]
Fix SVACE issue - 103419(NULL_AFTER_DEREF)

Change-Id: I36e834875f5aba0fd4aaa832e2cd8bd9a1cd949c
Signed-off-by: Haesu Gwon <haesu.gwon@samsung.com>
8 years agoenable multifile for hlssink
Eunhae Choi [Tue, 19 Jul 2016 07:52:54 +0000 (16:52 +0900)]
enable multifile for hlssink

Change-Id: I92a2bb42cec38848372ddb4365e3a359d2ac60d6

8 years agoflvdemux,qtdemux: post tag message during build pipeline
Eunhae Choi [Fri, 15 Jul 2016 11:34:22 +0000 (20:34 +0900)]
flvdemux,qtdemux: post tag message during build pipeline

Change-Id: I65813b85536524aef6da198448e26b4188e897b3

8 years agoMerge "Add property for expected trailer size" into tizen
Jeongmo Yang [Fri, 1 Jul 2016 08:35:10 +0000 (01:35 -0700)]
Merge "Add property for expected trailer size" into tizen

8 years agoAdd property for expected trailer size
Haesu Gwon [Wed, 29 Jun 2016 05:35:22 +0000 (14:35 +0900)]
Add property for expected trailer size

Change-Id: Ie42654106fbd42a8f64cfa565286cc67c363e6f2
Signed-off-by: Haesu Gwon <haesu.gwon@samsung.com>
8 years ago[pulse] cork when pop-timeout event received from pulseaudio
Seungbae Shin [Thu, 23 Jun 2016 11:23:31 +0000 (20:23 +0900)]
[pulse] cork when pop-timeout event received from pulseaudio

Change-Id: I3d75890f761bd96e3bf4b8b7cc300fe08514e8f5

8 years agodisable unused plugins with tv featuring
Eunhae Choi [Fri, 3 Jun 2016 03:46:54 +0000 (12:46 +0900)]
disable unused plugins with tv featuring

Change-Id: Id53a6445e6159fd4a3709493b2e18bd7b55ac7ac

8 years agoEnable libv4l2 for S/W color converting in v4l2src
Jeongmo Yang [Tue, 10 May 2016 06:26:53 +0000 (15:26 +0900)]
Enable libv4l2 for S/W color converting in v4l2src

Change-Id: Ic0ec265292ed28033c33d83e60e1203b0d784a2a
Signed-off-by: Jeongmo Yang <jm80.yang@samsung.com>
8 years agoModification the condition to disable v4l2 plugins - target-circle repository only
Jeongmo Yang [Wed, 6 Apr 2016 00:52:41 +0000 (09:52 +0900)]
Modification the condition to disable v4l2 plugins - target-circle repository only

Change-Id: I9a7e766da6a10c2d646f88dc74554bb518cf7539
Signed-off-by: Jeongmo Yang <jm80.yang@samsung.com>
8 years agoDisable v4l2 plugins on arm/aarch64 architecture
Jeongmo Yang [Tue, 22 Mar 2016 06:18:35 +0000 (15:18 +0900)]
Disable v4l2 plugins on arm/aarch64 architecture

Change-Id: I3192e40eeeaf1b6a0e27d17775f26cf61c002abf
Signed-off-by: Jeongmo Yang <jm80.yang@samsung.com>
8 years agoRevert "Remove libjpeg build dependency (uninstall jpeg enc/dec)"
Gilbok Lee [Wed, 16 Mar 2016 03:05:26 +0000 (20:05 -0700)]
Revert "Remove libjpeg build dependency (uninstall jpeg enc/dec)"

This reverts commit 69d818fe0922d59488043c09f28c7cedb815b9c1.

Change-Id: Iadccf13c7cf88b072da8dbb91cdb74117d2dbdf9

8 years agoRemove libjpeg build dependency (uninstall jpeg enc/dec)
Gilbok Lee [Tue, 2 Feb 2016 01:59:00 +0000 (10:59 +0900)]
Remove libjpeg build dependency (uninstall jpeg enc/dec)

Change-Id: I8374001fd98d766228438677701f58b0b3c5e1d5

8 years agoRemove libpng build dependency (uninstall png enc/dec)
Gilbok Lee [Fri, 29 Jan 2016 05:13:57 +0000 (14:13 +0900)]
Remove libpng build dependency (uninstall png enc/dec)

Change-Id: Ica157e8c4f6c7cd0238d385b27c4d3569a7cd49c
Signed-off-by: Gilbok Lee <gilbok.lee@samsung.com>
8 years agoRemove build dependecy (vpx, speex, flac)
Gilbok Lee [Wed, 30 Dec 2015 03:00:45 +0000 (12:00 +0900)]
Remove build dependecy (vpx, speex, flac)

Change-Id: Ib95db7bb17160c75aa67d46048ab598733e4ea05
Signed-off-by: Gilbok Lee <gilbok.lee@samsung.com>
8 years agoDidn't install libgsteffectv.so (fix build break)
Gilbok Lee [Wed, 30 Dec 2015 01:15:37 +0000 (10:15 +0900)]
Didn't install libgsteffectv.so (fix build break)

Change-Id: Ibde111e5a7fcd1703e98e45877a3c007079a533a
Signed-off-by: Gilbok Lee <gilbok.lee@samsung.com>
8 years agoDisable effectv plugins
Gilbok Lee [Tue, 29 Dec 2015 07:11:49 +0000 (16:11 +0900)]
Disable effectv plugins

Change-Id: Ic6626ca29eb609b82c1b660486c88755ed343e69
Signed-off-by: Gilbok Lee <gilbok.lee@samsung.com>
8 years agoMerge "Remove libv4l dependency" into tizen
Heechul Jeon [Tue, 29 Dec 2015 07:10:13 +0000 (23:10 -0800)]
Merge "Remove libv4l dependency" into tizen

8 years agoRemove libv4l dependency
Jeongmo Yang [Tue, 29 Dec 2015 05:57:46 +0000 (14:57 +0900)]
Remove libv4l dependency

Change-Id: Iba0673b6b90c03fbc638d8acce1d18a82e1722e7
Signed-off-by: Jeongmo Yang <jm80.yang@samsung.com>
8 years agoMerge "Disable unused plugins" into tizen
Heechul Jeon [Tue, 29 Dec 2015 05:23:57 +0000 (21:23 -0800)]
Merge "Disable unused plugins" into tizen

8 years agoaacparse: Fixed aac(adts) seek speed slow issue
Gilbok Lee [Mon, 28 Dec 2015 08:38:17 +0000 (17:38 +0900)]
aacparse: Fixed aac(adts) seek speed slow issue
Sync with tizen 2.4 fixed code (http://165.213.149.170/gerrit/#/c/71933)

Change-Id: Ia42f4501928f01d30a02db9a280082511a333282
Signed-off-by: Gilbok Lee <gilbok.lee@samsung.com>
8 years agompegaudioparse: Fixed http seek slow issue in case of souphttpsrc
Gilbok Lee [Mon, 28 Dec 2015 07:07:31 +0000 (16:07 +0900)]
mpegaudioparse: Fixed http seek slow issue in case of souphttpsrc
Sync with tizen 2.4 fixed code (http://165.213.149.170/gerrit/#/c/60290)

Change-Id: Icc1e1bafe4262dce4e1dc506cff45a60b264fbe1
Signed-off-by: Gilbok Lee <gilbok.lee@samsung.com>
8 years agowavparse: Modify GST_RIFF_TAG_acid tag's byte-align bug for appsrc push mode(streaming)
Gilbok Lee [Mon, 28 Dec 2015 06:03:22 +0000 (15:03 +0900)]
wavparse: Modify GST_RIFF_TAG_acid tag's byte-align bug for appsrc push mode(streaming)
Sync with tizen 2.4 fixed code (http://165.213.149.170/gerrit/#/c/34063)

Change-Id: Ic982954d517586a7473fdd48b58c449b83ae1c1e
Signed-off-by: Gilbok Lee <gilbok.lee@samsung.com>
8 years agoDisable unused plugins
Gilbok Lee [Thu, 24 Dec 2015 04:06:38 +0000 (13:06 +0900)]
Disable unused plugins

Change-Id: I21393502d1a012e024c052f7ce8fb31f30861626
Signed-off-by: Gilbok Lee <gilbok.lee@samsung.com>
8 years agoremove gdk_pixbuf
Eunhae Choi [Tue, 22 Dec 2015 08:12:47 +0000 (17:12 +0900)]
remove gdk_pixbuf

Change-Id: I41b6532c87e1251b5919514340e52c4af7ab45d6

8 years agoRelease version 1.6.1-2
Jeongmo Yang [Wed, 16 Dec 2015 06:18:07 +0000 (15:18 +0900)]
Release version 1.6.1-2

1. [v4l2] Fix prevent issue - Dereference before NULL check

Change-Id: I5581cd284892c8445a497a8d6f776d63f925f6e1
Signed-off-by: Jeongmo Yang <jm80.yang@samsung.com>
8 years agomodify spec file about build require
Eunhae Choi [Mon, 14 Dec 2015 08:27:37 +0000 (17:27 +0900)]
modify spec file about build require

Change-Id: I2f7f4f7117de82f71f10c91b78b0e6b7cc359189

8 years agoenable x for wearable profile
Eunhae Choi [Tue, 1 Dec 2015 05:47:38 +0000 (14:47 +0900)]
enable x for wearable profile

Change-Id: Ib8c9c713814a90663119f04138be55ed3745cf65

8 years agomodify spec to build and merge prev modification
Eunhae Choi [Thu, 26 Nov 2015 04:29:33 +0000 (13:29 +0900)]
modify spec to build and merge prev modification

Change-Id: I8cdc43e6a576f0f795f00fcb9f1a358782079fab

8 years agoMerge remote-tracking branch 'remotes/origin/upstream/1.6' into tizen
Eunhae Choi [Thu, 26 Nov 2015 04:20:18 +0000 (13:20 +0900)]
Merge remote-tracking branch 'remotes/origin/upstream/1.6' into tizen

Change-Id: I9cc7b5a7496dde5fd0f5ae4183be6d3c0b5f02f5

9 years agoflvmux: fix writing of creation time
Tim-Philipp Müller [Wed, 21 Oct 2015 10:16:01 +0000 (11:16 +0100)]
flvmux: fix writing of creation time

Don't write time as e.g. 11:9:42

9 years agoflvmux: GST_BUFFER_OFFSETs should be GST_BUFFER_OFFSET_NONE
Havard Graff [Tue, 22 Sep 2015 20:59:16 +0000 (22:59 +0200)]
flvmux: GST_BUFFER_OFFSETs should be GST_BUFFER_OFFSET_NONE

Or else flvdemux don't understand it

https://bugzilla.gnome.org/show_bug.cgi?id=754435

9 years agoflvmux: use time segment and copy timestamps when streamable
Havard Graff [Wed, 2 Sep 2015 08:44:59 +0000 (10:44 +0200)]
flvmux: use time segment and copy timestamps when streamable

Add a basic test using speex data to verify timestamping.

https://bugzilla.gnome.org/show_bug.cgi?id=754435

9 years agoRelease 1.6.1
Sebastian Dröge [Fri, 30 Oct 2015 14:40:55 +0000 (16:40 +0200)]
Release 1.6.1

9 years agoUpdate .po files
Sebastian Dröge [Fri, 30 Oct 2015 14:27:32 +0000 (16:27 +0200)]
Update .po files

9 years agopo: Update translations
Sebastian Dröge [Fri, 30 Oct 2015 12:28:04 +0000 (14:28 +0200)]
po: Update translations

9 years agov4l2: fix double-unref in the v4l2 device provider
Nirbheek Chauhan [Wed, 28 Oct 2015 18:17:30 +0000 (23:47 +0530)]
v4l2: fix double-unref in the v4l2 device provider

9 years agomatroskamux: don't drop JPEG frames that only have PTS but no DTS set
Nicola Murino [Tue, 27 Oct 2015 09:48:00 +0000 (10:48 +0100)]
matroskamux: don't drop JPEG frames that only have PTS but no DTS set

For the MS/VfW codec ids, we want to write DTS timestamps instead
of PTS because that's what everyone else seems to do (and it's also
how it is in AVI). So for those input formats we use the buffer DTS
instead of the PTS. However, if there's no DTS set but only the PTS
then just take the PTS instead of dropping the input buffer. This
is useful especially for I-frame only codecs like JPEG and huffyuv,
but should also be fine as fallback in general.

Fixes regression with input JPEG frames that only have PTS set on them.

https://bugzilla.gnome.org/show_bug.cgi?id=756967

9 years agoqtmux: Allow negotiating to S8 as a raw format but stop making it best choice
Thibault Saunier [Sun, 11 Oct 2015 11:06:26 +0000 (12:06 +0100)]
qtmux: Allow negotiating to S8 as a raw format but stop making it best choice

Negotiation to audio/x-raw,format=S8 was not possible because S8 does
not have a bit order so we ended up doing `if (!entry.fourcc) goto refuse_caps;`

https://bugzilla.gnome.org/show_bug.cgi?id=756387

9 years agoqtmux: Add prores support
Thibault Saunier [Sun, 11 Oct 2015 08:18:40 +0000 (09:18 +0100)]
qtmux: Add prores support

https://bugzilla.gnome.org/show_bug.cgi?id=756388

9 years agotests/check/splitmux: test that the release_pad vfunc of splitmuxsink actually releas...
George Kiagiadakis [Sat, 24 Oct 2015 21:57:38 +0000 (23:57 +0200)]
tests/check/splitmux: test that the release_pad vfunc of splitmuxsink actually releases pads

https://bugzilla.gnome.org/show_bug.cgi?id=753622

9 years agosplitmuxsink: do not destroy the multiqueue & muxer when going to NULL
George Kiagiadakis [Sat, 24 Oct 2015 21:57:29 +0000 (23:57 +0200)]
splitmuxsink: do not destroy the multiqueue & muxer when going to NULL

Instead, delay it until all request pads have been released. This is
because the release_pad() vfunc requires the multiqueue and muxer to
be there in order to release their request pads as well. If those
elements are destroyed earlier, release_pad() does not work, no
pads are released and some resources are leaked.

https://bugzilla.gnome.org/show_bug.cgi?id=753622

9 years agoscaletempo: F64 support is only available in GIT master at this point
Sebastian Dröge [Tue, 27 Oct 2015 15:32:22 +0000 (17:32 +0200)]
scaletempo: F64 support is only available in GIT master at this point

9 years agoscaletempo: Fix handling of rate < 0
Sebastian Dröge [Sat, 24 Oct 2015 14:14:07 +0000 (17:14 +0300)]
scaletempo: Fix handling of rate < 0

We have to reverse all samples in a buffer before processing them to properly
have continuous data from one buffer to another. As a result we will have a
negative applied rate and a rate of 1.0.

Also make sure that input buffers are correctly clipped to the segment,
otherwise our calculations are going to go wrong.

Also copy over the segment event's sequence number to the output segment while
we're at it.

https://bugzilla.gnome.org/show_bug.cgi?id=757033

9 years agocheck: Dist splitvideo0[012].ogg test files.
Jan Schmidt [Sun, 25 Oct 2015 13:41:28 +0000 (00:41 +1100)]
check: Dist splitvideo0[012].ogg test files.

9 years agortpj2kpay: update fragment offset
Thiago Santos [Tue, 13 Oct 2015 15:42:56 +0000 (12:42 -0300)]
rtpj2kpay: update fragment offset

It was always being set to 0, making the resulting stream broken
for the receiver

https://bugzilla.gnome.org/show_bug.cgi?id=756422

9 years agoqtmux: Don't unconditionally use strnlen()
Ryan Hendrickson [Mon, 19 Oct 2015 12:36:37 +0000 (15:36 +0300)]
qtmux: Don't unconditionally use strnlen()

It's not available on older OSX and we can as well use memchr() here.

https://bugzilla.gnome.org/show_bug.cgi?id=756154

9 years agompegaudioparse: Fix buffer memory leak during failures
Vineeth TM [Mon, 12 Oct 2015 01:48:23 +0000 (10:48 +0900)]
mpegaudioparse: Fix buffer memory leak during failures

mapped buffer is not being unmapped during failures

https://bugzilla.gnome.org/show_bug.cgi?id=756231

9 years agoauparse: Fix event memory leak
Vineeth TM [Mon, 19 Oct 2015 08:38:32 +0000 (17:38 +0900)]
auparse: Fix event memory leak

Free the event after being handled to prevent memory leak.

https://bugzilla.gnome.org/show_bug.cgi?id=756799

9 years agoqtdemux: fix caps leak
Guillaume Desmottes [Wed, 14 Oct 2015 10:03:15 +0000 (12:03 +0200)]
qtdemux: fix caps leak

If the QtDemuxStream are re-used they may already have caps which used
to be leaked.

Reproduced using the
validate.dash.playback.seek_forward.dash_exMPD_BIP_TC1 validate
scenario.

https://bugzilla.gnome.org/show_bug.cgi?id=756561

9 years agoqtdemux: Fix taglist memory leak
Vineeth TM [Wed, 14 Oct 2015 00:29:50 +0000 (09:29 +0900)]
qtdemux: Fix taglist memory leak

Free the stream and its sub items instead of just the stream

https://bugzilla.gnome.org/show_bug.cgi?id=756544

9 years agoqtmux: Fix date memory leak
Vineeth TM [Wed, 7 Oct 2015 08:14:57 +0000 (17:14 +0900)]
qtmux: Fix date memory leak

When getting date from taglist, the memory should be freed after
using it.

https://bugzilla.gnome.org/show_bug.cgi?id=756171

9 years agoqtmux: Fix sample memory leak
Vineeth TM [Mon, 5 Oct 2015 02:03:38 +0000 (11:03 +0900)]
qtmux: Fix sample memory leak

When getting sample from taglist, the memory should be freed after
using it.

https://bugzilla.gnome.org/show_bug.cgi?id=756068

9 years agocutter: Fix buffer leak
Vineeth TM [Mon, 5 Oct 2015 04:10:56 +0000 (13:10 +0900)]
cutter: Fix buffer leak

Buffer is added to the internal cache, and pushed only when accumulated
buffer duration crosses 200 ms. So when the chain ends, the buffer accumulated
is not freed. Freeing the cache when the state changes from PAUSED to READY.

https://bugzilla.gnome.org/show_bug.cgi?id=754212

9 years agosplitmuxsink: post messages when fragments are being opened and closed
George Kiagiadakis [Fri, 17 Apr 2015 12:25:43 +0000 (14:25 +0200)]
splitmuxsink: post messages when fragments are being opened and closed

This can be useful for applications that need to track the created fragments
(to log them in a recording database, for example)

https://bugzilla.gnome.org/show_bug.cgi?id=750108

9 years agosplitmuxsink: allow non-video streams to serve as reference
Ramiro Polla [Wed, 29 Apr 2015 17:23:28 +0000 (18:23 +0100)]
splitmuxsink: allow non-video streams to serve as reference

In the absence of a video stream, the first stream will be used as
reference.

https://bugzilla.gnome.org/show_bug.cgi?id=753617

9 years agosplitmuxsink: initialize mux_start_time properly
George Kiagiadakis [Wed, 22 Jul 2015 15:45:12 +0000 (17:45 +0200)]
splitmuxsink: initialize mux_start_time properly

mux_start_time refers to the running_time of the buffer
that goes first in the output file. Normally this time is
0, so this variable is initialized to 0 during the state
change to PAUSED.

However, when dealing with dynamic pipelines and starting
a recording while the pipeline has already run for a while,
the running_time of the first buffer is > 0 and this causes
a problem with detecting the end of the first file(s) when
splitting by duration, because the code will later compare
the threshold_time with (last buffer running_time - mux_start_time)
and will get it wrong until mux_start_time advances enough
to make this difference < threshold_time, creating empty files
in the meantime.

https://bugzilla.gnome.org/show_bug.cgi?id=753624

9 years agogdkpixbufsink: don't leak old pixel buffer when setting a new overlay
Tim-Philipp Müller [Tue, 29 Sep 2015 10:12:48 +0000 (11:12 +0100)]
gdkpixbufsink: don't leak old pixel buffer when setting a new overlay

https://bugzilla.gnome.org/show_bug.cgi?id=755773

9 years agoflacenc: avoid potential string overflow
Tim-Philipp Müller [Mon, 28 Sep 2015 19:25:22 +0000 (20:25 +0100)]
flacenc: avoid potential string overflow

We don't necessarily have full control over the input tags, so
it's possible that the ISRC tag contains a longer string than
expected, in which case we'd write over the end of the static-size
13 byte buffer that is FLAC__StreamMetadata_CueSheet_Track::isrc.
Make sure to only copy the ISRC if it's not too long, and make
sure the buffer we write to is always NUL-terminated by using
g_strlcpy().

CID 1324931.

9 years agomatroskademux: Remove leftover assertion from 0.10
Sebastian Dröge [Mon, 28 Sep 2015 16:03:51 +0000 (18:03 +0200)]
matroskademux: Remove leftover assertion from 0.10

We now allocate memory via GstAllocator and as such can handle arbitrary
alignments, not only <= G_MEM_ALIGN.

https://bugzilla.gnome.org/show_bug.cgi?id=755708

9 years agoaacparse: fix uninitialized variables in LOAS config reading
Guillaume Marquebielle [Fri, 25 Sep 2015 08:01:37 +0000 (10:01 +0200)]
aacparse: fix uninitialized variables in LOAS config reading

On reading LOAS config, flag v=1 and vA=1 combination can occur, leading to warning
"Spec says "TBD"...". Returning TRUE on this case while parameters 'sample_rate' and
'channels' are pointing to uninitialized values can end on setting random values as
rate and channels on src caps.

https://bugzilla.gnome.org/show_bug.cgi?id=755611

9 years agoRelease 1.6.0
Sebastian Dröge [Fri, 25 Sep 2015 21:15:55 +0000 (23:15 +0200)]
Release 1.6.0

9 years agoUpdate .po files
Sebastian Dröge [Fri, 25 Sep 2015 20:57:34 +0000 (22:57 +0200)]
Update .po files

9 years agosmptealpha: Do not set width/height before comparing with old values
Thibault Saunier [Fri, 25 Sep 2015 12:08:09 +0000 (14:08 +0200)]
smptealpha: Do not set width/height before comparing with old values

Otherwise we end up considering the values did not change and we wrongly
work with the old video format (which will lead to wrong
behaviour/segfaults).

https://bugzilla.gnome.org/show_bug.cgi?id=755621

9 years agoqtdemux: Accumulate segments for edit lists before activating the next segment
Sebastian Dröge [Wed, 23 Sep 2015 18:59:00 +0000 (20:59 +0200)]
qtdemux: Accumulate segments for edit lists before activating the next segment

eceb2ccc739092d964d78945e19c2ecedbd214e2 broke segment seeks by always
accumulating segments manually when activating a segment. This is only
needed when handling edit lists, not when activating a segment because of a
seek. Do the accumulation when switching edit list segments instead.

This fixes segment seeks again, while keeping edit lists playback working.

https://bugzilla.gnome.org/show_bug.cgi?id=755471

9 years agospectrum: send phase values in the GstMessage for Phase info
Vikram Fugro [Wed, 23 Sep 2015 12:13:51 +0000 (17:43 +0530)]
spectrum: send phase values in the GstMessage for Phase info

https://bugzilla.gnome.org/show_bug.cgi?id=755463

9 years agomatroska-mux: Don't output a warning on MONO multiview mode.
Jan Schmidt [Mon, 21 Sep 2015 14:46:01 +0000 (00:46 +1000)]
matroska-mux: Don't output a warning on MONO multiview mode.

9 years agortptheoradepay: Fix memory leaks
Sebastian Rasmussen [Sat, 19 Sep 2015 15:02:18 +0000 (17:02 +0200)]
rtptheoradepay: Fix memory leaks

The same memory leaks were fixed in identical fashion for
vorbisdepay in 06efeff5d979576a252e5dae57f46d6445b1df12 in 2009.

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=755277

9 years agortp{vorbis,theora}{pay,depay}: Cosmetic cleanup
Sebastian Rasmussen [Sat, 19 Sep 2015 15:04:07 +0000 (17:04 +0200)]
rtp{vorbis,theora}{pay,depay}: Cosmetic cleanup

* use g_list_free_full(), don't iterate elements maually when freeing
* call gst_rtp_*_pay_clear_packet(), don't duplicate its code
* use gst_buffer_unref() to clarify that it is buffers being released,
  instead of refering directly to gst_mini_object_unref()

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=755277

9 years agortp{vorbis,theora}pay: Store headers in the packet buffers lists, not a NULL buffer
Sebastian Dröge [Sat, 19 Sep 2015 16:44:22 +0000 (18:44 +0200)]
rtp{vorbis,theora}pay: Store headers in the packet buffers lists, not a NULL buffer

https://bugzilla.gnome.org/show_bug.cgi?id=755265

9 years agoRelease 1.5.91
Sebastian Dröge [Fri, 18 Sep 2015 17:33:13 +0000 (19:33 +0200)]
Release 1.5.91

9 years agoUpdate .po files
Sebastian Dröge [Fri, 18 Sep 2015 17:23:57 +0000 (19:23 +0200)]
Update .po files

9 years agopo: Update translations
Sebastian Dröge [Fri, 18 Sep 2015 09:50:31 +0000 (11:50 +0200)]
po: Update translations

9 years agoavidemux: Fix taglist leak
Eunhae Choi [Thu, 17 Sep 2015 01:50:01 +0000 (10:50 +0900)]
avidemux: Fix taglist leak

gst_tag_list_insert() does not take ownership of the inserted taglist.

https://bugzilla.gnome.org/show_bug.cgi?id=755138

9 years agoaacparse: Skip LOAS AAC until a valid config is seen.
Jan Schmidt [Tue, 15 Sep 2015 21:05:36 +0000 (07:05 +1000)]
aacparse: Skip LOAS AAC until a valid config is seen.

It's normal when dropping into the middle of a stream to
not always have the config available immediately, so skip LOAS
until a valid config is seen without either setting invalid
caps or erroring out.

https://bugzilla.gnome.org/show_bug.cgi?id=751386

9 years agortpjitterbuffer: reset just a bit more upon flush_stop
Mark Nauwelaerts [Sun, 13 Sep 2015 13:41:38 +0000 (15:41 +0200)]
rtpjitterbuffer: reset just a bit more upon flush_stop

9 years agortpjitterbuffer: remove dead struct member
Mark Nauwelaerts [Sun, 13 Sep 2015 13:40:09 +0000 (15:40 +0200)]
rtpjitterbuffer: remove dead struct member

9 years agomultiudpsink: fix GError memory leak when hostname resolution fails
Vineeth TM [Fri, 11 Sep 2015 08:09:28 +0000 (17:09 +0900)]
multiudpsink: fix GError memory leak when hostname resolution fails

https://bugzilla.gnome.org/show_bug.cgi?id=754869

9 years agomatroskamux: drop HEADER flag from output buffers
Thiago Santos [Thu, 10 Sep 2015 18:26:54 +0000 (15:26 -0300)]
matroskamux: drop HEADER flag from output buffers

Drop HEADER flag from output buffers if they are not indeed
headers.

Fixes resending of headers in tcp connection handling

https://bugzilla.gnome.org/show_bug.cgi?id=754768

9 years agomatroskamux: fix matroskamux ! matroskademux
Tim-Philipp Müller [Thu, 10 Sep 2015 15:00:50 +0000 (16:00 +0100)]
matroskamux: fix matroskamux ! matroskademux

Don't carry over DISCONT flags from the input buffers to the
output buffer, or the demuxer might reset its state when it
receives the first data buffer just after parsing the simple
block header, and then expect sane data to follow.
Fixes matroskamux ! demux erroring out.

https://bugzilla.gnome.org/show_bug.cgi?id=754768
https://bugzilla.gnome.org/show_bug.cgi?id=657805

9 years agortsp: fix small README typo
Martin Kelly [Wed, 9 Sep 2015 19:51:40 +0000 (12:51 -0700)]
rtsp: fix small README typo

https://bugzilla.gnome.org/show_bug.cgi?id=754807

9 years agowavpackparse: set both pts and dts so baseparse doesn't make up wrong dts after seeks
Tim-Philipp Müller [Fri, 4 Sep 2015 18:45:37 +0000 (19:45 +0100)]
wavpackparse: set both pts and dts so baseparse doesn't make up wrong dts after seeks

https://bugzilla.gnome.org/show_bug.cgi?id=752106

9 years agoflacparse: set both pts and dts so baseparse doesn't make up wrong dts after a seek
Tim-Philipp Müller [Fri, 4 Sep 2015 18:34:41 +0000 (19:34 +0100)]
flacparse: set both pts and dts so baseparse doesn't make up wrong dts after a seek

flac contains the sample offset in the frame header, so after a seek
without index flacparse will know the exact position we landed on and
timestamp buffers accordingly. It only set the pts though, which means
the baseparse-set dts which was set to the seek position prevails, and
since the seek was based on an estimate, there's likely a discrepancy
between where we wanted to land and where we did land, so from here on
that dts/pts difference will be maintained, with dts possibly multiple
seconds ahead of pts, which is just wrong. The easiest way to fix this
is to just set both pts and dts based on the sample offset, but perhaps
parsed audio should just not have dts set at all.

https://bugzilla.gnome.org/show_bug.cgi?id=752106

9 years agodocs: remove properties and signals that no longer exist
Tim-Philipp Müller [Sun, 6 Sep 2015 15:33:02 +0000 (16:33 +0100)]
docs: remove properties and signals that no longer exist

https://bugzilla.gnome.org/show_bug.cgi?id=726443

9 years agoflvmux: Make the element count in arrays not include end
George Chriss [Fri, 11 Oct 2013 15:13:00 +0000 (15:13 +0000)]
flvmux: Make the element count in arrays not include end

One-line removal of tags_written++

This should fix rtmp output to crtmpserver, and hopefully
noone is expecting that the element count includes the end
element, as different bits of documentation say different
things about whether it should or not.

https://bugzilla.gnome.org/show_bug.cgi?id=661624