platform/upstream/gstreamer.git
8 years agovideo-info: implement GstVideoInfo as boxed type
Marcin Kolny [Mon, 6 Jul 2015 09:36:58 +0000 (11:36 +0200)]
video-info: implement GstVideoInfo as boxed type

GstVideoInfo usually is created on the stack, but boxed type can be useful
for bindings.

https://bugzilla.gnome.org/show_bug.cgi?id=752011

8 years agortcpbuffer: Fix validation of packets with padding
Stian Selnes [Thu, 2 Jul 2015 18:50:00 +0000 (20:50 +0200)]
rtcpbuffer: Fix validation of packets with padding

The padding (if any) is included in the length of the last packet, see
RFC 3550.

Section 6.4.1:
   padding (P): 1 bit
      If the padding bit is set, this individual RTCP packet contains
      some additional padding octets at the end which are not part of
      the control information but are included in the length field. The
      last octet of the padding is a count of how many padding octets
      should be ignored, including itself (it will be a multiple of
      four).

Section A.2:
   *  The padding bit (P) should be zero for the first packet of a
      compound RTCP packet because padding should only be applied, if it
      is needed, to the last packet.

   *  The length fields of the individual RTCP packets must add up to
      the overall length of the compound RTCP packet as received.

https://bugzilla.gnome.org/show_bug.cgi?id=751883

8 years agovideodecoder: Fix setting default pixel-aspect-ratio
Stian Selnes [Wed, 1 Jul 2015 15:09:35 +0000 (17:09 +0200)]
videodecoder: Fix setting default pixel-aspect-ratio

It's needed to check if pixel-aspect-ratio exists before fixating.
It does not exist if input caps is not set yet and allowed caps
does not contain pixel-aspect-ratio (e.g. when using GST_VIDEO_CAPS_MAKE)

https://bugzilla.gnome.org/show_bug.cgi?id=751932

8 years agoAutomatic update of common submodule
Stefan Sauer [Fri, 3 Jul 2015 19:58:04 +0000 (21:58 +0200)]
Automatic update of common submodule

From f74b2df to 9aed1d7

8 years agodocs: order and canonicalize the -sections.txt file
Stefan Sauer [Fri, 3 Jul 2015 19:16:27 +0000 (21:16 +0200)]
docs: order and canonicalize the -sections.txt file

Have all sections in alphabetical order. Also make the macro order consistent.
This is a preparation for generating the file. Remove GET_CLASS macro for
some elements, since it is not used and the header is not installed.

8 years agocdparanoiasrc: remove unused defines
Stefan Sauer [Fri, 3 Jul 2015 19:09:29 +0000 (21:09 +0200)]
cdparanoiasrc: remove unused defines

8 years agovideoscale: fix debug categories
Stefan Sauer [Fri, 3 Jul 2015 19:08:03 +0000 (21:08 +0200)]
videoscale: fix debug categories

Use a local category for the default category and fix the import for the
performance category.

8 years agobasetextoverlay: Fix bug with unused upstream_has_meta
Nicolas Dufresne [Thu, 2 Jul 2015 14:47:45 +0000 (10:47 -0400)]
basetextoverlay: Fix bug with unused upstream_has_meta

The intention was to skip the allocation query if upstream has decided
to use the overlay meta feature in the caps. We can safely assume that
upstream have done that query already before making this decision. This
is an optimization since doing allocation queries is relatively
expensive.

CID #1308943

8 years agoRevert "basetextoverlay: remove dead code"
Nicolas Dufresne [Thu, 2 Jul 2015 14:27:39 +0000 (10:27 -0400)]
Revert "basetextoverlay: remove dead code"

This reverts commit e863e5f8a98ceec0ec0bd24274bbae8795e0ab75.

8 years agobasetextoverlay: remove dead code
Luis de Bethencourt [Thu, 2 Jul 2015 13:52:47 +0000 (14:52 +0100)]
basetextoverlay: remove dead code

upstream_has_meta is set to FALSE and never changed. The two checks for if
upstream_has_meta will never go to the true branch. Removing the boolean
and the true branches of these checks.

CID #1308943

8 years agoaudioencoder: Don't try to get buffers from an empty adapter
Sebastian Dröge [Thu, 2 Jul 2015 11:15:58 +0000 (13:15 +0200)]
audioencoder: Don't try to get buffers from an empty adapter

8 years ago{audio,video}{en,de}oder: Also copy POOL metas and make sure to copy over metas when...
Sebastian Dröge [Wed, 1 Jul 2015 08:58:07 +0000 (10:58 +0200)]
{audio,video}{en,de}oder: Also copy POOL metas and make sure to copy over metas when creating subbuffers

POOL meta just means that this specific instance of the meta is related to a
pool, a copy should be made when reasonable and the flag should just not be
set in the copy.

8 years agoaudiodecoder: Add transform_meta() vfunc with default implementation
Sebastian Dröge [Mon, 29 Jun 2015 16:00:17 +0000 (18:00 +0200)]
audiodecoder: Add transform_meta() vfunc with default implementation

The default implementation copies all metadata without tags, and metadata
with only the audio tag. Same behaviour as in GstAudioFilter.

https://bugzilla.gnome.org/show_bug.cgi?id=742385

8 years agoaudioencoder: Add transform_meta() vfunc with default implementation
Sebastian Dröge [Mon, 29 Jun 2015 15:38:38 +0000 (17:38 +0200)]
audioencoder: Add transform_meta() vfunc with default implementation

The default implementation copies all metadata without tags, and metadata
with only the audio tag. Same behaviour as in GstAudioFilter.

https://bugzilla.gnome.org/show_bug.cgi?id=742385

8 years agovideodecoder: Add transform_meta() vfunc with default implementation
Sebastian Dröge [Mon, 29 Jun 2015 13:58:38 +0000 (15:58 +0200)]
videodecoder: Add transform_meta() vfunc with default implementation

The default implementation copies all metadata without tags, and metadata
with only the video tag. Same behaviour as in GstVideoFilter.

This currently does not work if the ::parse() vfunc is implemented as all
metas are getting lost inside GstAdapter.

https://bugzilla.gnome.org/show_bug.cgi?id=742385

8 years agovideoencoder: Add transform_meta() vfunc with default implementation
Sebastian Dröge [Mon, 29 Jun 2015 11:59:25 +0000 (13:59 +0200)]
videoencoder: Add transform_meta() vfunc with default implementation

The default implementation copies all metadata without tags, and metadata
with only the video tag. Same behaviour as in GstVideoFilter.

https://bugzilla.gnome.org/show_bug.cgi?id=742385

8 years agortpbaseaudiopayload: Don't copy memory if not needed, just append payload to the...
Sebastian Dröge [Tue, 30 Jun 2015 08:37:27 +0000 (10:37 +0200)]
rtpbaseaudiopayload: Don't copy memory if not needed, just append payload to the RTP buffer

8 years agoplaybin: remove unnecessary break
danny song [Mon, 29 Jun 2015 22:26:00 +0000 (07:26 +0900)]
playbin: remove unnecessary break

https://bugzilla.gnome.org/show_bug.cgi?id=751690

8 years agovideoscaler: remove check for below zero for unsigned value
Luis de Bethencourt [Mon, 29 Jun 2015 15:16:06 +0000 (16:16 +0100)]
videoscaler: 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 a division of an unsigned integer (i). Removing that check
and only checking if it is bigger than max and setting it appropriately.

CID #1308950

8 years agoaudioresample: Also copy metas if their API has no tags attached to it
Sebastian Dröge [Mon, 29 Jun 2015 11:06:59 +0000 (13:06 +0200)]
audioresample: Also copy metas if their API has no tags attached to it

This is the default basetransform behaviour, being more strict than that
is not really useful.

8 years agoaudioconvert: Also copy metas if their API has no tags attached to it
Sebastian Dröge [Mon, 29 Jun 2015 11:06:49 +0000 (13:06 +0200)]
audioconvert: Also copy metas if their API has no tags attached to it

This is the default basetransform behaviour, being more strict than that
is not really useful.

8 years agoaudiofilter: Also copy metas if their API has no tags attached to it
Sebastian Dröge [Mon, 29 Jun 2015 11:06:33 +0000 (13:06 +0200)]
audiofilter: Also copy metas if their API has no tags attached to it

This is the default basetransform behaviour, being more strict than that
is not really useful.

8 years agovideofilter: Also copy metas if their API has no tags attached to it
Sebastian Dröge [Mon, 29 Jun 2015 11:05:54 +0000 (13:05 +0200)]
videofilter: Also copy metas if their API has no tags attached to it

This is the default basetransform behaviour, being more strict than that
is not really useful.

8 years agoBack to development
Sebastian Dröge [Wed, 24 Jun 2015 22:04:11 +0000 (00:04 +0200)]
Back to development

8 years agoRelease 1.5.2
Sebastian Dröge [Wed, 24 Jun 2015 21:24:01 +0000 (23:24 +0200)]
Release 1.5.2

8 years agoUpdate .po files
Sebastian Dröge [Wed, 24 Jun 2015 20:49:29 +0000 (22:49 +0200)]
Update .po files

8 years agopo: Update translations
Sebastian Dröge [Wed, 24 Jun 2015 09:14:21 +0000 (11:14 +0200)]
po: Update translations

8 years agostreamsynchronizer: Unblock EOS wait when track switching.
Song Bing [Wed, 17 Jun 2015 10:03:09 +0000 (18:03 +0800)]
streamsynchronizer: Unblock EOS wait when track switching.

sink_event () will blocked on EOS event. which will cause can't
send event when switch EOS track to non-EOS one.

https://bugzilla.gnome.org/show_bug.cgi?id=750761

8 years agostreamsynchronizer: Don't wait for sparse streams when doing stream switches
Sebastian Dröge [Mon, 22 Jun 2015 18:54:18 +0000 (20:54 +0200)]
streamsynchronizer: Don't wait for sparse streams when doing stream switches

Their stream-start event might come a bit later, like just before the first
buffer... and queues might run full before that happens.

8 years agostreamsynchronizer: Add some more debug output
Sebastian Dröge [Mon, 22 Jun 2015 18:29:52 +0000 (20:29 +0200)]
streamsynchronizer: Add some more debug output

8 years agostreamsynchronizer: Reset group start time when flushing
Sebastian Dröge [Mon, 22 Jun 2015 18:17:56 +0000 (20:17 +0200)]
streamsynchronizer: Reset group start time when flushing

We reset the group start time to the running time of the start of the other
streams that are not flushed. This fixes seeking in gapless mode after the
first track has played.

https://bugzilla.gnome.org/show_bug.cgi?id=750013

8 years agortspconnection: Only drop everything after the ; of a session header in requests
Sebastian Dröge [Mon, 22 Jun 2015 17:51:32 +0000 (19:51 +0200)]
rtspconnection: Only drop everything after the ; of a session header in requests

For responses it is actually allowed and used to signal the timeout to the
client!

https://bugzilla.gnome.org/show_bug.cgi?id=736267

8 years agoaudioringbuffer: Fix alaw/mulaw channel positions
Lyon Wang [Thu, 18 Jun 2015 09:38:09 +0000 (17:38 +0800)]
audioringbuffer: Fix alaw/mulaw channel positions

For alaw/mulaw we should also try to initialize the channel positions in the
ringbuffer's audio info. This allow pulsesink to directly use the channel
positions instead of using the default zero-initialized ones, which doesn't
work well.

https://bugzilla.gnome.org/show_bug.cgi?id=751144

8 years agotests: fix cpp directives
Wim Taymans [Mon, 22 Jun 2015 14:53:06 +0000 (16:53 +0200)]
tests: fix cpp directives

8 years agotests: add PPC64 abi struct sizes
Wim Taymans [Mon, 22 Jun 2015 13:59:42 +0000 (15:59 +0200)]
tests: add PPC64 abi struct sizes

8 years agoplaybin: Reset suburi also when receiving an error message from the sub uridecodebin
Sebastian Dröge [Mon, 22 Jun 2015 12:51:07 +0000 (14:51 +0200)]
playbin: Reset suburi also when receiving an error message from the sub uridecodebin

http://bugzilla.gnome.org/show_bug.cgi?id=751118

8 years agoplaybin: free group->suburi on failure
Brijesh Singh [Wed, 17 Jun 2015 15:20:54 +0000 (10:20 -0500)]
playbin: free group->suburi on failure

If suburidecodebin is failed to negotiate (e.g file does not exist)
then free internal suburi variable so that 'current-suburi' property
returns correct status.

https://bugzilla.gnome.org/show_bug.cgi?id=751118

8 years agooggdemux: set building_chain to NULL when deactivating chain
Guillaume Desmottes [Mon, 15 Jun 2015 14:08:10 +0000 (16:08 +0200)]
oggdemux: set building_chain to NULL when deactivating chain

The chain is about to be invalidated so we shouldn't keep it around.
Prevent a double free crash when the demuxer is being finalized.

https://bugzilla.gnome.org/show_bug.cgi?id=751000

8 years agotools: gst-play: fix seeking issue
Vineeth T M [Fri, 19 Jun 2015 10:51:25 +0000 (19:51 +0900)]
tools: gst-play: fix seeking issue

For positive seeking segment.stop value will be -1,
when we change rate to -1, then the stop value will be udpated
with the current position. And then again if we change rate to 1,
the segment.stop value does not get updated and remains as position
where we last changed rate to -1. Hence playback stops at that point.
In case of positive rates, call gst_element_new_seek with correct values

https://bugzilla.gnome.org/show_bug.cgi?id=751213

8 years agodoc: Fix gsttrtphdrext section name
Nicolas Dufresne [Fri, 19 Jun 2015 01:02:48 +0000 (21:02 -0400)]
doc: Fix gsttrtphdrext section name

8 years agovideo: Add missing new symbol to win32 def file
Nicolas Dufresne [Thu, 18 Jun 2015 22:23:45 +0000 (18:23 -0400)]
video: Add missing new symbol to win32 def file

Fixes make distcheck

8 years agoAdd gst_video_multiview_guess_half_aspect() to the docs
Jan Schmidt [Thu, 18 Jun 2015 16:19:12 +0000 (02:19 +1000)]
Add gst_video_multiview_guess_half_aspect() to the docs

8 years agomultiview: Add gst_video_multiview_guess_half_aspect()
Jan Schmidt [Mon, 15 Jun 2015 06:04:55 +0000 (16:04 +1000)]
multiview: Add gst_video_multiview_guess_half_aspect()

Add a utility function that, given a video size and a
packed stereoscopic mode, attempts to guess if the video
is packed at half resolution per view or not, since
very few videos provide the information.

8 years agovideo-converter: make sure we draw enough border for YUY2 formats
Wim Taymans [Wed, 17 Jun 2015 15:09:46 +0000 (17:09 +0200)]
video-converter: make sure we draw enough border for YUY2 formats

Round width up to 2 so that we draw all border pixels for YUY2 formats

8 years agovideo-scaler: fix scaling of odd width for YUY2 formats
Wim Taymans [Wed, 17 Jun 2015 14:43:03 +0000 (16:43 +0200)]
video-scaler: fix scaling of odd width for YUY2 formats

We need to scale groups of 4 bytes for YUY2 formats so round up to 4.
It's possible that there is no Y byte for the last pixel so make sure
we clamp correctly.

8 years agodiscoverer: Fix a wrong naming in the documentation
Thibault Saunier [Wed, 17 Jun 2015 08:02:08 +0000 (10:02 +0200)]
discoverer: Fix a wrong naming in the documentation

gst_discoverer_stream_get_missing_elements_installer_details does not
exist, one should use gst_discoverer_info_get_missing_elements_installer_details

8 years agogi: Use INTROSPECTION_INIT for --add-init-section
Nicolas Dufresne [Tue, 16 Jun 2015 22:04:57 +0000 (18:04 -0400)]
gi: Use INTROSPECTION_INIT for --add-init-section

This new define was added to common. The new init section fixed
compilation warning found in the init line that was spread across
all files.

8 years agoAutomatic update of common submodule
Nicolas Dufresne [Tue, 16 Jun 2015 21:47:24 +0000 (17:47 -0400)]
Automatic update of common submodule

From 6015d26 to f74b2df

8 years agotools: gst-play: error out instead of crashing if there's no playbin element
Tim-Philipp Müller [Tue, 16 Jun 2015 21:32:49 +0000 (22:32 +0100)]
tools: gst-play: error out instead of crashing if there's no playbin element

8 years agogi: Skip Scaler, Chroma, Conveter, Dither constructor
Nicolas Dufresne [Tue, 16 Jun 2015 20:08:39 +0000 (16:08 -0400)]
gi: Skip Scaler, Chroma, Conveter, Dither constructor

Please box these types before removing the skip mark.

8 years agogi: Add (transfer none) for various video meta
Nicolas Dufresne [Tue, 16 Jun 2015 20:07:27 +0000 (16:07 -0400)]
gi: Add (transfer none) for various video meta

These method chains gst_buffer_add_meta() which is also transfer
none.

8 years agogi: Fix warnings in GstRtsp
Nicolas Dufresne [Tue, 16 Jun 2015 19:50:13 +0000 (15:50 -0400)]
gi: Fix warnings in GstRtsp

* The custom GSource is not boxed (skip for now)
* The comment block has wrong name for _read_socket()

8 years agogi: Don't produce gir and typlib for GstRiff
Nicolas Dufresne [Tue, 16 Jun 2015 19:16:33 +0000 (15:16 -0400)]
gi: Don't produce gir and typlib for GstRiff

The API does not follow the type naming convention. Re-enable
only if one take the time to box and rename (see (rename-to SYMBOL)
annotation) all types.

8 years agogi: Skip fft constructor for now
Nicolas Dufresne [Tue, 16 Jun 2015 18:36:44 +0000 (14:36 -0400)]
gi: Skip fft constructor for now

These types have never been boxed, hence cannot be used
safely in interpreted languages. This fixes warnings.

8 years agogi: Fix warnings in libgstaudio
Nicolas Dufresne [Tue, 16 Jun 2015 18:34:04 +0000 (14:34 -0400)]
gi: Fix warnings in libgstaudio

* Duplicate section
* Miss-named parameter
* Missing transfer none annotation for meta

9 years agobasetextoverlay: add "draw-shadow" and "draw-outline" properties
Lazar Claudiu [Wed, 27 May 2015 09:20:19 +0000 (12:20 +0300)]
basetextoverlay: add "draw-shadow" and "draw-outline" properties

https://bugzilla.gnome.org/show_bug.cgi?id=749823

9 years agovideoencoder: fix gtk-doc chunk for new function
Tim-Philipp Müller [Sat, 13 Jun 2015 12:41:35 +0000 (13:41 +0100)]
videoencoder: fix gtk-doc chunk for new function

9 years agobasetextoverlay: Rewrite negotiation method
Nicolas Dufresne [Tue, 12 May 2015 18:12:52 +0000 (14:12 -0400)]
basetextoverlay: Rewrite negotiation method

This cleanup the negotiation function by properly splitting the probe
and the decisions. This allow handling correctly pipeline where upstream
caps have special memory type. An example pipeline is:

  gltestsrc ! textoverlay text=bla ! fakesink

The upstream caps will be memory:GLMemory, which isn't supported by the
blitter.

https://bugzilla.gnome.org/show_bug.cgi?id=749243

9 years agoxvimagesink: Don't share internal pool
Nicolas Dufresne [Fri, 5 Jun 2015 18:30:12 +0000 (14:30 -0400)]
xvimagesink: Don't share internal pool

Sharing the internal pool results in situation where the pool may have
two upstream owners. This creates a race upon deactivation. Instead,
always offer a new pool, and keep the internal pool internal in case
we absolutely need it.

https://bugzilla.gnome.org/show_bug.cgi?id=748344

9 years agoximagesink: Don't share internal pool
Nicolas Dufresne [Fri, 5 Jun 2015 18:28:41 +0000 (14:28 -0400)]
ximagesink: Don't share internal pool

Sharing the internal pool results in situation where the pool may have
two upstream owners. This create a race upon deactivation. Instead,
always offer a new pool, and keep the internal pool internal in case
we absolutely need it.

https://bugzilla.gnome.org/show_bug.cgi?id=748344

9 years agovideoencoder: Add gst_video_encoder_set_min_pts()
Matej Knopp [Wed, 26 Nov 2014 20:06:57 +0000 (21:06 +0100)]
videoencoder: Add gst_video_encoder_set_min_pts()

For streams with reordered frames this can be used to ensure that there
is enough time to accomodate first DTS, which may be less than first PTS

https://bugzilla.gnome.org/show_bug.cgi?id=740575

9 years agoUpdate .def file for new API
Tim-Philipp Müller [Fri, 12 Jun 2015 18:58:34 +0000 (19:58 +0100)]
Update .def file for new API

9 years agovideodecoder: Copy multiview-mode, flags and view count from ref info
Jan Schmidt [Fri, 12 Jun 2015 15:35:52 +0000 (01:35 +1000)]
videodecoder: Copy multiview-mode, flags and view count from ref info

When copying info from the reference input state, duplicate
all the fields of the video info. The sub-class will have the
chance to override them later.

9 years agovideo-scaler: enforce same taps when combining scalers
Wim Taymans [Fri, 12 Jun 2015 14:57:39 +0000 (16:57 +0200)]
video-scaler: enforce same taps when combining scalers

9 years agovideo-scaler: make sure to clamp to max width
Wim Taymans [Fri, 12 Jun 2015 14:52:27 +0000 (16:52 +0200)]
video-scaler: make sure to clamp to max width

When estimating the area that should first be vertically scaled, make
sure we clamp to the max input size or else we get invalid reads.

9 years agovideo-scaler: Enforce same taps on Y and UV scalers for merged formats
Wim Taymans [Fri, 12 Jun 2015 14:47:03 +0000 (16:47 +0200)]
video-scaler: Enforce same taps on Y and UV scalers for merged formats

Make sure we have the same number of taps for the Y and UV scalers so
that the scalers can be merged correctly.

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=749764

9 years agortsp: Add a FIXME 2.0 for gst_rtsp_connection_create_from_socket()
Arun Raghavan [Fri, 12 Jun 2015 07:20:35 +0000 (12:50 +0530)]
rtsp: Add a FIXME 2.0 for gst_rtsp_connection_create_from_socket()

There's a couple of redundant arguments from the pre-GIO days.

9 years agotypefinding: check for full UTF-8 BOM in MSS typefinder
Tim-Philipp Müller [Thu, 11 Jun 2015 22:32:55 +0000 (23:32 +0100)]
typefinding: check for full UTF-8 BOM in MSS typefinder

https://bugzilla.gnome.org/show_bug.cgi?id=750802

9 years agotypefindfunctions: UTF-8 MSS Manifest detection support
Philippe Normand [Thu, 11 Jun 2015 16:14:47 +0000 (18:14 +0200)]
typefindfunctions: UTF-8 MSS Manifest detection support

Check if the first bytes of data contain an UTF-8 BOM.

https://bugzilla.gnome.org/show_bug.cgi?id=750802

9 years agoplaybin: Check in autoplug_continue against the subtitle factory caps correctly
Sebastian Dröge [Thu, 11 Jun 2015 14:18:51 +0000 (16:18 +0200)]
playbin: Check in autoplug_continue against the subtitle factory caps correctly

6a2f017bfa9cb73c6db65eea0b84b1d5b56febb7 changed it to check the subtitle
factory caps if there is a text-sink but we fail to get its sinkpad. What
actually should be done here is to use the factory caps if there is no
text-sink at all.

https://bugzilla.gnome.org/show_bug.cgi?id=750785

9 years agoplaybin: Fix some warnings with clang around multiview enums
Jan Schmidt [Thu, 11 Jun 2015 13:01:48 +0000 (23:01 +1000)]
playbin: Fix some warnings with clang around multiview enums

There is the GstVideoMultiviewMode enum and the
GstVideoMultiviewFramePacking, which is a subset of the
multiview modes, with the same values as the corresponding
types from the full enum. Do some casts and use the right
times to avoid implicitly using/passing GstVideoMultiviewFramePacking
when a GstVideoMultiviewMode is needed.

9 years agotests: Fix video libs test for multiview GstVideoInfo change
Jan Schmidt [Thu, 11 Jun 2015 02:21:08 +0000 (12:21 +1000)]
tests: Fix video libs test for multiview GstVideoInfo change

The GstVideoInfo struct was changed late in integrating the
multiview changes, and I forgot to run and fix the unit test.

9 years agoplaybin: Implement multiview frame-packing overrides
Jan Schmidt [Thu, 11 Jun 2015 01:12:39 +0000 (11:12 +1000)]
playbin: Implement multiview frame-packing overrides

Add GstVideoMultiviewFramePacking enum, and the
video-multiview-mode and video-multiview-flags
properties on playbin.

Use a pad probe to replace the multiview information in
video caps sent out from uridecodebin.

This is a part implementation only - for full
correctness, it should also modify caps in caps events,
accept-caps and allocation queries.

https://bugzilla.gnome.org/show_bug.cgi?id=611157

9 years agovideo: Add multiview/stereo support
Jan Schmidt [Thu, 11 Jun 2015 01:12:39 +0000 (11:12 +1000)]
video: Add multiview/stereo support

Add flags and enums to support multiview signalling in
GstVideoInfo and GstVideoFrame, and the caps serialisation and
deserialisation.

videoencoder: Copy multiview settings from reference input state

Add gst_video_multiview_* support API and GstVideoMultiviewMeta meta

https://bugzilla.gnome.org/show_bug.cgi?id=611157

9 years agortpbaseaudiopayload: Use GST_BUFFER_PTS() instead of GST_BUFFER_TIMESTAMP()
Sebastian Dröge [Wed, 10 Jun 2015 12:33:01 +0000 (14:33 +0200)]
rtpbaseaudiopayload: Use GST_BUFFER_PTS() instead of GST_BUFFER_TIMESTAMP()

9 years agoplaysink: fix the channel of color balance element
Víctor Manuel Jáquez Leal [Wed, 10 Jun 2015 10:26:38 +0000 (12:26 +0200)]
playsink: fix the channel of color balance element

When traversing the color balance element channel list to find the one that
matches with the playsink proxy, the assignation was set to iterator of the
playsink proxy, not the balance element. Thus, the mapping to the values of
the balance element channel was wrong.

This patch fixes the assignation of the color balance element channel, so the
mapping to the channel of the color balance element is fixed.

https://bugzilla.gnome.org/show_bug.cgi?id=750691

9 years agoplaysink: cannot enable text flag while playing
Vineeth TM [Wed, 10 Jun 2015 06:50:12 +0000 (15:50 +0900)]
playsink: cannot enable text flag while playing

when text playbin is not enabled in the beginning, then
video_srcpad_stream_synchronizer gets linked to videochain->sinkpad
and when we try to enable text bin during play, since it is already linked to videochain,
text chain does not get linked properly. Hence unlinking the same
before linking to text chain

https://bugzilla.gnome.org/show_bug.cgi?id=748908

9 years agowin32: Update defs file
Edward Hervey [Wed, 10 Jun 2015 07:59:49 +0000 (09:59 +0200)]
win32: Update defs file

9 years agoGstRTSPConnection: Add GTlsInteraction support
Xavier Claessens [Sat, 6 Jun 2015 02:04:24 +0000 (22:04 -0400)]
GstRTSPConnection: Add GTlsInteraction support

https://bugzilla.gnome.org/show_bug.cgi?id=750471

9 years agotools: gst-play: don't print 64 whitespaces next to the time indication
Vivia Nikolaidou [Tue, 9 Jun 2015 18:24:07 +0000 (21:24 +0300)]
tools: gst-play: don't print 64 whitespaces next to the time indication

Printing 64 whitespaces to erase the "Paused" message (after \r) would make
it wrap to the next line on shorter terminals. Instead we only print the
amount of spaces needed. Also mark the "Paused" string for translation
while we're at it.

9 years agocruft: add the obsolete tmpl dir to cruft-dirs
Stefan Sauer [Tue, 9 Jun 2015 12:31:15 +0000 (14:31 +0200)]
cruft: add the obsolete tmpl dir to cruft-dirs

9 years agoUpdate win32 exports
Jan Schmidt [Tue, 9 Jun 2015 12:03:37 +0000 (22:03 +1000)]
Update win32 exports

9 years agoalsa: report recoverable device failures to base class
Carlos Rafael Giani [Mon, 9 Dec 2013 17:46:14 +0000 (18:46 +0100)]
alsa: report recoverable device failures to base class

This gives custom slave methods in the base class a chance to
resynchronize themselves

Signed-off-by: Carlos Rafael Giani <dv@pseudoterminal.org>
https://bugzilla.gnome.org/show_bug.cgi?id=708362

9 years agoaudiobasesink: added custom clock slaving method
Carlos Rafael Giani [Mon, 9 Dec 2013 16:08:15 +0000 (17:08 +0100)]
audiobasesink: added custom clock slaving method

This new clock slaving method allows for installing a callback that is
invoked during playback. Inside this callback, a custom slaving
mechanism can be used (for example, a control loop adjusting a PLL or an
asynchronous resampler). Upon request, it can skew the playout pointer
just like the "skew" method. This is useful if the clocks drifted apart
too much, and a quick reset is necessary.

Signed-off-by: Carlos Rafael Giani <dv@pseudoterminal.org>
https://bugzilla.gnome.org/show_bug.cgi?id=708362

9 years agoAutomatic update of common submodule
Edward Hervey [Tue, 9 Jun 2015 09:30:15 +0000 (11:30 +0200)]
Automatic update of common submodule

From d9a3353 to 6015d26

9 years agotools: gst-play: add shortcuts to switch audio/subtitle/video tracks
Tim-Philipp Müller [Tue, 9 Jun 2015 09:16:34 +0000 (10:16 +0100)]
tools: gst-play: add shortcuts to switch audio/subtitle/video tracks

9 years agoplayback: Skip 'ANY' capsfeature while finding the count of common capsfeatures
Sreerenj Balachandran [Wed, 5 Nov 2014 07:41:36 +0000 (09:41 +0200)]
playback: Skip 'ANY' capsfeature while finding the count of common capsfeatures

https://bugzilla.gnome.org/show_bug.cgi?id=687182

9 years agoplayback: Add gstplaybackutils.{h,c} to deploy the common subroutines
Sreerenj Balachandran [Wed, 5 Nov 2014 07:40:43 +0000 (09:40 +0200)]
playback: Add gstplaybackutils.{h,c} to deploy the common subroutines

Bring some of the helper functions in gstplaybin2.c to new files
gstplaybackutils.{h,c} which can be utilized by other files
in gst/playback too.

https://bugzilla.gnome.org/show_bug.cgi?id=687182

9 years agoAutomatic update of common submodule
Stefan Sauer [Mon, 8 Jun 2015 21:07:47 +0000 (23:07 +0200)]
Automatic update of common submodule

From d37af32 to d9a3353

9 years agotools: gst-play: sort directory entries
Vivia Nikolaidou [Mon, 8 Jun 2015 17:32:02 +0000 (20:32 +0300)]
tools: gst-play: sort directory entries

When adding a directory to the playlist, the order would be whatever
g_dir_read_name returned. Sorting these using natural sort order.

https://bugzilla.gnome.org/show_bug.cgi?id=750585

9 years agoximagesink, xvimagesink: fix string leaks when setting class hint
Tim-Philipp Müller [Mon, 8 Jun 2015 19:17:07 +0000 (20:17 +0100)]
ximagesink, xvimagesink: fix string leaks when setting class hint

https://bugzilla.gnome.org/show_bug.cgi?id=750455

9 years agovideo: Allow using bt2020 by name in colorimetry
Nicolas Dufresne [Mon, 8 Jun 2015 17:01:43 +0000 (13:01 -0400)]
video: Allow using bt2020 by name in colorimetry

As the lookup stops at the first element in the array with a NULL
name, bt2020 could not be used by name. Moving up this entry
fixes the issue.

9 years agoximagesink: set WM_CLASS of window
Luis de Bethencourt [Fri, 5 Jun 2015 15:01:05 +0000 (16:01 +0100)]
ximagesink: set WM_CLASS of window

Set WM_CLASS of the ximagesink window so window managers can apply rules
based on xprop filtering.

9 years agoxvimagesink: set WM_CLASS of window
Luis de Bethencourt [Fri, 5 Jun 2015 14:58:39 +0000 (15:58 +0100)]
xvimagesink: set WM_CLASS of window

Set WM_CLASS of the xvimagesink window so window managers can apply rules
based on xprop filtering.

9 years agoAutomatic update of common submodule
Stefan Sauer [Sun, 7 Jun 2015 21:06:08 +0000 (23:06 +0200)]
Automatic update of common submodule

From 21ba2e5 to d37af32

9 years agolibvisual: clean dereferences of private structures
Luis de Bethencourt [Sun, 7 Jun 2015 17:49:48 +0000 (18:49 +0100)]
libvisual: clean dereferences of private structures

https://bugzilla.gnome.org/show_bug.cgi?id=742875

9 years agolibvisual: make private all variable subclasses don't need
Luis de Bethencourt [Sun, 7 Jun 2015 17:23:23 +0000 (18:23 +0100)]
libvisual: make private all variable subclasses don't need

https://bugzilla.gnome.org/show_bug.cgi?id=742875

9 years agoAutomatic update of common submodule
Stefan Sauer [Sun, 7 Jun 2015 15:31:55 +0000 (17:31 +0200)]
Automatic update of common submodule

From c408583 to 21ba2e5

9 years agodocs: remove variables that we define in the snippet from common
Stefan Sauer [Sun, 7 Jun 2015 15:00:05 +0000 (17:00 +0200)]
docs: remove variables that we define in the snippet from common

This is syncing our Makefile.am with upstream gtkdoc.

9 years agoAutomatic update of common submodule
Stefan Sauer [Sun, 7 Jun 2015 15:16:13 +0000 (17:16 +0200)]
Automatic update of common submodule

From 241fcb7 to c408583