platform/upstream/gst-plugins-good.git
6 years agov4l2object: Always set the colorimetry in S_FMT
Nicolas Dufresne [Sun, 8 Apr 2018 17:43:56 +0000 (13:43 -0400)]
v4l2object: Always set the colorimetry in S_FMT

So far we were only setting colorimetry for OUTPUT devices (v4l2sink or
m2m sink pad). This prevented selecting through caps negotiation the
colorimetry for CAPTURE devices (v4l2src or m2m src pad). This is rarely
selectable, but trying is harmless.

6 years agomonoscope: Only fixate pixel-aspect-ratio if the field exists
Sebastian Dröge [Wed, 11 Apr 2018 19:41:58 +0000 (21:41 +0200)]
monoscope: Only fixate pixel-aspect-ratio if the field exists

6 years agosplitmuxsink: Don't send fragment-opened-closed message if the reference ctx is NULL
Vivia Nikolaidou [Wed, 11 Apr 2018 14:54:38 +0000 (17:54 +0300)]
splitmuxsink: Don't send fragment-opened-closed message if the reference ctx is NULL

It can happen during teardown that the reference context becomes NULL.
In that case, trying to send the fragment-opened-closed message would
lead to a crash.

6 years agosplitmuxsink: Run gst_iterator_foreach() as long as it returns GST_ITERATOR_RESYNC
Sebastian Dröge [Wed, 11 Apr 2018 07:12:09 +0000 (09:12 +0200)]
splitmuxsink: Run gst_iterator_foreach() as long as it returns GST_ITERATOR_RESYNC

CID 1434160

6 years agoqtdemux: Fix comparision for extra caption atom
Edward Hervey [Wed, 11 Apr 2018 06:51:32 +0000 (08:51 +0200)]
qtdemux: Fix comparision for extra caption atom

We want to make sure we have *enough* data for the potential 2nd
caption atom.

CID #1434161

6 years agoqtdemux: Handle bogus caption samples
Edward Hervey [Wed, 11 Apr 2018 06:42:54 +0000 (08:42 +0200)]
qtdemux: Handle bogus caption samples

Corrupted files could potentially have multiple cdat/cdt2 atoms in
a sample entry, which is unclear how to handle.

Ignore repeated ones.

CID #1434162
CID #1434159

6 years agomonoscope: Fixate pixel-aspect-ratio too and make sure the final caps are completely...
Sebastian Dröge [Tue, 10 Apr 2018 19:15:48 +0000 (21:15 +0200)]
monoscope: Fixate pixel-aspect-ratio too and make sure the final caps are completely fixated

Otherwise e.g. this fails with assertions:
gst-launch-1.0 audiotestsrc ! audioconvert ! monoscope ! videoconvert ! \
    videoscale ! video/x-raw,width=800,height=600 ! ximagesink

6 years agoqtmux: Add comments and doc about prefill mode
Edward Hervey [Thu, 8 Mar 2018 09:10:01 +0000 (10:10 +0100)]
qtmux: Add comments and doc about prefill mode

6 years agoqtmux: Refactor pad re-negotiation code
Edward Hervey [Tue, 6 Feb 2018 13:36:50 +0000 (14:36 +0100)]
qtmux: Refactor pad re-negotiation code

It was similar for all pads

https://bugzilla.gnome.org/show_bug.cgi?id=606643

6 years agoqtdemux: Detect and expose CEA 608/708 Closed Caption tracks
Edward Hervey [Wed, 31 Jan 2018 14:10:03 +0000 (15:10 +0100)]
qtdemux: Detect and expose CEA 608/708 Closed Caption tracks

https://bugzilla.gnome.org/show_bug.cgi?id=606643

6 years agortxsend: fix wrong memory layout assumption
Mathieu Duponchelle [Tue, 3 Apr 2018 23:48:44 +0000 (01:48 +0200)]
rtxsend: fix wrong memory layout assumption

The code responsible for creating retransmitted buffers
assumed the stored buffer had been created with
rtp_buffer_new_allocate when copying the extension data,
which isn't necessarily the case, for example when
the rtp buffers come from a udpsrc.

https://bugzilla.gnome.org/show_bug.cgi?id=794958

6 years agortpbin: new signal "get-storage"
Mathieu Duponchelle [Mon, 2 Apr 2018 21:04:06 +0000 (23:04 +0200)]
rtpbin: new signal "get-storage"

Similar to the get-session and get-internal-session signals,
we expose a get-storage signal in addition to the
get-internal-storage signal to give access to the actual
element for applications that need to set properties on the
element, in particular "size-time"

https://bugzilla.gnome.org/show_bug.cgi?id=794910

6 years agosplitmuxsink: Add new reset-muxer property
Sebastian Dröge [Thu, 29 Mar 2018 16:19:21 +0000 (19:19 +0300)]
splitmuxsink: Add new reset-muxer property

With this the muxer is not set to NULL after each segment but instead
only flush events are sent to it to reset the EOS state.

As a result, the muxer will keep stream state and e.g. mpegtsmux will
keep the packet continuity counter continuous between segments as needed
by hlssink2.

https://bugzilla.gnome.org/show_bug.cgi?id=794816

6 years agotests: remove broken and now pointless v4l2src-test
Tim-Philipp Müller [Mon, 2 Apr 2018 11:48:50 +0000 (12:48 +0100)]
tests: remove broken and now pointless v4l2src-test

This tests APIs that don't exist any longer and also doesn't
work at all, and was last touched in a meaningful way in 2006.

6 years agov4l2: Fix unknown type name ‘off_t’ error
Seungha Yang [Tue, 20 Mar 2018 15:19:37 +0000 (00:19 +0900)]
v4l2: Fix unknown type name ‘off_t’ error

Fix following build error

gstv4l2object.h:197:17: error: unknown type name ‘off_t’
       gint fd,  off_t offset);
                 ^

https://bugzilla.gnome.org/show_bug.cgi?id=794533

6 years agortspsrc: reject segment seeks
Mathieu Duponchelle [Thu, 25 May 2017 01:44:39 +0000 (03:44 +0200)]
rtspsrc: reject segment seeks

https://bugzilla.gnome.org/show_bug.cgi?id=784681

6 years agoqtdemux: Handle variant of vorbis in mp4
Edward Hervey [Tue, 13 Feb 2018 10:50:05 +0000 (11:50 +0100)]
qtdemux: Handle variant of vorbis in mp4

Comes from gpac apparently. The codec_data uses the same packing
mechanism as matroska.

https://bugzilla.gnome.org/show_bug.cgi?id=738244

6 years agoqtdemux: Check sample count is valid in PIFF parsing
Edward Hervey [Thu, 22 Mar 2018 14:20:47 +0000 (15:20 +0100)]
qtdemux: Check sample count is valid in PIFF parsing

The value stored in cenc_aux_sample_count wasn't in sync with the
parsing code that followed which checks whether all entries are
valid and present.

Only write the actual sample count when we know for sure.

CID #1427087

6 years agoqt: Get EGL native display from QPA if platform header is available
Carlos Rafael Giani [Sun, 4 Mar 2018 14:14:08 +0000 (15:14 +0100)]
qt: Get EGL native display from QPA if platform header is available

https://bugzilla.gnome.org/show_bug.cgi?id=792378

6 years agoudpsrc: switch to using a buffer pool
Petr Kulhavy [Tue, 6 Mar 2018 01:14:34 +0000 (02:14 +0100)]
udpsrc: switch to using a buffer pool

This exposes a new property, mtu, which is used to determine the
initial size of buffers from the buffer pool. If received data
exceeds this, the element gracefully handles that in a manner similar
to what we had previously: a large memory gets filled and reallocated
at the next call to "fill".

The default size is set to 1500, which should cover most use cases.

With contributions from Mathieu Duponchelle <mathieu@centricular.com>

https://bugzilla.gnome.org/show_bug.cgi?id=772841

6 years agoudpsrc: optimize GstUdpSrc object for cache performance
Petr Kulhavy [Tue, 15 Nov 2016 08:39:31 +0000 (09:39 +0100)]
udpsrc: optimize GstUdpSrc object for cache performance

Optimize GstUdpSrc for cache performance.

Move the hot properties, which are used by the read function, to the top:
@used_socket, @addr, @cancellable, @skip_first_bytes, @timeout,
@retrieve_sender_address.

Remove the unused property @ttl.

Where needed reorder so that holes are avoided (the 64-bit @timeout)

https://bugzilla.gnome.org/show_bug.cgi?id=772841

6 years agoqtdemux: Fix seeking on streams with frame reordering
Sebastian Dröge [Mon, 5 Mar 2018 10:48:15 +0000 (12:48 +0200)]
qtdemux: Fix seeking on streams with frame reordering

The samples table is sorted by DTS, not PTS. As such we can only get the
correct result when using a binary search on it, if we search for the
DTS.
Also if we only ever search for the frame, where the following frame is
the first one with a PTS after the search position, we will generally
stop searching too early if frames are reordered.

In forwards playback this is not really a problem (after the decoder
reordered the frames, clipping is happening), in reverse playback
it means that we can output one or more frames too few as we stop too
early and the decoder would never receive it.

https://bugzilla.gnome.org/show_bug.cgi?id=782118

6 years agortp: Fix compilation with non-C99 compilers
Sebastian Dröge [Tue, 20 Mar 2018 09:36:32 +0000 (11:36 +0200)]
rtp: Fix compilation with non-C99 compilers

By moving variable declarations out of loop headers.

6 years agoBack to development
Tim-Philipp Müller [Tue, 20 Mar 2018 09:24:19 +0000 (09:24 +0000)]
Back to development

6 years agoRelease 1.14.0
Tim-Philipp Müller [Mon, 19 Mar 2018 20:18:22 +0000 (20:18 +0000)]
Release 1.14.0

6 years agoUpdate docs
Tim-Philipp Müller [Mon, 19 Mar 2018 20:18:22 +0000 (20:18 +0000)]
Update docs

6 years agortpulpfecdec: fix build with older gcc
Tim-Philipp Müller [Mon, 19 Mar 2018 18:39:08 +0000 (18:39 +0000)]
rtpulpfecdec: fix build with older gcc

As on Ubuntu Trusty.

https://bugzilla.gnome.org/show_bug.cgi?id=794493

6 years agosplitmuxsink: Allow splitting at exactly the time/bytes threshold
Sebastian Dröge [Mon, 19 Mar 2018 08:58:28 +0000 (10:58 +0200)]
splitmuxsink: Allow splitting at exactly the time/bytes threshold

76e458a119926424e9dd5acf3210a592a314d713 changed the conditions from
"queued > threshold" to "queued >= threshold", which broke hlssink2 and
resulting in too small fragments being created although keyframes would
be at *exactly* the configured threshold.

https://bugzilla.gnome.org/show_bug.cgi?id=794440

6 years agortpulpfec: fix unconditional use of __attribute__ ((packed))
Tim-Philipp Müller [Sat, 17 Mar 2018 20:29:35 +0000 (20:29 +0000)]
rtpulpfec: fix unconditional use of __attribute__ ((packed))

Fix compilation with MSVC. We still assume that attribute
is supported by all other relevant compilers, which seems
to be the case since we haven't had any complaints about
similar code in rtpsbcpay.

6 years agortpulpfec: don't use non-portable notation for 64-bit int constants
Tim-Philipp Müller [Sat, 17 Mar 2018 13:04:47 +0000 (13:04 +0000)]
rtpulpfec: don't use non-portable notation for 64-bit int constants

Use GLib macro instead, even if it's a bit unwieldy.

6 years agortpulpfecdec: don't use __builtin_ctzll unconditionally
Tim-Philipp Müller [Sat, 17 Mar 2018 12:55:57 +0000 (12:55 +0000)]
rtpulpfecdec: don't use __builtin_ctzll unconditionally

Fixes build with MSVC, and possibly other compilers too.

6 years agoRelease 1.13.91
Tim-Philipp Müller [Tue, 13 Mar 2018 19:16:42 +0000 (19:16 +0000)]
Release 1.13.91

6 years agoUpdate docs
Tim-Philipp Müller [Tue, 13 Mar 2018 19:16:42 +0000 (19:16 +0000)]
Update docs

6 years agodocs: rtpbin: add some Since markers for new properties
Tim-Philipp Müller [Mon, 12 Mar 2018 13:21:08 +0000 (13:21 +0000)]
docs: rtpbin: add some Since markers for new properties

6 years agomeson: Add deviceprovider changes to directsoundsink
Nirbheek Chauhan [Sat, 10 Mar 2018 13:27:38 +0000 (18:57 +0530)]
meson: Add deviceprovider changes to directsoundsink

These were missed when they were added to Makefile.am

6 years agoconfigure.ac: enable largefile support if possible
Michael Tretter [Thu, 8 Mar 2018 09:12:16 +0000 (10:12 +0100)]
configure.ac: enable largefile support if possible

https://bugzilla.gnome.org/show_bug.cgi?id=793103

6 years agov4l2: Fix support for 32bit mmap
Nicolas Dufresne [Wed, 7 Mar 2018 19:16:02 +0000 (14:16 -0500)]
v4l2: Fix support for 32bit mmap

https://bugzilla.gnome.org/show_bug.cgi?id=793103

6 years agoRelease 1.13.90
Tim-Philipp Müller [Sat, 3 Mar 2018 22:19:36 +0000 (22:19 +0000)]
Release 1.13.90

6 years agoUpdate docs
Tim-Philipp Müller [Sat, 3 Mar 2018 22:19:36 +0000 (22:19 +0000)]
Update docs

6 years agoflvmux: Duration & unit tests
Olivier Crête [Thu, 1 Mar 2018 23:24:33 +0000 (18:24 -0500)]
flvmux: Duration & unit tests

The muxed buffers will not carry the duration of the
incoming buffers.

https://bugzilla.gnome.org/show_bug.cgi?id=793457

6 years agoflvmux: Set PTS based on running time
Olivier Crête [Thu, 1 Mar 2018 22:15:02 +0000 (17:15 -0500)]
flvmux: Set PTS based on running time

https://bugzilla.gnome.org/show_bug.cgi?id=793457

6 years agortspsrc: Ignore sendonly/recvonly attributes unless a backchannel is configured
Sebastian Dröge [Thu, 1 Mar 2018 16:13:20 +0000 (18:13 +0200)]
rtspsrc: Ignore sendonly/recvonly attributes unless a backchannel is configured

This works around a bug in various ONVIF cameras that implement the
attributes the wrong way around. They still won't work with a
backchannel but at least normal playback will work for the time being.
It restores pre-1.14 behaviour where we would fail to preroll on any SDP
that lists a recvonly stream. For 1.16 a better solution should be
found.

The problem here is that the ONVIF spec has the meaning of the two
attributes the wrong way around in the examples, compared to RFC4566.

https://bugzilla.gnome.org/show_bug.cgi?id=793715

6 years agomeson: enable more warnings
Mathieu Duponchelle [Thu, 1 Mar 2018 17:16:24 +0000 (18:16 +0100)]
meson: enable more warnings

https://bugzilla.gnome.org/show_bug.cgi?id=793961

6 years agoPort to latest GstAggregator segment API
Mathieu Duponchelle [Wed, 28 Feb 2018 23:34:20 +0000 (00:34 +0100)]
Port to latest GstAggregator segment API

The aggregator segment is now exposed on the src pad

https://bugzilla.gnome.org/show_bug.cgi?id=793945

6 years agodirectsoundsink: Downgrade rank to match directsoundsrc in -bad
Nirbheek Chauhan [Thu, 1 Mar 2018 10:04:13 +0000 (15:34 +0530)]
directsoundsink: Downgrade rank to match directsoundsrc in -bad

As stated in commit c2956036b8da4b8f22a63a4f5a254be03e870aa6 in -bad,
the wasapi elements are now better than directsound, and should be
preferred if they are available.

For a later release, once the elements have more testing, we can
consider moving them to -good.

6 years agomatroskamux: Only mark new clusters as keyframe if they start on a keyframe or we...
Sebastian Dröge [Wed, 28 Feb 2018 17:21:53 +0000 (19:21 +0200)]
matroskamux: Only mark new clusters as keyframe if they start on a keyframe or we're muxing only audio

Based on a patch by Nicola Murino <nicola.murino@gmail.com>

https://bugzilla.gnome.org/show_bug.cgi?id=792775

6 years agomatroskamux: Clip maximum cluster duration to the maximum possible value
Sebastian Dröge [Wed, 28 Feb 2018 17:19:10 +0000 (19:19 +0200)]
matroskamux: Clip maximum cluster duration to the maximum possible value

Only up to timescale * G_MAXINT16 is possible as cluster duration, which
is already higher than our default value. Using higher values would
cause overflows and broken files.

Based on the investigation by Nicola Murino <nicola.murino@gmail.com>

https://bugzilla.gnome.org/show_bug.cgi?id=792775

6 years agomatroska-mux: Refuse caps changes after starting to write headers
Sebastian Dröge [Mon, 26 Feb 2018 11:03:59 +0000 (13:03 +0200)]
matroska-mux: Refuse caps changes after starting to write headers

Matroska does not support changing the stream type and stream properties
after the headers were started to be written, and for example H264
codec_data changes can't be supported.

https://bugzilla.gnome.org/show_bug.cgi?id=782949

6 years agotests: fix redenc tests
Mathieu Duponchelle [Tue, 27 Feb 2018 15:33:53 +0000 (16:33 +0100)]
tests: fix redenc tests

The default of the allow-no-red-blocks property was changed in a
previous commit, thus breaking the test assumptions

6 years agortp: fix another debug log printf format warning on 32-bit systems
Tim-Philipp Müller [Tue, 27 Feb 2018 13:13:49 +0000 (13:13 +0000)]
rtp: fix another debug log printf format warning on 32-bit systems

rtpulpfeccommon.c:432:27: error: format ‘%lx’ expects argument of type
‘long unsigned int’, but argument 10 has type ‘guint64 {aka long long unsigned int}’

https://bugzilla.gnome.org/show_bug.cgi?id=793732

6 years agortpptdemux: provide example usage for ignored-payload-types
Mathieu Duponchelle [Mon, 26 Feb 2018 16:02:52 +0000 (17:02 +0100)]
rtpptdemux: provide example usage for ignored-payload-types

6 years agortpbin, rtpptdemux: Add missing Since markers
Mathieu Duponchelle [Mon, 26 Feb 2018 15:53:08 +0000 (16:53 +0100)]
rtpbin, rtpptdemux: Add missing Since markers

6 years agoFEC elements: document, remove irrelevant properties
Mathieu Duponchelle [Mon, 26 Feb 2018 14:57:28 +0000 (15:57 +0100)]
FEC elements: document, remove irrelevant properties

The ulpfecenc "mux-seq" and "ssrc" properties were initially added
because the element did more than implement ULPFEC. As it was
decided that FLEXFEC would be implemented in a separate element,
both properties are now unneeded and confusing.

Change the default for the ulpfecenc multi-packet property,
as it is expected that most users of this element will be protecting video
streams.

Change the default property for the rtpredenc allow-no-red-blocks
property, as it should also be its default mode of operation.

https://bugzilla.gnome.org/show_bug.cgi?id=793843

6 years agortpgstdepay: do not warn when caps were not yet received
Mathieu Duponchelle [Sat, 24 Feb 2018 19:05:05 +0000 (20:05 +0100)]
rtpgstdepay: do not warn when caps were not yet received

It is expected that when connecting to a stream that has
already started, the caps will only arrive at the interval
specified on rtpgstpay, we shouldn't be warning as this is
a normal mode of operation.

https://bugzilla.gnome.org/show_bug.cgi?id=793798

6 years agortpulpfec: fix debug log printf format warning on 32-bit platforms
Arnaud Bonatti [Thu, 22 Feb 2018 20:53:40 +0000 (21:53 +0100)]
rtpulpfec: fix debug log printf format warning on 32-bit platforms

https://bugzilla.gnome.org/show_bug.cgi?id=793732

6 years agodocs: hook up new RTP FEC elements
Tim-Philipp Müller [Thu, 22 Feb 2018 14:58:12 +0000 (14:58 +0000)]
docs: hook up new RTP FEC elements

https://bugzilla.gnome.org/show_bug.cgi?id=792696

6 years agodocs: update for git master
Tim-Philipp Müller [Thu, 22 Feb 2018 14:57:58 +0000 (14:57 +0000)]
docs: update for git master

6 years ago.gitignore more test binaries
Tim-Philipp Müller [Thu, 22 Feb 2018 10:54:02 +0000 (10:54 +0000)]
.gitignore more test binaries

6 years agotests: also dist new fec test header file
Tim-Philipp Müller [Wed, 21 Feb 2018 20:46:10 +0000 (20:46 +0000)]
tests: also dist new fec test header file

6 years agortp: dist new header files
Tim-Philipp Müller [Wed, 21 Feb 2018 20:44:26 +0000 (20:44 +0000)]
rtp: dist new header files

Fixes make distcheck

6 years agortp: fec: fix build with gstreamer debug log system disabled
Tim-Philipp Müller [Wed, 21 Feb 2018 18:52:44 +0000 (18:52 +0000)]
rtp: fec: fix build with gstreamer debug log system disabled

6 years agortpptdemux: do no assume sink caps are non NULL
Mathieu Duponchelle [Wed, 21 Feb 2018 18:59:04 +0000 (19:59 +0100)]
rtpptdemux: do no assume sink caps are non NULL

6 years agocheck: Fix ulpfec test build
Mathieu Duponchelle [Wed, 21 Feb 2018 17:51:17 +0000 (18:51 +0100)]
check: Fix ulpfec test build

The test name was updated but not the build definition

6 years agortpbin: Expose FEC support signals
Mathieu Duponchelle [Tue, 28 Nov 2017 05:02:05 +0000 (06:02 +0100)]
rtpbin: Expose FEC support signals

Also slightly refactor complete_session_src

https://bugzilla.gnome.org/show_bug.cgi?id=792696

6 years agortp: Implement ULPFEC (RFC 5109)
Mikhail Fludkov [Fri, 17 Nov 2017 02:52:03 +0000 (03:52 +0100)]
rtp: Implement ULPFEC (RFC 5109)

We expose a set of new elements:

* ULPFEC encoder / decoder
* A storage element, which should be placed before jitterbuffers,
  and is used to store packets in order to attempt reconstruction
  after the jitterbuffer has sent PacketLost events
* RED encoder / decoder (RFC 2198), these are necessary to
  use FEC in webrtc, as browsers will propose and expect ulpfec
  packets to be wrapped in red packets

With contributions from:

Mathieu Duponchelle <mathieu@centricular.com>
Sebastian Dröge <sebastian@centricular.com>

https://bugzilla.gnome.org/show_bug.cgi?id=792696

6 years agortpptdemux: Add ignored-payload-types property
Mathieu Duponchelle [Tue, 28 Nov 2017 00:11:54 +0000 (01:11 +0100)]
rtpptdemux: Add ignored-payload-types property

Packets with these payload types will be dropped. A use case
for this is FEC, where we want FEC packets to go through the
jitterbuffer, but not be output by rtpbin.

https://bugzilla.gnome.org/show_bug.cgi?id=792696

6 years agortpptdemux: Add ssrc to output caps
Mathieu Duponchelle [Mon, 20 Nov 2017 17:08:38 +0000 (18:08 +0100)]
rtpptdemux: Add ssrc to output caps

It may be useful downstream

https://bugzilla.gnome.org/show_bug.cgi?id=792696

6 years agogtk: fix compiler warning with recent glib
Arnaud Bonatti [Wed, 21 Feb 2018 10:12:10 +0000 (11:12 +0100)]
gtk: fix compiler warning with recent glib

https://bugzilla.gnome.org/show_bug.cgi?id=793688

6 years agoqt: don't use libEGL functions when we don't link to libEGL
Matthew Waters [Wed, 21 Feb 2018 00:35:33 +0000 (11:35 +1100)]
qt: don't use libEGL functions when we don't link to libEGL

Use the provided wrapper available from libgstgl.

https://bugzilla.gnome.org/show_bug.cgi?id=793547

6 years agomonoscope: Forward the SEGMENT event from the chain function
Sebastian Dröge [Sun, 18 Feb 2018 20:38:13 +0000 (21:38 +0100)]
monoscope: Forward the SEGMENT event from the chain function

Otherwise we'll break the event order and forward the SEGMENT event
before sending a CAPS event.

6 years agortspsrc: Fix missing read property of backchannel
James Stevenson [Fri, 16 Feb 2018 12:25:29 +0000 (12:25 +0000)]
rtspsrc: Fix missing read property of backchannel

Add missing read property code for backchannel

https://bugzilla.gnome.org/show_bug.cgi?id=793507

6 years agoexamples: rtsp: fix meson build take 2
Tim-Philipp Müller [Fri, 16 Feb 2018 09:42:59 +0000 (09:42 +0000)]
examples: rtsp: fix meson build take 2

6 years agortsp: Fix meson.build of the example
Sebastian Dröge [Fri, 16 Feb 2018 09:30:01 +0000 (11:30 +0200)]
rtsp: Fix meson.build of the example

6 years agortspsrc: Implement ONVIF backchannel support via TCP
Sebastian Dröge [Fri, 26 Jan 2018 14:33:21 +0000 (16:33 +0200)]
rtspsrc: Implement ONVIF backchannel support via TCP

6 years agortspsrc: Implement ONVIF backchannel support
Nirbheek Chauhan [Fri, 13 Oct 2017 15:05:54 +0000 (18:05 +0300)]
rtspsrc: Implement ONVIF backchannel support

Set backchannel=onvif to enable, and use the 'push-backchannel-sample'
action signal with the correct stream id.

6 years agosplitmuxsrc: Improve not-linked handling.
Jan Schmidt [Thu, 15 Feb 2018 14:49:57 +0000 (01:49 +1100)]
splitmuxsrc: Improve not-linked handling.

Don't report not-linked unless all pads have
returned not-linked.

6 years agoBack to development
Tim-Philipp Müller [Thu, 15 Feb 2018 19:44:19 +0000 (19:44 +0000)]
Back to development

6 years agoRelease 1.13.1
Tim-Philipp Müller [Thu, 15 Feb 2018 17:06:10 +0000 (17:06 +0000)]
Release 1.13.1

6 years agodocs: update plugin docs
Tim-Philipp Müller [Thu, 15 Feb 2018 17:05:23 +0000 (17:05 +0000)]
docs: update plugin docs

6 years agopo: update translations
Tim-Philipp Müller [Thu, 15 Feb 2018 13:32:20 +0000 (13:32 +0000)]
po: update translations

6 years agortpjitterbuffer: allow timestamps to move backwards
Patrick Radizi [Wed, 14 Feb 2018 15:38:07 +0000 (16:38 +0100)]
rtpjitterbuffer: allow timestamps to move backwards

The original solution for #784002 incorrectly assumed that timestamps
may not move backwards and changed timestamps that did so.

https://bugzilla.gnome.org/show_bug.cgi?id=784002

6 years agodocs: remove pointless Since: 0.10.x markers
Tim-Philipp Müller [Thu, 15 Feb 2018 00:58:38 +0000 (00:58 +0000)]
docs: remove pointless Since: 0.10.x markers

6 years agortpvorbisdepay: fix unbounded memory usage
Alban Bedel [Wed, 27 Sep 2017 14:01:35 +0000 (16:01 +0200)]
rtpvorbisdepay: fix unbounded memory usage

All received configurations are parsed and added to a list, this lead
to an unbounded memory usage. As the configuration is resent every
second this quickly lead to a large memory usage.

Add a check to only add the config if it is not already available in
the list. This fix only handle the typical case of a well behaved
stream, a malicious server could still send many useless
configurations to raise the client memory usage.

6 years agodocs: add qt plugin
Tim-Philipp Müller [Mon, 12 Feb 2018 18:41:41 +0000 (18:41 +0000)]
docs: add qt plugin

https://bugzilla.gnome.org/show_bug.cgi?id=754094

6 years agoqt: hook up to build
Tim-Philipp Müller [Mon, 12 Feb 2018 18:34:16 +0000 (18:34 +0000)]
qt: hook up to build

https://bugzilla.gnome.org/show_bug.cgi?id=754094

6 years agoMove qt plugin from -bad
Tim-Philipp Müller [Mon, 12 Feb 2018 18:13:17 +0000 (18:13 +0000)]
Move qt plugin from -bad

https://bugzilla.gnome.org/show_bug.cgi?id=754094

6 years agoconfigure: fix build with --disable-external
Tim-Philipp Müller [Mon, 12 Feb 2018 15:44:35 +0000 (15:44 +0000)]
configure: fix build with --disable-external

6 years agodocs: add moved gtk plugin to docs
Tim-Philipp Müller [Sat, 10 Feb 2018 20:31:49 +0000 (20:31 +0000)]
docs: add moved gtk plugin to docs

6 years agodocs: update for git master
Tim-Philipp Müller [Sat, 10 Feb 2018 20:28:46 +0000 (20:28 +0000)]
docs: update for git master

6 years agogtk: hook up to meson build
Tim-Philipp Müller [Mon, 12 Feb 2018 11:02:12 +0000 (11:02 +0000)]
gtk: hook up to meson build

6 years agogtk: hook up to autotools build
Tim-Philipp Müller [Sat, 10 Feb 2018 13:20:43 +0000 (13:20 +0000)]
gtk: hook up to autotools build

6 years agoMove gtk plugin from -bad
Tim-Philipp Müller [Sat, 10 Feb 2018 12:49:36 +0000 (12:49 +0000)]
Move gtk plugin from -bad

https://bugzilla.gnome.org/show_bug.cgi?id=754094

6 years agoqtdemux: Fix comment typo in previous commit
Edward Hervey [Fri, 9 Feb 2018 10:26:56 +0000 (11:26 +0100)]
qtdemux: Fix comment typo in previous commit

6 years agoqtdemux: More 'meta' atom parsing fixes
Edward Hervey [Fri, 9 Feb 2018 10:20:38 +0000 (11:20 +0100)]
qtdemux: More 'meta' atom parsing fixes

Turns out everybody is doing it their own way, so peek into the
meta atom itself to figure out which spec it is following

6 years agoqtmux: Add support for muxing svmi atom for stereoscopic video information
Sebastian Dröge [Fri, 2 Feb 2018 11:51:49 +0000 (13:51 +0200)]
qtmux: Add support for muxing svmi atom for stereoscopic video information

https://bugzilla.gnome.org/show_bug.cgi?id=793120

6 years agoqtdemux: Minor cleanup
Edward Hervey [Fri, 9 Feb 2018 07:59:56 +0000 (08:59 +0100)]
qtdemux: Minor cleanup

Just move variables to the blocks where they are used.

That function is massive, could do with some splitting up for
readability :(

6 years agoqtdemux: Cope with difference between QTFF and ISO BMFF specs
Edward Hervey [Fri, 9 Feb 2018 07:54:05 +0000 (08:54 +0100)]
qtdemux: Cope with difference between QTFF and ISO BMFF specs

The 'meta' atom is defined differently in QTFF and BMFF, so try
to guess which spec the current stream applies to by looking
at the major file type.

6 years agoisomp4: Make 'hdlr' atom dump more flexible
Edward Hervey [Fri, 9 Feb 2018 07:35:52 +0000 (08:35 +0100)]
isomp4: Make 'hdlr' atom dump more flexible

The smallest possible is 24 (and not 25) bytes.

The last "name" field can according to QTFF specifications not be present
at all. The parser will handle this fine and so will the rest of
the qtdemux code.

6 years agoUpdate ORC files
Edward Hervey [Fri, 9 Feb 2018 07:35:25 +0000 (08:35 +0100)]
Update ORC files

6 years agomeson: make version numbers ints and fix int/string comparison
Tim-Philipp Müller [Thu, 8 Feb 2018 19:09:45 +0000 (19:09 +0000)]
meson: make version numbers ints and fix int/string comparison

WARNING: Trying to compare values of different types (str, int).
The result of this is undefined and will become a hard error
in a future Meson release.