Matthew Waters [Wed, 14 Oct 2020 13:28:36 +0000 (00:28 +1100)]
isomp4/mux: fixes for fragmented mp4 output
Various buffer offset calculations were not quite correct in all cases.
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/issues/866
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1060>
Mathieu Duponchelle [Mon, 9 Aug 2021 22:53:57 +0000 (00:53 +0200)]
matroska-mux: support H264 avc3 / H265 hev1
The matroska codec specs is unfortunately vague on the subject,
stating for H264:
AVC/H.264 stored as described in [@!ISO.14496-15]
and for H265:
HEVC/H.265 stored as described in [@!ISO.14496-15]
This spec however specifies multiple stream formats, our
implementation has opted for interpreting this as avc1 / hvc1,
both of which disallow in-band SPS.
Most decoders however will support in-band SPS / PPS, and
this commit gives the option to explicitly mux in avc3 / hev1,
which allows changing stream parameters on the fly, that is
useful for smart encoding for example.
When either of these stream formats are picked as the input,
changes in codec_data / tier / level / profile do not cause
renegotiation failure, a warning is logged however as it isn't
clear how compliant such a stream is.
The stream-format field is correctly ordered in the template
caps to avoid selecting potentially non-compliant options on
automatic negotiation.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1047>
Mathieu Duponchelle [Mon, 9 Aug 2021 22:51:36 +0000 (00:51 +0200)]
isomp4/qtmux: allow renegotiating when tier / level / profile change
Those are carried either in codec_data or in-band SPS (for avc3),
and it is OK for those to change, though decoders obviously need
to support it.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1047>
Mathieu Duponchelle [Fri, 6 Aug 2021 21:36:48 +0000 (23:36 +0200)]
isomp4/qtmux: accept video/x-h264, stream-format=avc3
The main difference between avc1 and avc3 is that avc3 is allowed
to contain in-band SPS / PPS. In practice decoders will always use
in-band parameter sets anyway, but it is cleaner to explicitly
advertise it.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1047>
Mathieu Duponchelle [Fri, 6 Aug 2021 20:59:23 +0000 (22:59 +0200)]
isomp4/qtmux: make sure to switch to next chunk on new caps
For example, with single video sink pad, and new codec_data is
received, current_chunk_offset must be reset to -1 for the
aggregate loop to open a new chunk.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1047>
Mathieu Duponchelle [Fri, 6 Aug 2021 20:55:32 +0000 (22:55 +0200)]
isomp4/atoms: fix multiple stsd entries
stsd entries are serialized in reverse order (starting from
g_list_last()), and must be prepended to the entry list for their
index to be correct when referenced from stsc entries.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1047>
Arun Raghavan [Thu, 12 Aug 2021 15:03:58 +0000 (11:03 -0400)]
matroska-mux: Add a timestamp-offset property
Adds a user-controllable timestamp offset to clusters and blocks. This
should be useful if we want to have timestamps that have significance
outside of the current file (for example, we might set the offset to the
wallclock when the file is being created, or some other common base, if
we want to correlate streams across multiple files).
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1051>
Stéphane Cerveau [Thu, 15 Jul 2021 10:02:40 +0000 (12:02 +0200)]
matroska: demux: update stream_start_time
The stream_start_time can be less than the first detected.
In case of B-Frame based media, the first frame PTS might be
greater than the next one.
Need to keep the segment.start if a seek has been performed.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1030>
Nicolas Dufresne [Tue, 17 Aug 2021 20:08:33 +0000 (16:08 -0400)]
mastrokademux: Remove redundant assignment
The segment.position is unconditionnaly set few lines below.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1030>
Víctor Manuel Jáquez Leal [Tue, 17 Aug 2021 14:49:47 +0000 (16:49 +0200)]
videocrop: Fix icles tests.
Internally videcrop can call gst_video_crop_set_info() with NULL as in
caps. Then critical messages are raised when the in caps are
processed.
To fix this the in caps are checked, and if they are present, its
capsfeature is extracted, otherwise, the previous raw caps detection
remains as before.
Also the videocrop-test removes the format field in the structure
because now its always passed.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1056>
Jakub Adam [Fri, 8 Jan 2021 16:34:02 +0000 (17:34 +0100)]
rtp: Color Space header extension
Implements WebRTC header extension defined in
http://www.webrtc.org/experiments/rtp-hdrext/color-space.
It uses RTP header to communicate color space information and optionally
also metadata that is needed in order to properly render a high dynamic
range (HDR) video stream.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/853>
Hou Qi [Mon, 9 Aug 2021 02:46:30 +0000 (10:46 +0800)]
v4l2: Add protection when set decoder capture fps accroding to output fps
Some v4l2 drivers don't have the capacity to change framerate. There is
chance to make decoder capture fps to be 0/0 if numerator and denominator
returned by G_PARM ioctl are both 0. It causes critical warning
"passed '0' as denominator for `GstFraction'".
In order to fix this, add protection when set decoder capture fps according
to output fps.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1048>
Per Förlin [Tue, 1 Jun 2021 13:33:01 +0000 (15:33 +0200)]
rtspsrc: Add support to ignore x-server HEADER reply
When connecting to an RTSP server in tunnled mode (HTTP) the server
usually replies with a x-server header. This contains the address
of the intended streaming server. However some servers return an
"invalid" address. Here follows two examples when it might happen.
1. A server use Apache combined with a separate RTSP process to handle
Https request on port 443. In this case Apache handle TLS and
connects to the local RTSP server, which results in a local
address 127.0.0.1 or ::1 in the x-server reply. This address is
returned to the actual RTSP client in the x-server header.
The client will receive this address and try to connect to it
and fail.
2. The client use a ipv6 link local address with a specified scope id
fe80::aaaa:bbbb:cccc:dddd%eth0 and connects via Http on port 80.
The RTSP server receives the connection and returns the address
in the x-server header. The client will receive this address and
try to connect to it "as is" without the scope id and fail.
In the case of streaming data from RTSP servers like 1. and 2. it's
useful to have the option to simply ignore the x-server header reply
and continue using the original address.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1007>
Dmitry Shusharin [Wed, 4 Aug 2021 05:33:06 +0000 (12:33 +0700)]
gstqmlgl: fix indent
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1032>
Dmitry Shusharin [Fri, 30 Jul 2021 09:52:23 +0000 (16:52 +0700)]
gstqmlgl: wrap raw GstGLContext into GWeakRef
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1032>
Dmitry Shusharin [Fri, 30 Jul 2021 09:32:13 +0000 (16:32 +0700)]
gstqmlgl: add multisink test application
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1032>
Dmitry Shusharin [Fri, 30 Jul 2021 10:21:46 +0000 (17:21 +0700)]
gstqmlgl: refactoring: rename ambiguous variables, clean up unused and duplicated ones
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1032>
Dmitry Shusharin [Fri, 30 Jul 2021 10:20:59 +0000 (17:20 +0700)]
gstqmlgl: rework WGL-specific context init code
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1032>
Dmitry Shusharin [Fri, 30 Jul 2021 10:20:49 +0000 (17:20 +0700)]
gstqmlgl: retrieve correct device bound to current GL context (+ minor code cleanup)
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1032>
Dmitry Shusharin [Fri, 30 Jul 2021 10:20:25 +0000 (17:20 +0700)]
gstqmlgl: correct validation for Qt GL context
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1032>
Dmitry Shusharin [Fri, 30 Jul 2021 10:20:07 +0000 (17:20 +0700)]
gstqmlgl: create helper QRunnable-based class for render jobs
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1032>
Tulio Beloqui [Fri, 6 Aug 2021 14:25:02 +0000 (16:25 +0200)]
rtpjitterbuffer: fixed stall on gap when using rtx
Co-authored-by: Håvard Graff <havard@pexip.com>
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1055>
Nirbheek Chauhan [Fri, 13 Aug 2021 14:02:53 +0000 (19:32 +0530)]
flv: use g_memdup2() as g_memdup() is deprecated
g_memdup() is deprecated since GLib 2.68 and we want to avoid
deprecation warnings with recent versions of GLib.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1052>
Sebastian Dröge [Sun, 15 Aug 2021 09:26:38 +0000 (12:26 +0300)]
souphttpsrc: Always use the content decoder but set `Accept-Encoding: identity` if no compression should be used
Some servers respond with gzip-encoded responses regardless of whether
the request allowed it to be used in the response. By always having the
content decoder enabled, these invalid responses can be decoded
correctly while for well-behaving servers the `compress` property
selects between allowing compressed responses or not.
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/issues/833
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1053>
Matthew Waters [Thu, 12 Aug 2021 12:57:01 +0000 (22:57 +1000)]
qt: always update the sink_retrieved flag when the sink retrieves
Fixes a case where adding a qmlgloverlay element after an existing
qmlglsink elements was already in the pipeline would create an entirely
separate GstGLDisplay pointing to the same underlying display resource.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1050>
Víctor Manuel Jáquez Leal [Wed, 11 Aug 2021 12:52:52 +0000 (14:52 +0200)]
videocrop: Resurrect logging category.
Fix for a regression from commit
8f1384c9. That commit moved the debug
category definition, as static, into a gstvideocropelement.c, but that
category was used as default, in gstvideocrop.c, so it was never used
at logging, so the debug selector never showed the logs for
videocrop.
This patch move back the category definition into gstvideocrop.c and
leaving the function videocrop_element_init() as a noop.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1049>
Seungha Yang [Sat, 31 Jul 2021 14:14:34 +0000 (23:14 +0900)]
jpeg: Add support for meson fallback
Allow building jpeg plugin by using meson fallback
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1041>
Hou Qi [Tue, 27 Jul 2021 02:43:21 +0000 (10:43 +0800)]
v4l2: Keep decoder capture fps same as output fps if it's not set
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1035>
Seungha Yang [Tue, 27 Jul 2021 09:33:18 +0000 (18:33 +0900)]
jack: Add port-names property to select ports explicitly
By this new property, user can select physical port to connect,
and element will pick requested port instead of random ones.
User should provide full port name including "client_name:" prefix.
An example is
jackaudiosrc port-names="system:capture_1,system:capture_3" ! ...
jackaudiosink port-names="system:playback_2"
In addition to "port-names" property, a new connect type "explicit"
is added so that element can post error message if requested
"port-names" contains invalid port(s).
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1037>
Kai Uwe Broulik [Fri, 23 Jul 2021 09:04:00 +0000 (11:04 +0200)]
qt: Support RGB format
In GstQSGTexture::hasAlphaChannel return value based on
whether the video format has alpha channel.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1040>
Seungha Yang [Mon, 26 Jul 2021 11:14:32 +0000 (20:14 +0900)]
jack: Add low-latency property for automatic latency-optimized setting
Similar to wasapi/wasapi2 plugins on Windows, adding low-latency
option so that jack element can optimize GstAudioRingBufferSpec
setting for low latency.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1034>
Seungha Yang [Mon, 26 Jul 2021 10:55:25 +0000 (19:55 +0900)]
jack: Remove trailing whitespace
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1034>
Víctor Manuel Jáquez Leal [Tue, 27 Jul 2021 15:58:15 +0000 (17:58 +0200)]
videocrop: Resurrect any caps feature negotiation.
Commit
e31cbce4 brought a regression to negotiate featured caps. But
only by removing the entry in the caps template. This commit brings it
back.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1039>
Seungha Yang [Mon, 26 Jul 2021 09:43:04 +0000 (18:43 +0900)]
jack: Fix assertion fail when device supports only mono channel
MAX should be larger than MIN for GST_TYPE_INT_RANGE.
GStreamer-CRITICAL **: 18:26:27.912:
gst_value_collect_int_range: assertion 'collect_values[0].v_int < collect_values[1].v_int' failed
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1033>
Matthew Waters [Wed, 21 Jul 2021 10:14:46 +0000 (20:14 +1000)]
qmlglsrc: fix operation without any qmlglsink
E.g. a pipeline like qmlglsrc ! gldownload ! ... would currently fail to
run because the OpenGL context are not created in the correct order.
The QtWindow also needs to know the OpenGL context used by downstream
elements in order to set optimize for the correct GstGLSyncMeta for
synchonisation purposes.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1036>
Jan Schmidt [Mon, 26 Jul 2021 07:55:24 +0000 (17:55 +1000)]
splitmuxsink: Fix some reference leaks in error cases.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1023>
Jan Schmidt [Wed, 7 Jul 2021 14:12:52 +0000 (00:12 +1000)]
splitmuxsink: Prevent hang going back to NULL after failures
Prevent a condition where splitmuxsink won't go back to NULL state
after a child element fails to change state by making sure that
a READY->READY state change doesn't fail, and by returning
GST_FLOW_ERROR or GST_FLOW_FLUSHING upstream to shut down streaming
as quickly as possible.
This can happen after (for example) setting an invalid filename
on the sink element. In that case, the READY->PAUSED transition
fails, but with internal elements still in the NULL state. Trying
to set splitmuxsink back to NULL then ends up trying to bring
those NULL elements up to READY with a READY->READY transition,
(which fails, prevent splitmuxsink from getting to NULL)
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1023>
Mathieu Duponchelle [Mon, 12 Jul 2021 23:27:45 +0000 (01:27 +0200)]
deinterlace: reduce noise when gst_pad_set_caps fails
It may be that downstream is simply flushing, in which case logging
an error is misleading.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1029>
Mathieu Duponchelle [Thu, 8 Jul 2021 00:22:20 +0000 (02:22 +0200)]
splitmuxsink: always use factory property when set
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1024>
Yacine Bandou [Mon, 21 Jun 2021 11:47:50 +0000 (13:47 +0200)]
qtdemux: No need for new "application/x-cbcs" caps
Instead of using the new "application/x-cbcs" caps, we are just adding
a new structure field "ciphe-mode", to indicate which encryption scheme
is used: "cenc", "cbcs", "cbc1" or "cens".
Similarly for the protection metadata, we add the "cipher-mode" field
to specify the encryption mode with which the buffers are encrypted.
"cenc": AES-CTR (no pattern)
"cbc1": AES-CBC (no pattern)
"cens": AES-CTR (pattern specified)
"cbcs": AES-CBC (pattern specified, using a constant IV)
Currently only "cenc" and "cbcs" are supported.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1013>
Philippe Normand [Mon, 5 Jul 2021 15:12:57 +0000 (16:12 +0100)]
qt: Fix clang build
The updatePaintNode method is part of the QQuickItem class interface, so needs
to be flagged as overriding the default implementation.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/567>
Philippe Normand [Wed, 15 Apr 2020 09:38:04 +0000 (10:38 +0100)]
qt: Add navigation events support
Currently handles only mouse events.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/567>
Philippe Normand [Wed, 15 Apr 2020 09:33:22 +0000 (10:33 +0100)]
gtk: Scroll events dispatch support
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/567>
Jan Schmidt [Thu, 1 Jul 2021 15:41:05 +0000 (01:41 +1000)]
matroskamux: Always write a tags element into seekhead
If there are only stream tags, we still want to write the
tags entry into the seekhead, so that tags can be found
quickly in the player.
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/issues/905
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1020>
Seungha Yang [Wed, 30 Jun 2021 14:52:26 +0000 (23:52 +0900)]
qtmux: Don't need to update track per GstCaps if it's not changed
Skip GstQTMuxPad::set_caps() call for duplicated caps.
All the processing done in set_caps() method for duplicated caps
are redundant.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1019>
Sebastian Dröge [Thu, 1 Jul 2021 10:18:45 +0000 (13:18 +0300)]
rtpssrcdemux: Remove pads and reset the element also in READY->NULL
Mostly for completeness.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1018>
Sebastian Dröge [Thu, 1 Jul 2021 10:18:09 +0000 (13:18 +0300)]
rtpptdemux: Remove pads also in PAUSED->READY
They're based on per-stream information and that should be reset
whenever going to READY state.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1018>
Jakub Adam [Tue, 16 Feb 2021 15:39:34 +0000 (16:39 +0100)]
docs: update plugins cache for vp9enc
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/874>
Jakub Adam [Fri, 9 Apr 2021 17:22:29 +0000 (19:22 +0200)]
vpx: add enum for adaptive quantization modes
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/874>
Jakub Adam [Tue, 16 Feb 2021 12:28:00 +0000 (13:28 +0100)]
vp9enc: expose frame-parallel-decoding property
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/874>
Jakub Adam [Tue, 16 Feb 2021 11:57:55 +0000 (12:57 +0100)]
vp9enc: expose aq-mode property
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/874>
Seungha Yang [Sat, 26 Jun 2021 11:00:03 +0000 (20:00 +0900)]
multiudpsink: Fix broken SO_SNDBUF get/set on Windows
SO_SNDBUF has been undefined on Windows because of missing WinSock2.h
include. And don't use native socket functions (e.g., setsockopt())
if code is expected to be built on Windows. We don't link ws2_32.lib
for this plugin.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1016>
Olivier Crête [Thu, 24 Jun 2021 18:57:14 +0000 (14:57 -0400)]
rtpmanager: Access GstRTPHdrExt fields through accessor
This way, the implementation can be private.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1017>
Jan Schmidt [Tue, 22 Jun 2021 07:19:19 +0000 (17:19 +1000)]
qtdemux: Refuse seeks in BYTES format
If downstream tries to seek in BYTES format, don't pass that through
to upstream. The byte positions downstream requests won't make any
sense in the muxed stream. There might be other formats we want to
pass through to upstream, but BYTES is not one of them. If we get a
seeking query about BYTES format, refuse that too.
This fixes a situation where we're playing a fragmented mp4 over http
and qtdemux refuses the initial seek (in TIME format), but then
h264parse/baseparse send a seek in BYTES format and everything falls
apart.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1014>
Nirbheek Chauhan [Wed, 16 Jun 2021 11:00:59 +0000 (16:30 +0530)]
rtph265depay: update codec_data in caps regardless of format
Updating of codec_data in the caps is important to propagate changes
in sps/pps/vps via NALs. Without this, downstream does not renegotiate
when upstream changes resolution.
The comment referring to rtph264pay is from 2015 and is out of date.
rtph264pay stopped doing that in 2017 with commit
dabeed52a995d27e16eba9e4617e61eb0bcd44c4
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1011>
Jordan Petridis [Fri, 4 Jun 2021 10:56:05 +0000 (13:56 +0300)]
doc: update gst_plugins_cache.json
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1006>
Matthew Waters [Thu, 3 Jun 2021 10:33:45 +0000 (20:33 +1000)]
qtitem: don't potentially leak a large number of buffers
The only other place where these queued buffers are removed, is in
setCaps() but that is not called at all on shutdown so this list of
buffers could not be removed.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1004>
Nicolas Dufresne [Fri, 28 May 2021 13:54:12 +0000 (09:54 -0400)]
jpegenc: Remove arbitrary encoding size limitation
The encoder is happy to encode with sizes less then 16x16, so remove this
arbitrary limitation. This also fixes the fact the sink and src template caps
disagree.
Fixes #888
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/998>
Tim-Philipp Müller [Sun, 23 May 2021 14:42:38 +0000 (15:42 +0100)]
qtdemux: use g_memdup2() as g_memdup() is deprecated
- atom nodes/bytereader sizes are already checked
- palettes: are fixed/known size
g_memdup() is deprecated since GLib 2.68 and we want to avoid
deprecation warnings with recent versions of GLib.
Also use gst_buffer_new_memdup() instead of _wrapped(g_memdup(),..).
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/993>
Tim-Philipp Müller [Sun, 23 May 2021 00:28:11 +0000 (01:28 +0100)]
matroskademux: use g_memdup2() as g_memdup() is deprecated
- ebml-read: add some sanity checks when going from 64-bit
to 32-bit length
- matroska-ids: codec_data_size has been checked via
gst_ebml_read_binary(), is existing allocation.
- matroska-demux: alloc size is from existing allocations
g_memdup() is deprecated since GLib 2.68 and we want to avoid
deprecation warnings with recent versions of GLib.
Also use gst_buffer_new_memdup() instead of _wrapped(g_memdup(),..).
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/993>
Tim-Philipp Müller [Sat, 22 May 2021 18:39:32 +0000 (19:39 +0100)]
Use g_memdup2() where available and add fallback for older GLib versions
- png: alloc size variable is a png type that's always 32-bit
- vpx: alloc size based on existing allocation
- wavpack: alloc size based on existing allocation
- icles: gdkpixbufoverlay: trusted and hard-coded input data
- rtp tests: rtp-payloading, vp8, vp9, h264, h265: trusted and/or static input data
g_memdup() is deprecated since GLib 2.68 and we want to avoid
deprecation warnings with recent versions of GLib.
Also use gst_buffer_new_memdup() instead of _wrapped(g_memdup(),..)
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/993>
Tim-Philipp Müller [Tue, 1 Jun 2021 14:28:36 +0000 (15:28 +0100)]
Back to development
Tim-Philipp Müller [Mon, 31 May 2021 23:11:44 +0000 (00:11 +0100)]
Release 1.19.1
Tim-Philipp Müller [Sat, 29 May 2021 11:54:22 +0000 (12:54 +0100)]
rtpjpegpay: fix image corruption when compiled with MSVC on Windows
On Windows with MSVC, jpeg_header_size would end up 2 bytes larger
than it should be. This then leads to the first 2 bytes of the
actual jpeg image data to be dropped, because we think those
belong to the header, which results in an undecodable image when
reconstructed in the depayloader.
What happens is that when the compiler evaluates
jpeg_header_size = mem.offset + read_u16_and_inc_offset_by_2(&mem);
it actually uses the mem.offset value after it has been increased
by the function call on the right hand size of the equation.
From section 6.5 of the C99 spec:
3. The grouping of operators and operands is indicated by the syntax [74].
Except as specified later (for the function-call (), &&, ||, ?:, and
comma operators), the order of evaluation of subexpressions and the
order in which side effects take place are both unspecified.
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/issues/889
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/999>
Hou Qi [Tue, 25 May 2021 08:19:20 +0000 (16:19 +0800)]
v4l2videoenc: Set default latency if the frame duration is invalid
If the duration of the v4l2object is invalid, use default 25fps instead.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/996>
Seungha Yang [Tue, 25 May 2021 15:23:56 +0000 (00:23 +0900)]
deinterlace: Drop "field-order" field while transforming caps
Like other basetransform subclasses are doing, drop field
which can be converted by deinterlace.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/997>
Seungha Yang [Tue, 25 May 2021 11:10:34 +0000 (20:10 +0900)]
deinterlace: Drop field-order field if outputting progressive
Progressive with field-order doesn't make sense
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/997>
Havard Graff [Fri, 21 May 2021 12:19:29 +0000 (14:19 +0200)]
rtpssrcdemux: fix "data flow before segment event" crash
This crash could happen at any time a RTP and RTCP buffer arrived
simultaneously in ssrcdemux.
The problem was that sticky-event arriving while the rtp and rtcp pads
were being set up could arrive just too late to be included in the initial
forwarding.
The fix checks if the stickies have been sent on the srcpad about to be
pushed on, and if not sends them. It also blocks any stickes from
being forwarded *prior* to this happening, to avoid them arriving on
the srcpad multiple times.
Since the test loops 1000 times, this will make running under valgrind
take forever, so use the RUNNING_ON_VALGRIND variable to detect we
are running under valgrind, and reduce the loop-count to 2 in that case.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/992>
Havard Graff [Fri, 21 May 2021 16:45:17 +0000 (18:45 +0200)]
rtpssrcdemux: refactor destruction of GstRtpSsrcDemuxPads
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/992>
Havard Graff [Fri, 21 May 2021 16:30:28 +0000 (18:30 +0200)]
rtpssrcdemux: make naming consistent
Use plural for GstRtpSsrcDemuxPads, since it contains two pads, and
use the variable-name 'dpads' everywhere.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/992>
Tim-Philipp Müller [Sun, 23 May 2021 14:14:11 +0000 (15:14 +0100)]
wavparse: use g_strndup() for copying text data
So we don't rely on NUL terminators inside the data.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/994>
Tim-Philipp Müller [Sun, 23 May 2021 12:29:07 +0000 (13:29 +0100)]
wavparse: clean up adtl/note/labl chunk parsing
We were passing the size of the adtl chunk to the note/labl
sub-chunk parsing function, which means we may memdup lots of
data after the chunk string's NUL terminator that doesn't
really belong to it.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/994>
Tim-Philipp Müller [Sun, 23 May 2021 12:24:21 +0000 (13:24 +0100)]
wavparse: guard against overflow when comparing chunk sizes
Could be rewritten as lsize > (size - 8) a well, but the
extra check seems clearer. Doesn't look like it was problematic,
lsize wasn't actually used when parsing the sub-chunks.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/994>
Daniel Almeida [Fri, 21 May 2021 16:31:12 +0000 (13:31 -0300)]
doc: update gst_plugins_cache.json
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/991>
Stéphane Cerveau [Wed, 5 May 2021 11:20:04 +0000 (13:20 +0200)]
matroskademux: fix decoder glitches with H264 content
To avoid decoder starvation causing glitches on screen,
the demuxer shall clip only when the buffer is a key frame
and the lace time is greater than the stop time.
Fixes gst-editing-services#128
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/973>
Matthew Waters [Tue, 11 May 2021 10:41:38 +0000 (20:41 +1000)]
qml: don't use buffers that have invalid contents
If the GL context is not shareable, ignore it.
A future change may also not output the relevant output either.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/983>
Matthew Waters [Tue, 11 May 2021 10:38:52 +0000 (20:38 +1000)]
qml: also use the dummy texture when no buffer has been set
Fixes corrupted texture output when changing OpenGL display/contexts.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/983>
Nicolas Dufresne [Tue, 11 May 2021 21:20:00 +0000 (17:20 -0400)]
doc: Update cache for RGBP format addition
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/968>
Nicolas Dufresne [Fri, 23 Apr 2021 18:37:46 +0000 (14:37 -0400)]
matroskademux: Advertise codec-alpha in caps
This will be used to select the appropriate decoders. We also only attach the
GstVideoCodecAlphaMeta if the AlphaMode element is set, this is to stay on the
safe side and mimic what browsers (verified in Firefox and Chromium code) do.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/968>
Nicolas Dufresne [Mon, 22 Mar 2021 20:58:26 +0000 (16:58 -0400)]
matroskademux: Store alpha stream in VideoCodecAlphaMeta
This generalize the feature over using mini object quark data. If
that feature was Matroska specifc, using the new CustomMeta would have
been enough and arguably cleaner then QData, though it seems that
similar technique is use with AV1 Image Format (AVIF).
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/968>
Tim-Philipp Müller [Sat, 3 Dec 2016 14:27:57 +0000 (14:27 +0000)]
matroska-demux: extract VP8 alpha from BlockAdditionals
And put it on buffers as qdata (which is easier in this
case than a private custom meta because it can be picked
up easily in other modules).
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/968>
Matthew Waters [Mon, 3 May 2021 07:39:05 +0000 (17:39 +1000)]
qt: return a different GstGLDisplay object when the first sink requests
This allows the 'replace-gstreamer-opengl-context' context machinery to
correctly replace the OpenGL context used by the pipeline when the first
qmlglsink is added to the pipeline.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/982>
Jan Alexander Steffens (heftig) [Fri, 7 May 2021 09:16:47 +0000 (11:16 +0200)]
udpsrc: Plug leaks of saddr in error cases
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/977>
Jan Alexander Steffens (heftig) [Fri, 7 May 2021 09:16:21 +0000 (11:16 +0200)]
udpsrc: Whitespace
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/977>
Jan Alexander Steffens (heftig) [Thu, 6 May 2021 22:43:44 +0000 (00:43 +0200)]
deinterlace: Plug a method subobject leak
Changing the method would leak the previous method.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/976>
Nicolas Dufresne [Thu, 6 May 2021 19:04:42 +0000 (15:04 -0400)]
vp9enc: Add color range support
When setting the colorspace, we now clear the range to reduced range,
the default, and then we also set the range so the VP9 encoder encodes
the right information in the bitstream.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/975>
Nicolas Dufresne [Thu, 6 May 2021 18:51:31 +0000 (14:51 -0400)]
vp9enc: Move colorspace configuration in VP9 enc
This is not supported by VP8 and was causing a warning.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/975>
Nicolas Dufresne [Thu, 6 May 2021 18:48:36 +0000 (14:48 -0400)]
vpxdenc: Add a GstVideoCodecState to configure_encoder virtual
This will be needed to configure the VP9 specific colorimetry, which is
currently configured for VP8 casing warning.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/975>
Bastien Nocera [Wed, 5 May 2021 14:48:10 +0000 (16:48 +0200)]
gtk: Remove coordinates double-translation
Remove our own translation in the mouse event capture code, as that
translation will be done through the navigation interface.
Tested by resizing the window created by:
gst-launch-1.0 -v videotestsrc ! navigationtest ! glupload ! glcolorconvert ! tee name=t ! gtkglsink
and checking that the cursor follows the mouse as expected.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/974>
Bastien Nocera [Wed, 5 May 2021 12:28:15 +0000 (14:28 +0200)]
gtk: Translate navigation events coordinates
If the application passed down some pointer coordinates, translate those
from display coordinates to stream coordinates, so things work as
expected even if the video is resized.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/974>
Bastien Nocera [Wed, 5 May 2021 12:24:31 +0000 (14:24 +0200)]
gtk: Export _display_size_to_stream_size()
Export _display_size_to_stream_size() so that GstNavigation implementors
can translate from display coordinates to stream coordinates before
pushing the events upstream to the DVD source, for example.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/974>
David Fernandez [Mon, 26 Feb 2018 16:26:41 +0000 (17:26 +0100)]
matroska-mux: Change accepted caps width and height from [16, MAX] to [1, MAX]
There are cases where the video size might be less than 16x16.
This change allows the Matroska muxer to accept this cases.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/539>
François Laignel [Tue, 20 Apr 2021 20:08:23 +0000 (22:08 +0200)]
Use gst_element_request_pad_simple
Instead of the deprecated gst_element_get_request_pad.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/958>
Jan Schmidt [Thu, 29 Apr 2021 22:12:47 +0000 (08:12 +1000)]
qtmux: Make sure to write 64-bit STCO table when needed.
qtmux attempts to choose between writing a 32-bit stco chunk offset table
when it can, but switch to a 64-bit co64 table when file offsets go over
4GB.
This patch fixes a problem where the atom handling code was checking
mdat-relative offsets instead of the final file offset (computed by
adding the mdat position plus the mdat-relative offset) - leading to
problems where files with a size between 4GB and 4GB+offset-of-the-mdat
would write incorrect STCO tables with some samples having truncated
32-bit offsets.
Smaller files write STCO correctly, larger files would switch to
co64 and also output correctly.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/970>
Hou Qi [Thu, 22 Apr 2021 07:01:32 +0000 (15:01 +0800)]
v4l2object: Add interlace-mode back to caps for camera
skip_try_fmt_probes is set to TRUE for v4l2src to skip interlace-mode and
colorimetry when probe caps. gst_v4l2_object_set_format_full() will add
colorimetry back to caps when iterating over the negotiated caps. There is
one case that v4l2src is first in preview state then starts recording.
v4l2src caps will change with an additional interlace-mode structure after
renegotiation, then v4l2src needs to reset. But this camera driver can't
orphan buffer pool, it causes require buffer failed as streaming is still
in active state.
To fix this, also need to add interlace-mode back to caps for camera to
avoid reset.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/961>
Guillaume Desmottes [Fri, 2 Apr 2021 16:41:28 +0000 (18:41 +0200)]
rtpopuspay: set MARKER flag
Set MARKER flag on first buffer after DTX.
According to RFC 3551 section 4.1 the marker bit needs to be set on
"the first packet after a silence period during which packets have
not been transmitted contiguously".
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/967>
Guillaume Desmottes [Wed, 31 Mar 2021 09:18:30 +0000 (11:18 +0200)]
rtpopuspay: add DTX support
If enabled, the payloader won't transmit empty frames.
Can be tested using:
opusenc dtx=true bitrate-type=vbr ! rtpopuspay dtx=true
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/967>
Doug Nazar [Sat, 24 Apr 2021 15:15:50 +0000 (11:15 -0400)]
taglib: Update createFrame() to non-deprecated version.
ID3v2::FrameFactory::createFrame() versions not taking a Header have
been deprecated since v1.5 (Jan 2008).
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/965>
Havard Graff [Sun, 25 Apr 2021 00:16:45 +0000 (02:16 +0200)]
rtpjitterbuffer: fix divide-by-zero
The estimated packet-duration can sometimes end up as zero, and dividing
by that is never a good idea...
The test reproduces the scenario, and the fix is easy.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/966>
Havard Graff [Tue, 2 Jun 2020 17:38:33 +0000 (19:38 +0200)]
rtpjitterbuffer: clean up and improve missing packets handling
* Try to make variable and function names more clear.
* Add plenty of comments describing the logic step-by-step.
* Improve the logging around this, making the logs easier to read and
understand when debugging these issues.
* Revise the logic of packets that are actually beyond saving in doing
the following:
1. Do an optimistic estimation of which packets can still arrive.
2. Based on this, find which packets (and duration) are now hopelessly
lost.
3. Issue an immediate lost-event for the hopelessly lost and then add
lost/rtx timers for the ones we still hope to save, meaning that if
they are to arrive, they will not be discarded.
* Revise the use of rtx-delay:
Earlier the rtx-delay would vary, depending on the pts of the latest
packet and the estimated pts of the packet it being issued a RTX for,
but now that we aim to estimate the PTS of the missing packet accurately,
the RTX delay should remain the same for all packets.
Meaning: If the packet have a PTS of X, the delay in asked for a RTX
for this packet is always a constant X + delay, not a variable one.
* Finally ensure that the chaotic "check-for-stall" tests uses timestamps
that starts from 0 to make them easier to debug.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/952>