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The comment referring to rtph264pay is from 2015 and is out of date.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
dabeed52a9
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/1011>
- 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>
- 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>
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>
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>
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>
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>
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>
* 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>
Seek events are sent upstream on each sink, so if we receive multiple
seeks with the same seqnum, we must only perform one seek, not N seeks
where N = the number of sinks in the pipeline connected to rtspsrc.
This is the same thing done by demuxers like qtdemux or matrsokademux.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/938>
This code can be fragile, since it is very exacting in the timestamps
that it will accept. Add more logging so it's easier to debug issues
and figure out whether it's a bug in the calculation or something
wrong in the incoming buffers.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/934>
gst_rtp_sbc_pay_flush_buffers() is a misleading name. A better name would
be gst_rtp_sbc_pay_drain_buffers(), because that's what it does, it drains
any leftover queued data and pushes it downstream. "Flushing" in GStreamer
typically means to throw away any queued data and not process/push it
downstream.
Signed-off-by: Michal Dzik <michal.dzik@streamunlimited.com>
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/700>
GstAdapter must be flushed in some cases (flush, new segment, state change)
Without it, it may, for example, push some leftover buffer from old
segment in new segment. This, in general, breaks timestamps.
See GstAdapter documentation for more.
Signed-off-by: Michal Dzik <michal.dzik@streamunlimited.com>
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/merge_requests/700>
