rtpgstpay: Don't push packets before the first input buffer is received

It's not possible to create a valid RTP timestamp for them, which would
cause a potentially very big RTP timestamp discontinuity between those
first packets (created from initial events) and the packet based on the
first input buffer.

As a side-effect, also simplify the packet aggregation code a bit and
work with only a single level of buffer lists.

Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/1157

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2250>

rtpbasepayload: always store input buffer meta before negotiation

The decision to store the input buffer depends on whether extensions
are to be added to the output buffer, I assume as an optimization.

This creates an issue for subclasses that call negotiate(), where
header_exts is actually populated, from their handle_buffer()
implementation: at chain time, no header extension has been negotiated
yet, which means that we don't add extensions to the first batch of
buffers that comes out.

Keep track of whether negotiate has been called (this is different
from the negotiated field) and always store the input buffer until
then. This fixes the issue while largely preserving the optimization.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2304>

tests/onvif: improve robustness

The previous iteration of the code was inferring the type of the
frame by looking at the overall size of the gst-payloaded packet.

It is more robust to actually parse the payload and look at the
actual data buffers it contains.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2303>

tests/onvif: don't push buffers outside segment

segment->stop is exclusive, so in reverse playback mode we do not
need to output a buffer at that position as it will simply get
clipped in basesink.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2303>

rtponviftimestamp: add extension data to all packets ..

regardless of whether they are input as individual buffers or
buffer lists.

The ONVIF specification requires all packets to hold the extension,
it makes no sense to behave differently when handling buffer lists.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2303>

cudaconvert: Enhance debugging when setting video info fails

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2300>

codecparsers: bitwriter: Calculate PicSizeInCtbsY when writing slice header.

 The slice->pps->PicWidthInCtbsY/PicHeightInCtbsY values are calculated when
 we parse the slice header. But they are not calculated in bitwriter and so
 may not have valid value when writing slice header.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2287>

codecparsers: bitwriter: Fix some coverity issues of H265 bitwriter.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2287>

videoscale: Fix the src video info error in transfer_colorimetry_from_input()

Pipeline such as:
gst-launch-1.0 -vf videotestsrc ! video/x-raw,format=NV12,colorimetry=\(string\)bt709 \
 ! videoscale ! video/x-raw,format=I420 ! fakesink
Always trigger a error:
ERROR             video-info video-info.c:556:gst_video_info_from_caps: no width property given

Because it is called before the fixate_size(), the src caps' resolution
may be absent or not fixed. That causes that the src video info can not
be created correctly and we can not inherit the colorimetry and chroma-site
from the input caps.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2289>

videorate: fix assertion when pushing last and only buffer without duration

Fixing this pipeline:
  gst-launch-1.0 filesrc location=sample.png ! pngdec ! videorate ! fakesink

- videorate receives a single buffer with pts = 0, duration = invalid;
- then it receives eos triggering this buffer to be pushed downstream;
- the pushing code was assuming that a duration was set, which is
  impossible as we received a single buffer and no output framerate was
  set either. So the best we can do is to push the buffer without
  duration.

Fix #1177

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2296>

va: dec: Use gst_buffer_pool_config_set_va_alignment() to set alignment.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2282>

va: pool: Replace all tabs with spaces in header file.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2282>

va: pool: Delete the GST_BUFFER_POOL_OPTION_VIDEO_ALIGNMENT.

The va pool is used for GPU side surface/image, its alignment should
not be changed arbitrarily by others. So we decide not to expose the
GST_BUFFER_POOL_OPTION_VIDEO_ALIGNMENT flag anymore.
Instead, user can call gst_buffer_pool_config_set_va_alignment() to
set its surface/image alignment.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2282>

va: pool: Add set_va_alignment() API.

We want to use gst_buffer_pool_config_set_va_alignment() to replace
gst_buffer_pool_config_get_video_alignment(). The later one is specific
for GST_BUFFER_POOL_OPTION_VIDEO_ALIGNMENT option.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2282>

rtptwcc: don't map the buffer twice

...and use the pt extracted rather than the one from RTPPacketInfo
when logging.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2271>

rtpsession: Emit "notify::stats" when we update stats from RR or SR

Sensibily optimizing caching the pspecs and using them directly

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2266>

oggdemux: Protect against invalid framerates

This check wasn't done for all mappings.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2295>

av1parse: Fix a latent memory leak in colorimetry setting.

Also delete the useless "have_cinfo" judgement.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2288>

av1parse: Correct the meaning of color_range flag.

According to spec:
color range equal to 0 shall be referred to as the studio swing
representation and color range equal to 1 shall be referred to as
the full swing representation.

The current status is just the opposite.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2288>

qsvencoder: Workaround for invalid DTS with VP9

SDK runtime seems to report zero DTS for all frames in case of VP9 but
we don't need to use DTS for the codec.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2291>

nvcodec: Stop building for non-{Windows,Linux} target

The other platforms, specifically macOS have not been supported already
because this plugin loads so or dll. Moreover, NVIDIA dropped
support for macOS as of CUDA 11.0. See also
https://developer.nvidia.com/nvidia-cuda-toolkit-11_0_0-developer-tools-mac-hosts

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2283>

srtp: add unit tests

Enable unit tests in meson.build
Add test_play_key_error to check the stats

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2027>

srtpdec: add counts in stats

In order to count the buffers which have been received and dropped for
decryption reason, add a stats to track it.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2027>

d3d11videosink: Fix for unhandled mouse double click events

Only window created with CS_DBLCLKS style can receive those mouse
double click events, so we need to use the style for internal/external
windows can get double click events.

Also, passthrough mouse events to parent window in the same message pumping
threads instead of manually forwarding each mouse event.

Fixes: https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/1172
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2277>

rtpredenc: quieten warning about ignoring header extensions

Turn it into a FIXME, and only log once

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2279>

Meson: Fix deprecation warnings

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1876>

rvsg: fix cairo include

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2276>

Revert "videorate: Update the base time on segment updates"

This reverts commit 75b4809ebc23814009bebc70c775ab44d85decf2.

See https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2186

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2254>

Revert "videorate: Add test for segment update"

This reverts commit a76f38b2c7ddbed546bb058c32ebcf8a553c003f.

See https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2186

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2254>

Revert "videorate: Only "close" the segment if it is discontinous"

This reverts commit 6f7922b4dbba5ed780e7b0988669a81848a9e333.

See https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2186

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2254>

Revert "videorate: Drop incoming buffers that are outside of the segment"

This reverts commit 24fd80344dbc059b72e13d813ca82f414a9d6cce.

See https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2186

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2254>

Revert "videorate: Add unit test for closing a segment and opening a separate one"

This reverts commit 98f2a84a289ed4d4cfac9bc5c73182a56eefa99c.

See https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2186

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2254>

rtprtxsend: mark RTX buffers with GST_RTP_BUFFER_FLAG_RETRANSMISSION

It is useful for elements downstream from rtxsend to know if the RTP
buffer they are dealing with is an RTX buffer or not.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2272>

cuda-converter: fix nvrtc compilation on non-English locale systems

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2268>

rtspclientsink: fix possible shutdown deadlock collect_streams()

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1696>

videoaggregator: keep old buffer when processing a MISSING_DATA gap

GAP events flagged with MISSING_DATA are transformed into GAP buffers
flagged with CORRUPTED.

In these cases, it is preferable to simply keep rendering the previous
buffer (if there was one) instead of flashing the pad in and out of
view.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/708>

compositor: fix prepare_frame obscuring check

A pad without a buffer or with a GAP buffer cannot obscure a
pad below it. Ignore those when considering whether a pad should
be drawn.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/708>

compositor: a pad without a frame can't obscure the background

Skip those when considering whether the background should be
drawn

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/708>

aggregator: forward gap event information to gap buffer

When the GAP event was flagged with MISSING_DATA, subclasses
may want to adopt a different behaviour, for example by repeating
the last buffer.

As we turn these gap events into gap buffers, we need to flag
those, we do so with a new custom meta.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/708>

Update proxy-libintl wrap to 0.4

It has fixes for Windows static build.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2262>

webrtcbin: Avoid access of freed memory

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2256>

mp4mux: fix spelling

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2241>

va: h264enc: Do not let L0 number surplus the HW limitation.

The algorithm to calculate the L0/L1 number may let the L0 number
surplus the HW limitation. We should ensure that limitation after
that calculation.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2202>

videoscale: Add test to ensure that non-scaled metas are kept

Just make sure that we're not breaking non-scaled metas

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1630>

videoscale: Add test for transform_meta function

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1630>

videoscale: Don't copy scaled metas

Returning TRUE from the `transform_meta` function tells
GstBaseTransform to copy the meta into the new buffer. If videoscale
has already transformed a meta by scaling it, it should always return
FALSE to avoid duplicating the meta.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1630>

x11: Fix unused variable warnings

These are emitted when XInput 2 is not available.

```
../subprojects/gst-plugins-base/sys/ximage/ximagesink.c: In function ‘gst_x_image_sink_handle_xevents’:
../subprojects/gst-plugins-base/sys/ximage/ximagesink.c:696:29: warning: unused variable ‘state’ [-Wunused-variable]
  696 |   GstNavigationModifierType state = GST_NAVIGATION_MODIFIER_NONE;
      |                             ^~~~~
../subprojects/gst-plugins-base/sys/ximage/ximagesink.c:694:35: warning: unused variable ‘touch_frame_open’ [-Wunused-variable]
  694 |   gboolean pointer_moved = FALSE, touch_frame_open = FALSE;
      |                                   ^~~~~~~~~~~~~~~~
../subprojects/gst-plugins-base/sys/xvimage/xvimagesink.c: In function ‘gst_xv_image_sink_handle_xevents’:
../subprojects/gst-plugins-base/sys/xvimage/xvimagesink.c:427:35: warning: unused variable ‘touch_frame_open’ [-Wunused-variable]
  427 |   gboolean pointer_moved = FALSE, touch_frame_open = FALSE;
      |                                   ^~~~~~~~~~~~~~~~
../subprojects/gst-plugins-base/sys/xvimage/xvimagesink.c:426:29: warning: unused variable ‘state’ [-Wunused-variable]
  426 |   GstNavigationModifierType state = GST_NAVIGATION_MODIFIER_NONE;
      |                             ^~~~~
```

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2093>

mingw: Remove deprecated section in example cross file

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2080>

meson: Add PYTHONPATH to load GDB helper module

Meson generates a gdbinit file that will automatically load gstreamer
script. However that script uses a helper python module that needs
PYTHONPATH to be pointing into the right location in the source
tree to be able to find gst_gdb.py.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1796>

multiudpsink: allow binding to IPv6 address

When the sink is configured to create sockets with an explicit bind
address, then the created socket gets set to the udp_socket field
irregardless of whether the bind address indicated that the socket
family should be IPv4 or IPv6.  When binding to an IPv6 address, this
results in the following error:

gstmultiudpsink.c:1285:gst_multiudpsink_configure_client:<rtcpsink>
error: Invalid address family (got 10)

This patch adds a check of the address family being bound to and sets
the created socket to used_socket or used_socket_v6, accordingly.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1551>

rtphdrextsdes: fixup test trying to g_free a local variable

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2235>

hls/m3u8: Fix starting segment for live playlist

RFC 8216 6.3.3 "Playing the Media Playlist File" : states that for live media
playlists "the client SHOULD NOT choose a segment that starts less than three
target durations from the end of the Playlist file"

This is an off-by-one error. Since we are looking for the "index" of the
segment, we need to subtract 1 from the searched position.

Ex: For a playlist with 12 entries, we want to start playback on the 9th segment
... which is at index 8.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2259>

test: Add test cases for the H265 bitwriter.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1831>

codecparsers: bitwriter: Add the common bit writer functions for H265.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1831>

va: vpp: Update the colorimetry and HDR fields when fixup src caps.

When we fixup src caps, the current way of handling the HDR fields is not
correct.
1. We trim the HDR fields only when the input caps is not a subset of the
fixup src caps. But in fact, the input caps with HDR fields such as the
"mastering-display-info" can possibly be the subset of the fixup src caps,
if they have all same other fields.
2. We always copy the colorimetry from input caps to src caps if it is
absent. But when hdr-tone-mapping is enabled, the HDR->SDR conversion makes
the colorimetry change. We should use downstream's setting, or just use the
default colorimetry of SDR.

We changes to:
1. If hdr-tone-mapping is enabled, we trim all HDR fields and add a correct
colorimetry.
2. Copy colorimetry from input if it is still absent.
3. Consider the subset replacement.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2244>

validate: Do not print Known error tests

Those are passing tests in practice and it doesn't give any useful information

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2264>

tests: convertscale: Do not log EOS events in validateflow

We already checked that we are getting an EOS before going back to ready
and adding the EOS at the end of the .validateflow is racy as the ERROR
message might reach the bus before it is logged.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2264>

ci: Trigger cerbero on user project if a branch with the same name exists there

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2264>

docs:examples: Fix building iOS examples

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2264>

validate:pipeline-monitor: Minor cleanup setting unused variable

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/898>

videoconvertscale: Add properties to disable scaling/converting in videoconvert/videoscale

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/898>

validate: scenario: Add a 'allow-error' configuration

This new configuration allows the scenario to keep running even
after receiving an "error" message on the bus.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/898>

videoconvertscale: Don't claim we can support any kind of memory

Since d0133a2d11566ff4c0cded7af8dfdff0046e0e8b "videoconvert: Allow
passthrough for ANY caps features" videoconvert will always claim that
it supports any kind of memory which is true in very specific case (when
it is running in passthrough mode). To get elements that autoplug
converters depending on the caps running in the pipeline (like
autovideoconvert), we need to have converters no lie about what they can
do when queried `accept_caps` or `query_caps`.

This still accepts any caps feature as before but it introduces
a restriction in the way we handle memory capsfeatures.

We keep previous behaviour in videoconvert and videoscale.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/898>

Introduce the videocolorscale element

Now that videoconvert and videoscale's are both based on
GstVideoConverter and are using the exact same code, it makes much more
sense to have one element doing the two operation, and it can be
more efficient in some cases (one single path for both operations).

This removes the `videoscale` and `videoconvert` plugins but keeps the element
but makes them also do both operations (adding some APIs to each element).

There is a small change in API for the `videoscale:dither` property which
was previously a totally unused boolean, it is now an enum and is used.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/898>

qsv: Add GPU description to element long-name on Windows

Would be useful for a multi-Intel-GPU system

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2249>

hls: Relax webvtt checks

If no hour field is present (which is allowed), the remaining data can be less
than 15 character.

Fix time translation failures if the hour field wasn't present

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2248>

vapostproc: Notify if hdr-tone-mapping property changes.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2247>

gstleaks: log the number of alive objects

Can help checking if the number of alive objects is growing over time or stays stable.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2245>

va: Expose a couple va surface/buffer methods.

And use them in va plugin.

Also, set as default logging category the vadisplay one.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2077>

rtpmanager: Move some duplicated constant and helper function to a single place

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2132>

gstreamer: Document various caps for the reference timestamp meta

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2132>

rtpbin/rtpjitterbuffer: Don't parse RTCP SRs twice unless needed

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2132>

rtpjitterbuffer: Add property to throttle handling of RTCP SR / NTP-64 syncing

This proxies the "rtcp-sync-interval" property of rtpbin.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2132>

rtpsession: Handle RTCP-SR-REQ (RFC6051) RTCP feedback message

This causes an RTCP SR to be sent at the earliest possible time.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2132>

rtpbin/rtpjitterbuffer: Allow syncing to an SR without CNAME if the CNAME is already known

The RTCP SR packet might be without SDES in case of a reduced-size RTCP
packet. For syncing purposes the CNAME is needed but it might be known
already from an earlier RTCP packet or out of band, via the SDP for
example.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2132>

rtpbin/jitterbuffer: Use inband 64-bit NTP timestamps according to RFC6051 for faster synchronization

When signalled via the caps that the header extension is used, it will
be read and used in the same way as the RTP/NTP time mapping from RTCP
SRs.

If the CNAME of the stream's SSRC is provided out of band via e.g. the
SDP then this allows streams to be synchronized immediately on the first
packet instead of having to wait for the first RTCP SR to arrive.

Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/issues/383

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2132>

rtpsession: Only add send latency to the running time if it is actually known

Otherwise we can't know the running time yet if rtcp-sync-send-time is
set, and have to wait until the latency is known later.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2132>

rtpsession: Update 64-bit NTP header extensions with the actual NTP time in senders

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2132>

rtpmanager: Add header extension implementation for the 64-bit RFC6051 NTP header extension

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2132>

sdp: Parse the RFC5576 Source-specific media SDP attributes into caps

The format of the caps fields is
    ssrc-(SSRC_VALUE)-(ATTRIBUTE_NAME)=(ATTRIBUTE_VALUE)
.

Parsing of the attributes from the caps into the SDP is not implemented
as this depends not only a single stream's caps but on the whole rtpbin
configuration.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2132>

rtsp-server: Add RFC5576 Source-specific media attribute to the SDP media for signalling the CNAME

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2132>

dashsink: Unlock when failed to get content

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2242>

gstglmixerbin: minor refactor of _find_element_pad_template

As suggested by @slomo, make the loop more readable and prevent returning
a garbage value to the caller from the previous implementation.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2232>

gstav1parse: fixup various possible logical errors

Found via an analyzed build for Clang. Specifically we had:

gstav1parse.c[1850,11] in gst_av1_parse_detect_stream_format: Logic error: The left operand of '==' is a garbage value
gstav1parse.c[1606,11] in gst_av1_parse_handle_to_small_and_equal_align: Logic error: The left operand of '==' is a garbage value

Also a couple of false-positives:

gstav1parse.c[1398,24] in gst_av1_parse_handle_one_obu: Logic error: Branch condition evaluates to a garbage value
gstav1parse.c[1440,37] in gst_av1_parse_handle_one_obu: Logic error: The left operand of '-' is a garbage value

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2230>

Always define ENABLE_NLS

GLib guarantees libintl API is always available, provided by
proxy-libintl as last resort. GLib itself unconditionally define
ENABLE_NLS.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2028>

Delete unused i18n headers

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2028>

Replace gst-i18n-*.h with gi18n-lib.h

GLib guarantees libintl is always present, using proxy-libintl as
last resort. There is no need to mock gettex API any more.

This fix static build on Windows because G_INTL_STATIC_COMPILATION must
be defined before including libintl.h, and glib does it for us as part
as including glib.h.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2028>

gst-play: Do not split translatable string

Concatenating N_() strings does not work with gi18n.h macro, was working
only with GStreamer's redefinition.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2028>

cuda: Fix typo in gst_cuda_buffery_copy_type_to_string

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1472>

cuda: Add a `gst_cuda_context_new_wrapped` constructor

So that application or people implementing cuda element from
outside GStreamer itself can ensure that their contexts are used
by GStreamer.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1472>

cuda: Factor out a public GstCUDA library

So applications and elements implemented outside GStreamer can reuse
our infrastructure

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1472>

nvenc: Give more details about errors

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1472>

video: Fix possible overrun when iterating comp[] array

Fix 2 iterations that can overrun the array if the number of component is
equal to the size of the array.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2239>

Revert "video-frame: avoid possible out of bound memory access"

This reverts commit c4255f08f787ff6a41504b538fa4c2e911d40ec7.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2239>

meson_options: declare glib and gobject checks at top level

This makes it easier to specify this features for all subprojects.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2238>

meson: Add -Wl,-rpath,${libdir} on macOS

We made the gstreamer installation prefix relocatable by picking up
plugins relative to the location of libgstreamer-1.0.dylib, similar to
how it's done for Windows:

https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1627

This had a lot of side-effects:

https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/1051
https://gitlab.freedesktop.org/gstreamer/cerbero/-/issues/363
https://gitlab.freedesktop.org/gstreamer/cerbero/-/issues/371
https://gitlab.freedesktop.org/gstreamer/cerbero/-/issues/362

A partial fix for the cerbero side of these was:

https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/807

However, this relied on the consumers knowing that they need to add
`LC_RPATH` entries to the libdir of the prefix. This is done
automatically by build systems like Meson, but not by others, such as
Autotools, CMake, Cargo, XCode, etc. For those, we need to add the
RPATH entries to the gstreamer-1.0.pc file.

This also has the side-effect of fixing the loading of gstreamer rust
plugins on macOS:

Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/1159

Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/1149

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2218>

video-frame: avoid possible out of bound memory access

Although the components' initialization code would fill in -1 to all
unset components, make the code a bit more defensive and check for an
index bound first.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2231>

tests: dash_mpd: fix linker issues with non-optimizing compilers

undefined reference to `download_request_take_buffer'

Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2117#note_1344646

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2228>

gst_plugin_load_file: force plugin reload if diff filename

If a file includes a new version of a plugin that exits in the
registry, the output of gst-inspect is incorrect. The output has the
correct version but incorrect filename, and element description.

This seems to have also fixed some documentation issues.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1344>

docs: design: fix up some document titles

These show up in the index.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2222>