Commit graph

265 commits

Author SHA1 Message Date
Edward Hervey
18b941dfd5 decodebin3: Unify collection switching checks
We only want to switch to a selection of an output collection if all streams are
present.

This was previously only done in one place (when triggering by new incoming
streams) but not when triggered by user/application.

Avoid this by moving the check to handle_stream_switch()

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7941>
2024-11-25 08:03:38 +00:00
Philippe Normand
827caa662c playbackutils: Fix caps leak in get_n_common_capsfeatures()
The gst_static_caps_get() return value is transfer-full.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7908>
2024-11-16 18:24:23 +00:00
Edward Hervey
09be1ab863 urisourcebin: Also use event probe for HLS use-cases
Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/3957

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7822>
2024-11-08 09:14:21 +00:00
Jan Schmidt
c02d41c259 subtitleoverlay: Check for memory capsfeature on video/x-raw
Don't assume that video/x-raw caps means buffers are mappable
or can be processed by videoconvert and friends. Only plug
those converters for real system memory, and treat other
memory capsfeatures as hardware surfaces

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7741>
2024-10-30 06:13:23 +00:00
Edward Hervey
39c770af90 playback: Improve stream list search
There is the possibility than an element/code/helper creates an identical
`GstStream` (same type and stream-id) instance instead of re-using a previous
one.

For those cases, when detecting whether a `GstStream` is already present in a
collection, we need to do more checks than just comparing the pointer.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7716>
2024-10-28 10:09:47 +00:00
Edward Hervey
0fda426656 urisourcebin: Aggregate collections from multiple parsebin
In the case where multiple parsebin are present (ex: from rtsp sources), we want
to aggregate the collections provided by the different parsebin and expose a
single "unified" collection.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7716>
2024-10-28 10:09:47 +00:00
Edward Hervey
a26984a72b parsebin: Store caps on parsepad as early as possible
When analyzing a new pad, we can store the updated caps on the GstStream if they
are fixed.

If they are not fixed, this function will be called again once the element will
provide them.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7716>
2024-10-28 10:09:47 +00:00
Edward Hervey
3fdfef15a1 parsebin: Don't store bogus caps on GstStream
If we can't get the current caps when receiving a stream-start, that's fine,
they can/will be provided by other means at a later time.

What we definitely should not do is provide the starting caps of the chain,
which are potentially completely different from the end ones (like for example
`application/x-rtp`)

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7716>
2024-10-28 10:09:47 +00:00
Edward Hervey
1814aec0d5 parsebin: Minor debug statement updates
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7716>
2024-10-28 10:09:47 +00:00
Sebastian Dröge
073831128b uridecodebin3: Remove "source" property
There is not necessarily a single urisourcebin inside uridecodebin3 so having a
single source property makes little sense. Additionally, this property was never
hooked up at all and always returned NULL.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7712>
2024-10-23 12:29:06 +00:00
Tim-Philipp Müller
f5169670bc playback: remove mention of gconf sinks from playbin docs
Fixes #3916

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7676>
2024-10-19 06:12:14 +00:00
Edward Hervey
189cbe7fdd urisourcebin: Don't remove probe if not present
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7609>
2024-10-18 08:49:50 +00:00
Edward Hervey
b2c0f490f0 urisourcebin: Set pad probe before linking pad
We want to grab all events that pass through, so we need to set the
probe **before** the pad is linked

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7609>
2024-10-18 08:49:50 +00:00
Edward Hervey
c97126212f urisourcebin: Streamline event forwarding code
Instead of two different functions for copying events and rewriting the
stream-start event, just have a single one.

This is needed, since we want to do both of those in one go, with the pad lock
taken.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7609>
2024-10-18 08:49:50 +00:00
Edward Hervey
f130382b08 parsebin: Set stream collection on pad before exposing it
We want to ensure the stream-collection is present on the pad (as a sticky
event) before we expose the pad.

This is more reliable since it will ensure it is present before any other event
is pushed through.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7609>
2024-10-18 08:49:49 +00:00
Edward Hervey
cb25b1e95e uridecodebin3: Use lock when checking input/output items
Otherwise there is the risk that the play items list could have changed in
between

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7609>
2024-10-18 08:49:49 +00:00
Edward Hervey
e1e48b9fea decodebin3: Free pending events on input pads when unlinking
Otherwise we could end up with *old* events leaking through on the next linking.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7609>
2024-10-18 08:49:49 +00:00
Théo Maillart
aaf9b46e2f decodebin3: do not attempt to remove a null stream
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7586>
2024-10-03 07:08:37 +00:00
Théo Maillart
4065ac5dcb decodebin3: protect internal reset with SELECTION_LOCK
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7586>
2024-10-03 07:08:37 +00:00
Théo Maillart
ed264d64b0 decodebin3: remove output event probe on remove input stream
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7586>
2024-10-03 07:08:37 +00:00
Edward Hervey
b21ffc6b9f urisourcebin: Ensure all stream-start are handled
In order to ensure all initial events (stream-start, caps, ..) are present on
pads that we expose, those various sticky events are propagated (from parsebin
to multiqueue output, from multiqueue output to exposed pads).

The problem was that the "hack" in `urisourcebin` to inform downstream elements
that the stream is parsed data and a collection will be present was only done in
one place : a probe on the output of parsebin ... but the stream-start could
potentially have already been propagated to the output pads before that.

In order to fix that, we make sure any pending sticky stream-start event is
updated before being propagated.

Fixes #3788

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7598>
2024-10-02 21:29:00 +00:00
Edward Hervey
32bf29d64d urisourcebin: Allow more cases for posting stream-collection
Previously urisourcebin only allows stream-collections messages from adaptive
demuxers or sources to be posted.

This commit also allows the case where they come from a single parsebin. We
still want to prevent it in the case where they are multiple parsebins, since
that would require some form of aggregation to show a single/unified collection.

In order to avoid a regression with uridecodebin3 behavior, we also implement
support for GST_QUERY_SELECTABLE, so that uridecodebin3 can figure out whether
it should let GST_MESSAGE_STREAM_COLLECTION flow upwards (because app/user could
react on it) or whether it drops it in order for decodebin3 to do the collection
aggregation and posting.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7597>
2024-10-02 19:40:48 +00:00
Edward Hervey
9a59fc7168 decodebin3: Make update/posting of collection messages atomic
The presence (or not) of a collection on an input will determine whether events
will be throttled so that there are only forwarded when that input gets a valid
collection.

Therefore the input lock should be used.

In addition to that, we want to ensure that the application/user has a chance to
reliably (i.e. synchronously) specify what streams it is interested in by
sending a GST_EVENT_SELECT_STREAMS.

But we cannot allow anything to go forward until that message posting has come
back, otherwise we run in various races.

Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/3872

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7594>
2024-10-02 18:50:06 +00:00
Elliot Chen
6c830c5bd3 decodebin3: check and send selected stream message even if no decoder is selected
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7579>
2024-10-02 10:21:56 +00:00
Sebastian Dröge
6233eb0ff3 common: Stop using GQuark-based GstStructure field name API
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7432>
2024-09-26 19:21:29 +03:00
Sebastian Dröge
0c1611d31d common: Stop using GQuark-based GstStructure name API
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7432>
2024-09-26 19:21:29 +03:00
Edward Hervey
ea59c921d6 decodebin3: Fix collection identity check
Collections can be auto-generated from upstream and yet have exactly the same
streams in it.

Therefore do a more in-depth check for equality.

Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/3742

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7401>
2024-08-22 09:27:02 +00:00
Edward Hervey
de6de83986 urisourcebin: Actually drop EOS on old-school pad switch
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7375>
2024-08-19 09:49:12 +02:00
Edward Hervey
7546975856 urisourcebin: Don't hold lock when emitting about-to-finish
This could trigger actions that re-enter this element

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7368>
2024-08-17 05:50:26 +00:00
Edward Hervey
701165227f decodebin3: Remove custom EOS handling
Initially introduced in 2017 by 4fcbcf4e48 (and
follow up commits) and ec7d81f67c

See https://bugzilla.gnome.org/show_bug.cgi?id=773341 and
https://bugzilla.gnome.org/show_bug.cgi?id=792693 for more details.

This was made to support the legacy behaviour of adaptive demuxers (regarding
streams added/removed dynamically).

Since then, a lot of things have changed in decodebin3 and related elements
regarding how dynamic streams are handled and this custom behaviour is no longer
needed.

This also removes weird behaviours like EOS being delayed until *all* streams
were EOS, which could cause deadlocks downstream.

Fixes #3666

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7194>
2024-08-16 13:26:14 +00:00
Alicia Boya García
9b0e951512 streamsynchronizer: Add documentation
I didn't find the behavior and purpose of streamsynchronizer documented
or intuitive. Eventually I got Edward to explain it to me, which was
very helpful. Now I'm contributing some docs so that the next person
doesn't have to figure it out by asking around and hoping for an answer.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7084>
2024-08-14 13:55:37 +00:00
Philippe Normand
bd64374174 parsebin: accept-caps handling for elements with unusual pad names
In case the last element of the parse chain doesn´t have a sink pad named
"sink", send the accept-caps query to the first sink pad of the element.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7255>
2024-07-29 12:41:33 +00:00
George Hazan
9e1abc0797 parsebin: fix compilation problem with Visual C++ 2017
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7089>
2024-06-22 20:56:03 +03:00
Edward Hervey
11ff2b2835 decodebin3: Fix keyframe drop probe handling
We were storing the probe id in a different structure (DecodebinOutputStream)
than the pad it is targetting (which is in MultiQueueSlot).

The problem is that when re-targetting outputs (to a different slot)... we would
end up having an invalid probe id, or not have a reference to an existing one.

Instead, store the probe id in the same structure as the pad it's targetting

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7069>
2024-06-20 12:56:46 +00:00
Edward Hervey
6aa0c8442d decodebin3: Fix detection of selection done
We should not assert if there are still some old streams that are waiting to be
deactivated.

Instead wait for them to be gone before posting the selection done message

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7069>
2024-06-20 12:56:46 +00:00
Alicia Boya García
223b63b1d2 decodebin3: Add missing INPUT_LOCK() before ensure_input_parsebin()
ensure_input_parsebin() has a top comment saying it must be called with
INPUT_LOCK taken, but 2 out of 3 usages of the function call it without
taking that mutex.

This patch adds locking in these two remaining usages.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/5279>
2024-06-17 10:15:28 +00:00
Edward Hervey
c5a39f2dfa uridecodebin3: Don't hold PLAY_ITEMS lock when activating them
Once the item is configured it can be activated without holding that lock

Fixes #3610

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7015>
2024-06-11 16:42:27 +00:00
Edward Hervey
74dbbe0091 decodebin3: Always ensure we end up with parsebin or identity
This fixes a regression introduced by 6c4f52ea20

There are cases where the input stream will be push-based, time-segment and not
have a collection nor caps. This means the event-based checks are not sufficient
to decide when/where to plug in a identity or parsebin to process the input.

For those corner cases we setup a buffer probe to ensure we always end up with
at least a parsebin

Fixes #3609

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7010>
2024-06-11 14:57:34 +00:00
Edward Hervey
6c4f52ea20 decodebin3: Avoid usage of parsebin even more
When dealing with push-based inputs, we are now delaying the creation of
parsebin/identity until we get all pre-buffer events.

We therefore can simplify the handling of new pads being linked and only have to
check if upstream can handle pull-based or not.

Avoids creating parsebin for parsed upstream data altogether

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6953>
2024-06-04 14:44:14 +00:00
Edward Hervey
cb95ba72cf decodebin3: Include the stream-id in ERROR/WARNING/INFO messages
Allows application and other users to know to which stream the
error/warning/info message originated from.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6964>
2024-05-31 11:58:21 +00:00
Edward Hervey
686d980955 uridecodebin3: Specify URI for INFO/WARNING/ERROR messages
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6929>
2024-05-29 22:24:07 +00:00
Edward Hervey
94646c642d decodebin3: Specify stream-id for missing decoder
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6886>
2024-05-29 11:44:06 +00:00
Edward Hervey
39f62862d8 decodebin3: Ensure we get a collection for parsed inputs
When we are dealing with parsed inputs (i.e. using identity), we need to ensure
that we have a valid stream collection (and therefore DBCollection) before
anything flows dowsntream.

In those cases, we hold onto those events until we get such a collection.

Fixes #3356

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6774>
2024-05-28 08:31:17 +00:00
Edward Hervey
daa022b9ee decodebin3: New mechanism for handling collection and selections
This commit separates collection and selections into a new separate structure:
DecodebinCollection.

This provides a much cleaner/saner way of dealing with collections being
updated, gapless playback, etc...

There is now a list of DecodebinCollection in flight, of which two are special:
* input_collection, the currently inputted/merged collection
* output_collection, the currently active collection on the output of multiqueue

Handling GST_EVENT_SELECT_STREAMS is split, by looking for the collection to
which it applies. And the requested streams are stored in it. IIF that
collection is output_collection we can do the switch, else it will be updated
when it becomes active.

Detecting which collection/selection is active is done by looking at the
GST_EVENT_STREAM_START on the output of the multiqueue.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6774>
2024-05-28 08:31:17 +00:00
Edward Hervey
c4b625a3fe decodebin3: minor refactoring to identify selected stream
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6774>
2024-05-28 08:31:17 +00:00
Edward Hervey
bfb64f7f44 decodebin3: Debug line cleanups
Use identifiable items in log lines instead of random pointers

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6774>
2024-05-28 08:31:17 +00:00
Edward Hervey
39f2d96105 decodebin: Remove unused includes
* config.h is not used, plugin/element is registered in another file
* play-enum.h is not used

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6774>
2024-05-28 08:31:17 +00:00
Edward Hervey
48cbb1c96f decodebin3: Remove un-needed variable
We don't do anything with the unknown streams. Detecting that a list of
requested streams don't apply to a given collection should be handled
before-hand

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6774>
2024-05-28 08:31:17 +00:00
Edward Hervey
3acb219b76 decodebin3: Remove un-needed variable
pending_select_streams was only set just before releasing/taking the selection
lock in a single place. That temporary lock release is not needed and therefore
the variable isn't needed either

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6774>
2024-05-28 08:31:17 +00:00
Edward Hervey
1ab0936196 decodebin3: Remove active_selection list
It's a duplicate of the list of slots which have an output. Use that instead.

Also when we fail to (re)configure an output, remove it.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6774>
2024-05-28 08:31:17 +00:00