Instead of doing it at the time of resetting `object->in_composition`
when user calls `gst_bin_remove` do it after we actually removed
it from the object thread, and do it in the `nle_object_reset`
method where it belongs
Fixes https://gitlab.freedesktop.org/gstreamer/gst-editing-services/issues/96
nleobject.duration defaults to 0, but this is pretty unintuitive for
end user in the case nlesource is use standalone, just consider
duration=0 equivalent to duration=GST_CLOCK_TIME_NONE as it makes
the element much simpler to use, we could actually forbid 0 as a value
in the future.
Also take into account potential CLOCK_TIME_NONE
Basically the tags we send before STREAM_START are now ignored, meaning
that we could not detect nested timelines anymore, this commits makes
sure that we send our tag event after getting pushing STREAM_START.
When a composition is nested into anotherone, we *know* that the
toplevel composition is going to send a stack initializing seek,
we can thus avoid sending it on the subcomposition itself when
prerolling. This avoid seeking roundtrips and we now have one and
only one seek travelling in the overall pipeline (after it has
prerolled).
The 'start' of nleobject is in the 'composition' scale, inpoint is in
the media scale, when outside a composition, a nleobject->start value
doesn't mean anything.
Instead of seeking each nleobject separately to setup new stack, wait
for the whole stack to preroll and then seek that newly setup stack,
leading to the same code path and seek 'tweaking' as when processing
a seek on the composition (without stack changes).
This is mandatory to properly handle filter that tweak segments to handle
time remapping for example.
By passing NULL to `g_signal_new` instead of a marshaller, GLib will
actually internally optimize the signal (if the marshaller is available
in GLib itself) by also setting the valist marshaller. This makes the
signal emission a bit more performant than the regular marshalling,
which still needs to box into `GValue` and call libffi in case of a
generic marshaller.
Note that for custom marshallers, one would use
`g_signal_set_va_marshaller()` with the valist marshaller instead.
Implement lazy loading asset cache so gesdemux use the formatters
assets while GES hasn't been initialized.
And set extensions to temporary files as some formatters require
the information (otio)
Adding a property to let the application know
Also make sure that the duration of nested timeline assets is reported
as CLOCK_TIME_NONE as those are extended as necessary.
And make a difference between asset duration and their max duration
As nested timelines can be extended 'infinitely' those max duration
is GST_CLOCK_TIME_NONE, but their duration is the real duration of
the timeline.
When we have nested timelines, we need to make sure the underlying
formatted file is reloaded when commiting the main composition to
take into account the new timeline.
In other to make the implementation as simple as possible we make
sure that whenever the toplevel composition is commited, the decodebin
holding the gesdemux is torn down so that a new demuxer is created
with the new content of the timeline.
To do that a we do a NleCompositionQueryNeedsTearDown query to which
gesdemux answers leading to a full nlecomposition stack
deactivation/activation cycle.
Make sure that an event resulting from the seek happens before removing
the pad probe, dropping anything while it is not the case.
This guarantees that the seek happens before `nlesource` outputs
anything. This was not necessary as with decodebin or usual source
flushing seeks lead to synchronous flush_start/flush_stop and we could
safely assume that once the seek is sent, it was happenning.
With nested `nlecomposition` this assumption is simply not true as
in the composition seeks are basically cached and happen later in
the composition updating thread.
This fixes races where we ended up removing the blocking probe before
the seek actually started to be executed in the nlecomposition
nested inside an nlesource which leaded to data from *before* the seek
to be outputed which means we could display wrong frames,
and it was leading to interesting deadlocks.
Seeks that lead to a stack change lead to deactivating the current
stack. At that point we explicitely flush downstream as a reaction to
the flushing seek. Until now those flushes had a random seqnum, this
fails if we are a nested compostion as the parent composition will end
up dropping that flush which in turns might lead to deadlocks. For
example, the flush goes through a `compositor` which wants to flush
downstream to stop its srcpad task, but that flush wouldn't have
"released" its srcpad thread if the composition srcpad drops it, meaning
it won't be able to stop the task ever.
Otherwise if we shutdown a composition whith an nested composition
(inside a source in the test) and leak it, we end up with the nested
composition task still running (in READY) which is bad.
Add a test for that which leaks the pipeline on purpose.
First marshilling it to the main thread is dangerous as it is a blocking
operation and it should never happen there.
The asset cache is MT safe now so it is possible to load the timeline
from that thread directly
Making sure to have 1 GstDiscoverer per thread.
Use that new feature in gesdemux by loading the timeline directly from
the streaming thread. Modifying the timeline is not supported allowed
anyway.
