Make sure to push the frame for a still frame, with discont = true and
timestamp=none, so that it gets displayed by the sink. Also, don't leak
each rendered video frame during still menus.
If the estimated bitrate is lower than 188 bytes, there's most likely
something completely wrong with the two samples. If that happens,
force recalculation.
Use guint64 for observation PCR, I saw cases where it would overflow.
This allows debugging with mpegtsdemux:4 while being able to track what's
going on (and avoid taking up as much cpu for debugging as for the actual
demuxing process).
This ensures that the metadata items are always written
in the same order and that first comes the preface,
then the identification linked from the preface, ...
Some demuxers can't handle files where the metadata
items are in random order.
Sync from gst-fluendo-mpegdemux and have seeking/duration query
improvements in. No support however for wrapped around pcrs etc. but a
start nonetheless.
Also fix indentation issues.
We usually only get SR packets in our chain function but if an invalid packet
contains the SR packet after the RR packet, we must not fail but simply ignore
the malformed packet.
Fixes#581375
Pluggin the text-ovelay causes some slowdowns and might need a colorspace
converter. We now only check the text-overlay property whne goint to ready and
plug or unplug the text-overlay at that time. If the property changes during
playing, its not switched immediately.
We shouldn't register a new GstTag for every unknown tag
we find as this might lead to conflicts and also those
tags are essentially unknown.
Add mappings for some known tags and also convert string
dates to GDate, as found in many FLV files.
Since neither rtpmanager nor any of the payloaders properly implement
pad allocation, there is no way for the rtpmanager to inform downstream elements
of the new SSRC if there is an SSRC collision. So the warning is emitted all the
time and it is confusing.
Fixes#580144
When we get bus messages for setting the xid, we have the real sink element
instance as message_src. No need to proxy the xoverlay iface therefore
(autovideosink does not do it either). Also we don't need to rewrite the
message src of all messages from that sink.
When a segment event is received on the active pad, forward it downstream
immediately instead of deferring it until the next data buffer arrives. This
fixes problems with segment updates never being sent downstream, like those
needed for sparse streams, or for closing previously opened segments.
This fixes playback of DVD menus with a still video frame and an audio track,
for example.
Fixes: #577843
Add a virtual method in rawparse to set buffer flags. This doesn't
use API from unreleased -base, since it defines GST_VIDEO_BUFFER_TFF
if it's not defined yet.
This should remove the bogus error messages while still keeping the original
intent of this, which is to inform the pipeline/application/user that we
could not find any valid streams.
There are many reasons why pushing an event can fail, and not all of them are
because there's no link downstream (it could be because it was blocked, or
flushing).
itvhd masks its h264 video stream as a private stream making it harder for
other set top boxes to decode. this checks for specific program number, video
pid and stream type combination before declaring it as h264.
For this add a "mode" property that defaults to "interlaced" for now as
most decoders/demuxers don't properly set the "interlaced" field on the
caps yet.
If this property is set to "auto" the element will work in passthrough
mode unless the caps contain the "interlaced" field.
Don't allow setting filename via img-done signal parameter but force app
use filename property. Don't stop capture when setting filename property.
Update check unit test based on the change.
Sending the flush-start event forward before taking the stream lock actually
works, in contrast to deadlocking in downstream preroll_wait (hunk 1).
After that we get the chain function being stuck in a busy loop. This is fixed
by updating the minimum frame size inside the synchronization loop because the
subclass asks for more data in this way (hunk 2).
Finally, this leads to a very probable crash because the subclass can find a
valid frame with a size greater than the currently available data in the
adapter. This makes the subsequent gst_adapter_take_buffer call return NULL,
which is not expected (hunk 3).
If one side has a preference for a particular sample rate or set of sample rates, we
should honour this in the caps we advertise and transform to and from, so that elements
actually know about the other side's sample rate preference and can negotiate to it
if supported. Also add unit test for this.
