Change the pad health calculation to consider a pad 'healthy'
if it has received data within the last 'timeout' window. Previously,
inactive pads were constantly flip-flopping between healthy and not
healthy depending on whether they were slightly ahead of or behind
the active pad running_time.
When the health status of a pad changes, make sure to always notify
the property, so that applications that are manually controlling
the active pad can make their switching decisions.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs/-/merge_requests/1286>
We cannot continue with this buffer, because we cannot calculate the
time when the recording stopped or started. We also cannot safely drop
it, because that might break the stream, especially if it's encoded.
Therefore, we return an element error.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs/-/merge_requests/1266>
Previously livesync was waiting for the start timestamp of the current
buffer after looking at the queue and right before pushing it
downstream. This meant that it generally looked too early at the queue
and especially that upstream had to provide the next buffer already at
the start timestamp of the previous one.
Instead, now wait before looking at the queue and wait for the end
timestamp of the previous buffer. Once the previous buffer has expired,
a new buffer really needs to be available or otherwise a filler buffer
has to be pushed downstream.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs/-/merge_requests/1266>
I've looked at the GstQueue code again and tried making livesync behave
better with EOS. This isn't very well tested, though. My goal was to
make this look saner but I think this should be reviewed by someone who
knows the queue code.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs/-/merge_requests/1266>
Fix the following use case:
- main input of fallbackswitch is finite (a media file)
- fallback input is infinite (videotestsrc)
- main input is done and send eos, which is propagated downstream
- fallbackswitch switches to fallback, sending STREAM_START which reset
EOS downstream (aggregator does that)
- fallback input keeps pushing buffers forever.
Solve it by adding a 'stop-on-eos' property so fallbackswitch stops
pushing property once the main input is eos.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs/-/merge_requests/1248>
The logic of the element requires the next buffer to be available
immediately after we are done pushing the previous, otherwise we insert
a repeat.
Making the src loop handle events and queries broke this, as upstream is
almost guaranteed not to deliver a buffer in time if we allow non-buffer
items to block upstream's push.
To fix this, replace our single-item `Option` with a `VecDeque` that we
allow to hold an unlimited number of events or queries, but only one
buffer at a time.
In addition, the code was confused about the current caps and segment.
This wasn't an issue before making the src loop handle events and
queries, as only the sinkpad cared about the current segment, using it
to buffers received, and only the srcpad cared about the current caps,
sending it just before sending the next received buffer.
Now the sinkpad cares about caps (through `update_fallback_duration`)
and the srcpad cares about the segment (when not in single-segment
mode).
Fix this by
- making `in_caps` always hold the current caps of the sinkpad,
- adding `pending_caps`, which is used by the srcpad to store
caps to be sent with the next received buffer,
- adding `in_segment`, holding the current segment of the sinkpad,
- adding `pending_segment`, which is used by the srcpad to store
the segment to be sent with the next received buffer,
- adding `out_segment`, holding the current segment of the srcpad.
Maybe a fix for
https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs/-/issues/298.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs/-/merge_requests/1082>
