| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
Other layers will seek as appropriate, we shouldn't fight them I don't think.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Trying to repeat in VideoDecoder is the wrong side of the distinction
between content and DCP time; the repeat is for the DCP and VideoDecoder
should be emitting in terms of the source.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
jobs queued in the batch converter don't start too early; e.g.
if you add the same job twice the check of the existing video
won't start on the second job until the first is complete.
|
|
|
|
|
|
|
|
|
|
at the end of the film (with no corresponding video).
|
|
|
|
to the next video frame). Fix thinko in ::done(). Fix initial value
of _position when the first empty period does not start at time 0.
|
|
|
|
of things that we are ignoring; this fixes strange behaviour (delays)
when analysing audio as it used to keep pass()ing decoders to get
data that would never come.
|
|
|
|
content.
|
