In this patch, we first deal with the case of MediaElement. Now we replace |PlayVideo| with |VideoFrameContainer::SetCurrentFrames| in |SourceMediaStream::AppendToTrack|. The MSG use TimeStamp::Now() for the TimeStamp of each video frame in most of case except MediaElement case. Becasue the MediaElement has its own VideoQueue, we need to calucalte the correct Timestamp based on the StartTimeStamp of this MediaStream and the elpased time of the video frame in DecodedStream.
MozReview-Commit-ID: 2bm2AHkFXHu
In this patch, we first deal with the case of MediaElement. Now we replace |PlayVideo| with |VideoFrameContainer::SetCurrentFrames| in |SourceMediaStream::AppendToTrack|. The MSG use TimeStamp::Now() for the TimeStamp of each video frame in most of case except MediaElement case. Becasue the MediaElement has its own VideoQueue, we need to calucalte the correct Timestamp based on the StartTimeStamp of this MediaStream and the elpased time of the video frame in DecodedStream.
MozReview-Commit-ID: 2bm2AHkFXHu
HTMLVideoElement can expose its thread-safe FrameStatistics object, so that
HTMLMediaElement can access more adequate data for its telemetry, without
having to use an intermediary (and potentially less accurate)
VideoPlaybackQuality object.
This will also help with accessing other/new FrameStatistics members later on.
MozReview-Commit-ID: AT7mEGy0zGr
Removed test blocking execution of ReportTelemetry in non-MSE case.
EME currently depends on MSE, so ReportEMETelemetry won't do anything in the
non-MSE case anyway. But I think it's good to remove the MSE test as well, in
case we change that limitation in the future.
MozReview-Commit-ID: 8sTg5F13KMs
Note that this is a simple duration counter based on the existing (non-hidden)
play-time, but it only counts when video is playing while hidden.
There is no bucketing yet, future probes may add finer-grained information.
MozReview-Commit-ID: DUfryXjGBAN
This is so that if media autoplay is disabled, a site can capture user intent
to play in their click handlers by calling load() on an empty video element to
"bless" the video element with the ability to play later when the video has
loaded.
Large video sites typically have a catalog view of videos to chose from on
their main page, and when the user clicks on one, with this patch, the site can
bless a video element so that it will play by beginning the load of that video
in the click handler. Often these sites do a bunch of asynchronous things
before they actually get around to playing the video, they don't call play in
the click handler, so we need another way to bless the videos so they can play.
We allow seeks in a click handler to capture user intent to play, so I don't
see why we should not also allow a load() to capture user intent.
MozReview-Commit-ID: KzjNcn3s6od
This can reduce the include header dependency. MediaStreamVideoSink will inherit from DirectMediaStreamTrackListener. But we can't use forward declaration on MediaStreamListener because the usage of nsTArray<RefPtr<MediaStreamVideoSink>>.
MozReview-Commit-ID: 328s4Kw9NvW
1. If mHasUserInteraction MediaElement is false, don't run the TimeMarchesOn because the element is not played. 2. Update the activeCueList only in TimeMarchesOn(). 3. Run TimeMarchesOn() at the beginning of play. r=rillian
MozReview-Commit-ID: BhwsIfRm3B2
