1293186 - currentTime can be different to seek time while seeking

Status: RESOLVED FIXED

(Core :: Audio/Video: Playback, defect)

Description

[Jean-Yves Avenard [:jya]]

Noticed during a compilation (-j14) and running the test:
http://www.w3c-test.org/media-source/mediasource-play-then-seek-back.html

failure was https://github.com/w3c/web-platform-tests/blob/master/media-source/mediasource-play-then-seek-back.html#L34

assert_equals(mediaElement.currentTime, 0.0, 'Current time is 0.0');

in the 'seeking' event handler.

What is the proper behaviour here. Is it correct to assume that currentTime will be the seek time as the seeking event is fired?

Reading the spec, I only see requirement for the currentTime to be stable within the same event loop

Comment 1

[Jean-Yves Avenard [:jya]]

Attachment: Bug 1293186: Do not assume that currentTime is still at seeking position when seeking event is fired.

The seeking attribute is changed synchronously, however, the seeking event is fired asynchronously. It is possible that by the time the seeking event is fired, the seeking operation has already completed and that the "time marches on", causing currentTime to be past the seeking position.

Review commit: https://reviewboard.mozilla.org/r/69956/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/69956/

Comment 2

[JW Wang [:jwwang] [:jw_wang]]

Comment on attachment 8778786 [details]
Bug 1293186: Do not assume that currentTime is still at seeking position when seeking event is fired.

https://reviewboard.mozilla.org/r/69956/#review67092

Comment 3

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/4fa9a49c57aa
Do not assume that currentTime is still at seeking position when seeking event is fired. r=jwwang

Comment 4

[Jean-Yves Avenard [:jya]]

Conclusion: the test is invalid, fixing the test.

Comment 5

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://hg.mozilla.org/mozilla-central/rev/4fa9a49c57aa