1038494 - Test case for track interfaces when consuming a MediaStream

Status: RESOLVED FIXED

(Core :: Audio/Video, defect)

Description

[Shelly Lin [:shelly]]

The current test case (test_mediatrack_events_and_consuming_ms.html) is kindda fragile, so re write it.

Comment 1

[Shelly Lin [:shelly]]

Attachment: Rewrite test cases of track interfaces when consuming a media stream

Though consuming a media stream won't meet the case where playback has ended before events are fired, I think it would be better to make the two test cases as identical as possible, since they are testing the same scenario but consuming different resources. I also break out the part of testing event identities, mixing two testing goal into one test seem not good.

Comment 2

[Robert O'Callahan (:roc) (email my personal email if necessary)]

Comment on attachment 8456049 [details] [diff] [review]
Rewrite test cases of track interfaces when consuming a media stream

Review of attachment 8456049 [details] [diff] [review]:
-----------------------------------------------------------------

Sorry about the delay

::: content/media/test/test_mediatrack_events.html
@@ +18,5 @@
> +            'HTMLMediaElement::AudioTracks() property should be available.');
> +      isnot(element.videoTracks, undefined,
> +            'HTMLMediaElement::VideoTracks() property should be available.');
> +
> +      function verify_event(e, type) {

verifyEvent

Comment 3

[Shelly Lin [:shelly]]

Attachment: (revised)Rewrite test cases of track interfaces when consuming a media stream

This patch has updated the function name to verifyEvent.
Thanks!

Comment 4

[Shelly Lin [:shelly]]

Try result:
https://tbpl.mozilla.org/?tree=Try&rev=289780b43c0a
https://tbpl.mozilla.org/?tree=Try&rev=2edbafcf0cb1

Comment 5

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/dec3449b055e

Comment 6

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

https://hg.mozilla.org/mozilla-central/rev/dec3449b055e

Comment 7

[Shelly Lin [:shelly]]

Attachment: New fix for this test case

Comment 8

[Shelly Lin [:shelly]]

Comment on attachment 8465279 [details] [diff] [review]
New fix for this test case

Review of attachment 8465279 [details] [diff] [review]:
-----------------------------------------------------------------

::: content/media/test/test_mediatrack_consuming_mediastream.html
@@ -52,2 @@
>            is(element.audioTracks.length, 0, 'The length of audioTracks should be 0.');
> -          is(videoOnremovetrack, 1, 'Calls of onremovetrack on videoTracks should be 1.');

I think it should be fine to remove the check of onremovetrack on video element, because 1. It should really be checked at the onended event handler of MediaStream or the onended event handler of MediaStreamTrack, and we implement them neither, and 2. We are still checking the length of video.audioTracks and video.videoTracks.

Comment 9

[Shelly Lin [:shelly]]

Try server result with failing platform:
(Android 2.3 Emulator try opt test mochitest-4 )
https://tbpl.mozilla.org/?tree=Try&rev=0bc5c4782685

(b2g_emulator_vm try opt test mochitest-3)
https://tbpl.mozilla.org/?tree=Try&rev=72b4bd387fe4

Comment 10

[Shelly Lin [:shelly]]

Move the fix to bug 1047307.