Bug 134976 – REGRESSION(r171033) [GTK] media/media-controls-invalid-url.html and media/video-fast-seek.html fail

• Component: Tools / Tests
• Status: RESOLVED FIXED

[Carlos Alberto Lopez Perez]

Revision r171033 <http://trac.webkit.org/r171033> has caused the 2 following tests to fail on the GTK port:

  media/media-controls-invalid-url.html [ Failure ]
    * Diff: http://build.webkit.org/results/GTK%20Linux%2064-bit%20Release%20%28Tests%29/r171040%20%281383%29/media/media-controls-invalid-url-pretty-diff.html
  media/video-fast-seek.html [ Failure ]
    * Diff: http://build.webkit.org/results/GTK%20Linux%2064-bit%20Release%20%28Tests%29/r171040%20%281383%29/media/video-fast-seek-pretty-diff.html


It also has caused the following diff on the test http/tests/w3c/webperf/approved/navigation-timing/html/test_timing_xserver_redirect.html : http://build.webkit.org/results/GTK%20Linux%2064-bit%20Release%20%28Tests%29/r171040%20%281383%29/http/tests/w3c/webperf/approved/navigation-timing/html/test_timing_xserver_redirect-pretty-diff.html 
But I understand this last one only needs to be rebaselined.

[Carlos Alberto Lopez Perez]

I double checked this: I tested to locally revert r171033 and this two test pass again.

I have marked this two test as failing on the GTK TestExpectations, as also rebaselined the test http/tests/w3c/webperf/approved/navigation-timing/html/test_timing_xserver_redirect.html on r171169 <http://trac.webkit.org/r171169>

[Michael Catanzaro]

media/media-controls-invalid-url.html is now timing out as well. Updating expectations accordingly.

[Ms2ger (he/him; ⌚ UTC+1/+2)]

(In reply to Carlos Alberto Lopez Perez from comment #0)
> Revision r171033 <http://trac.webkit.org/r171033> has caused the 2 following
> tests to fail on the GTK port:
> 
>   media/media-controls-invalid-url.html [ Failure ]

Current result:

This tests that media element controls are reset to their default state when the src is changed to an invalid url.

This test only runs in DRT!


EVENT(canplaythrough)
EXPECTED (getTimeLineValue() == '0') OK
EVENT(seeked)
EXPECTED (getTimeLineValue() == '1') OK
EVENT(error)
EXPECTED (getTimeLineValue() == '0'), OBSERVED '1' FAIL
END OF TEST


Two problems here:

- We don't update the displayed duration after an error event.
- Even if we did update it from the error handler, the failing check runs before mediaControls's error handler. (A setTimeout(0) in the test would probably be enough to fix that, though.)

Also note that the test fails the same way on mac: bug 67716.

[Ms2ger (he/him; ⌚ UTC+1/+2)]

Created attachment 318627 [details]
Reduced test case for media-controls-invalid-url.html

[Eric Carlson]

(In reply to Ms2ger from comment #3)
> 
> Two problems here:
> 
> - We don't update the displayed duration after an error event.
> - Even if we did update it from the error handler, the failing check runs
> before mediaControls's error handler. (A setTimeout(0) in the test would
> probably be enough to fix that, though.)
> 
> Also note that the test fails the same way on mac: bug 67716.

This test now passes on macOS.

[Diego Pino]

There are no references to this bug in any TestExpectations. It's probable this bug was solved at some point but it wasn't marked as closed. I'm closing this bug now. If you think this bug report is still valid, please reopen it and add an entry to TestExpectations.

[Radar WebKit Bug Importer]

<rdar://problem/104678737>