When progressive download of the media is enabled it is currently impossible to seek pas to the maxTimeLoaded value because maxTimeSeekable() returns it. Making that method return the value of duration() should fix the issue.
Created attachment 59904 [details] proposed patch
Works good for me, thanks. Please get this into the next 1.2 release too
Comment on attachment 59904 [details] proposed patch Makes sense to me.
Landed in http://trac.webkit.org/changeset/62371 Thanks!
http://trac.webkit.org/changeset/62371 might have broken GTK Linux 32-bit Release
Breaks media/controls-drag-timebar.html on 32-Bit GTK Release. --- /var/lib/buildbot/build/gtk-linux-32-release/build/layout-test-results/media/controls-drag-timebar-expected.txt 2010-07-02 02:44:46.000000000 -0700 +++ /var/lib/buildbot/build/gtk-linux-32-release/build/layout-test-results/media/controls-drag-timebar-actual.txt 2010-07-02 02:44:46.000000000 -0700 @@ -8,6 +8,6 @@ EVENT(seeked) Time: 2.7 EVENT(seeked) -Time: 3.2 +Time: 3.5 END OF TEST
Comment on attachment 59904 [details] proposed patch Clearing flags
(In reply to comment #6) > Breaks media/controls-drag-timebar.html on 32-Bit GTK Release. > And of course this test passes fine for me :P I'll try to reproduce it on my laptop and investigate the failure if I reproduce it. Otherwise I'd appreciate a gst debug log from kov's bot when it runs DRT with the patch applied for the failing test.
I was able to reproduce the failure on my laptop. The patch introduced a change of behavior of updateStates() that triggered the issue. A new patch is coming ;)
Created attachment 60365 [details] proposed patch
Comment on attachment 60365 [details] proposed patch OK!
Landed as http://trac.webkit.org/changeset/62561 with a new comment, as discussed with Xan. Thanks!