MSE: Speculative fix for flaky mediasource-errors layout test

MSE: Speculative fix for flaky mediasource-errors layout test

The media engine may progress beyond HAVE_METADATA readyState before
MEDIA_ERR_DECODE is visible to the web app.  This change speculatively
removes the assertion that readyState is still HAVE_METADATA to see if
this was indeed the cause of flaky mediasource-errors layout test.

R=philipj@opera.com
BUG=488184
TEST=http/tests/media/media-source/mediasource-errors.html

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=196386

[wolenetz, 2015-06-03 01:35:13]

Hi Philip - please take a look.
Thanks!
Matt

[wolenetz, 2015-06-03 01:48:07]

On 2015/06/03 01:35:13, wolenetz wrote:
> Hi Philip - please take a look.
> Thanks!
> Matt

In particular, I think the FlakyTests piece of this CL may be incorrect (since
it is updated by a tool according to the comment at the top of the file.)
I'm guessing the recommendation would be to *NOT* include either update
(FlakyTests or TestExpectations) in this CL since the fix is speculative.
I'll upload a second patch set shortly that contains just the speculative fix.

[wolenetz, 2015-06-03 01:51:22]

The CQ bit was checked by wolenetz@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-06-03 01:52:17]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1159163004/10004

[wolenetz, 2015-06-03 01:52:27]

On 2015/06/03 01:48:07, wolenetz wrote:
> On 2015/06/03 01:35:13, wolenetz wrote:
> > Hi Philip - please take a look.
> > Thanks!
> > Matt
> 
> In particular, I think the FlakyTests piece of this CL may be incorrect (since
> it is updated by a tool according to the comment at the top of the file.)
> I'm guessing the recommendation would be to *NOT* include either update
> (FlakyTests or TestExpectations) in this CL since the fix is speculative.
> I'll upload a second patch set shortly that contains just the speculative fix.

Sorry for the spam.
Please take a look at patch set 2.
I would also like some clarification about when and how FlakyTests should be
updated if this indeed fixes the flakiness.

[commit-bot: I haz the power, 2015-06-03 03:11:29]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-06-03 03:11:29]

Dry run: This issue passed the CQ dry run.

[philipj_slow, 2015-06-03 09:30:01]

The CQ bit was checked by philipj@opera.com

[philipj_slow, 2015-06-03 09:30:01]

lgtm

https://codereview.chromium.org/1159163004/diff/10004/LayoutTests/http/tests/...
File LayoutTests/http/tests/media/media-source/mediasource-errors.html (left):

https://codereview.chromium.org/1159163004/diff/10004/LayoutTests/http/tests/...
LayoutTests/http/tests/media/media-source/mediasource-errors.html:162:
assert_equals(mediaElement.readyState, HTMLMediaElement.HAVE_METADATA);
Yeah, this is a very likely cause, there's hardly a single case I can think of
where it's safe to assert anything about readyState or networkState in an event
handler.

[commit-bot: I haz the power, 2015-06-03 09:30:38]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1159163004/10004

[commit-bot: I haz the power, 2015-06-03 09:35:32]

Committed patchset #2 (id:10004) as
https://src.chromium.org/viewvc/blink?view=rev&revision=196386