VTTCue: Add TypeChecking=Unrestricted

VTTCue: Add TypeChecking=Unrestricted

Enables proper checking of the startTime and endTime parameters to the
constructor.
Fix one test (media/track/track-cue-negative-timestamp.html) which
previously tested the wrong thing, and drop one subtest of
media/track/track-add-remove-cue.html since it's pretty useless.

BUG=448000
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=188621

[fs, 2015-01-15 18:11:27]

fs@opera.com changed reviewers:
+ philipj@opera.com

[philipj_slow, 2015-01-19 08:49:36]

lgtm % nits

https://codereview.chromium.org/854863002/diff/1/LayoutTests/media/track/oper...
File LayoutTests/media/track/opera/interfaces/VTTCue/constructor.html (right):

https://codereview.chromium.org/854863002/diff/1/LayoutTests/media/track/oper...
LayoutTests/media/track/opera/interfaces/VTTCue/constructor.html:10:
assert_throws(new TypeError, function() { new VTTCue('tomorrow', 'christmas',
'foo'); });
s/christmas/0/ unless I'm missing a reason.

https://codereview.chromium.org/854863002/diff/1/LayoutTests/media/track/trac...
File LayoutTests/media/track/track-add-remove-cue.html (right):

https://codereview.chromium.org/854863002/diff/1/LayoutTests/media/track/trac...
LayoutTests/media/track/track-add-remove-cue.html:68: testException("oldStyleCue
= new VTTCue('sausage-cue', 33, 3.4, 'Sausage?')", '"TypeError: Failed to
construct \'VTTCue\': double parameter 1 is non-finite."');
Drop the 'sausage-cue' arguemnt and s/33/NaN/ to avoid giving the impression
that the 4th argument does something?

[fs, 2015-01-19 10:18:16]

https://codereview.chromium.org/854863002/diff/1/LayoutTests/media/track/oper...
File LayoutTests/media/track/opera/interfaces/VTTCue/constructor.html (right):

https://codereview.chromium.org/854863002/diff/1/LayoutTests/media/track/oper...
LayoutTests/media/track/opera/interfaces/VTTCue/constructor.html:10:
assert_throws(new TypeError, function() { new VTTCue('tomorrow', 'christmas',
'foo'); });
On 2015/01/19 08:49:36, philipj wrote:
> s/christmas/0/ unless I'm missing a reason.

Yes, this was just my fun-loving (pun-loving?) side getting the upper hand.
Fixed.

https://codereview.chromium.org/854863002/diff/1/LayoutTests/media/track/trac...
File LayoutTests/media/track/track-add-remove-cue.html (right):

https://codereview.chromium.org/854863002/diff/1/LayoutTests/media/track/trac...
LayoutTests/media/track/track-add-remove-cue.html:68: testException("oldStyleCue
= new VTTCue('sausage-cue', 33, 3.4, 'Sausage?')", '"TypeError: Failed to
construct \'VTTCue\': double parameter 1 is non-finite."');
On 2015/01/19 08:49:36, philipj wrote:
> Drop the 'sausage-cue' arguemnt and s/33/NaN/ to avoid giving the impression
> that the 4th argument does something?

Based on "Create an old-style cue with an id" just above, I think this is
supposed to be a valid constructor call per some older spec. version - hence
making it valid would seem to miss the point. If we are to change it any way,
than I'd rather just remove it - we have coverage on argument checking
elsewhere.

[philipj_slow, 2015-01-19 10:31:10]

https://codereview.chromium.org/854863002/diff/1/LayoutTests/media/track/trac...
File LayoutTests/media/track/track-add-remove-cue.html (right):

https://codereview.chromium.org/854863002/diff/1/LayoutTests/media/track/trac...
LayoutTests/media/track/track-add-remove-cue.html:68: testException("oldStyleCue
= new VTTCue('sausage-cue', 33, 3.4, 'Sausage?')", '"TypeError: Failed to
construct \'VTTCue\': double parameter 1 is non-finite."');
On 2015/01/19 10:18:16, fs wrote:
> On 2015/01/19 08:49:36, philipj wrote:
> > Drop the 'sausage-cue' arguemnt and s/33/NaN/ to avoid giving the impression
> > that the 4th argument does something?
> 
> Based on "Create an old-style cue with an id" just above, I think this is
> supposed to be a valid constructor call per some older spec. version - hence
> making it valid would seem to miss the point. If we are to change it any way,
> than I'd rather just remove it - we have coverage on argument checking
> elsewhere.

Oh, right. Seems like a pretty useless test to keep around, so removing is OK by
me.

[fs, 2015-01-19 11:42:44]

https://codereview.chromium.org/854863002/diff/1/LayoutTests/media/track/trac...
File LayoutTests/media/track/track-add-remove-cue.html (right):

https://codereview.chromium.org/854863002/diff/1/LayoutTests/media/track/trac...
LayoutTests/media/track/track-add-remove-cue.html:68: testException("oldStyleCue
= new VTTCue('sausage-cue', 33, 3.4, 'Sausage?')", '"TypeError: Failed to
construct \'VTTCue\': double parameter 1 is non-finite."');
On 2015/01/19 10:31:10, philipj wrote:
> On 2015/01/19 10:18:16, fs wrote:
> > On 2015/01/19 08:49:36, philipj wrote:
> > > Drop the 'sausage-cue' arguemnt and s/33/NaN/ to avoid giving the
impression
> > > that the 4th argument does something?
> > 
> > Based on "Create an old-style cue with an id" just above, I think this is
> > supposed to be a valid constructor call per some older spec. version - hence
> > making it valid would seem to miss the point. If we are to change it any
way,
> > than I'd rather just remove it - we have coverage on argument checking
> > elsewhere.
> 
> Oh, right. Seems like a pretty useless test to keep around, so removing is OK
by
> me.

Removed. Description updated.

[philipj_slow, 2015-01-19 12:06:47]

Still LGTM.

[fs, 2015-01-19 12:25:36]

The CQ bit was checked by fs@opera.com

[commit-bot: I haz the power, 2015-01-19 12:25:50]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/854863002/40001

[commit-bot: I haz the power, 2015-01-19 12:54:19]

Committed patchset #3 (id:40001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=188621