The WebVTT parsing algorithm allows empty cue text

The WebVTT parsing algorithm allows empty cue text

Adjust tests:
  tc027-empty-cue - PASS -> FAIL, because opposite assumptions
  opera/track/webvtt/parsing:newlines.vtt - one more cue (empty) observed
  opera/interfaces/TextTrackCue/text.html - FAIL -> PASS

(The newlines.vtt file is truncated compared to the
original/upstream version, so chose to update the
expected result rather than modifying the test. Aim to
update the testfile at a later point.)

BUG=305317
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=161924

[jochen (gone - plz use gerrit), 2013-11-11 12:27:27]

https://codereview.chromium.org/62833005/diff/1/LayoutTests/media/track/opera...
File LayoutTests/media/track/opera/track/webvtt/parsing/001-expected.txt
(right):

https://codereview.chromium.org/62833005/diff/1/LayoutTests/media/track/opera...
LayoutTests/media/track/opera/track/webvtt/parsing/001-expected.txt:33: FAIL
WebVTT parser tests, newlines.vtt assert_equals: number of cues expected 4 but
got 3
why do you update the baselines instead of updating the tests?

[fs, 2013-11-11 12:53:51]

On 2013/11/11 12:27:27, jochen wrote:
>
https://codereview.chromium.org/62833005/diff/1/LayoutTests/media/track/opera...
> File LayoutTests/media/track/opera/track/webvtt/parsing/001-expected.txt
> (right):
> 
>
https://codereview.chromium.org/62833005/diff/1/LayoutTests/media/track/opera...
> LayoutTests/media/track/opera/track/webvtt/parsing/001-expected.txt:33: FAIL
> WebVTT parser tests, newlines.vtt assert_equals: number of cues expected 4 but
> got 3
> why do you update the baselines instead of updating the tests?

In this particular case, because it's an imported test (minor), and that the
testfile for this particular file is broken (truncated at some point - probably
at initial import?). So it seemed wrong to fix the test because of this
particular change. I'd like to import the "correct" (untruncated) .vtt file for
the test, but since it's one of those tests which is supposed to have mixed line
endings I didn't want to deal with it here...

[jochen (gone - plz use gerrit), 2013-11-11 12:54:46]

lgtm

can you mention this in the CL description please?

[fs, 2013-11-11 13:09:28]

On 2013/11/11 12:54:46, jochen wrote:
> lgtm
> 
> can you mention this in the CL description please?

CL description updated.

[commit-bot: I haz the power, 2013-11-11 13:09:53]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/fs@opera.com/62833005/1

[philipj_slow, 2013-11-13 14:16:40]

https://codereview.chromium.org/62833005/diff/1/LayoutTests/media/track/opera...
File LayoutTests/media/track/opera/track/webvtt/parsing/001-expected.txt
(right):

https://codereview.chromium.org/62833005/diff/1/LayoutTests/media/track/opera...
LayoutTests/media/track/opera/track/webvtt/parsing/001-expected.txt:33: FAIL
WebVTT parser tests, newlines.vtt assert_equals: number of cues expected 4 but
got 3
On 2013/11/11 12:27:28, jochen wrote:
> why do you update the baselines instead of updating the tests?

Fixing this is going to require some Subversion magic, I'm taking care of it in
https://code.google.com/p/chromium/issues/detail?id=318734

[commit-bot: I haz the power, 2013-11-13 14:21:14]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/fs@opera.com/62833005/1

[commit-bot: I haz the power, 2013-11-13 15:45:49]

Change committed as 161924