Convert before-*, broken-* and constructors.html tests to testharness.js

Convert before-*, broken-* and constructors.html tests to testharness.js

Cleaning up before-*, broken-* and constructors.html tests
in media/ to use testharness.js instead of video-test.js.
This will enable to upstream these tests to web-platform-tests.

BUG=588956
Committed: https://crrev.com/99f16ba7f1252ab27dab926d328421b5cdc55c36
Cr-Commit-Position: refs/heads/master@{#396447}

[Srirama, 2016-05-27 07:45:20]

Description was changed from

==========
fix

BUG=
==========

to

==========
Convert before-*, broken-* and constructors.html tests to testharness.js

Cleaning up before-*, broken-* and constructors.html tests in media/
to use testharness.js instead of video-test.js.
This will enable to upstream these tests to web-platform-tests.

BUG=588956
==========

[Srirama, 2016-05-27 07:45:51]

Description was changed from

==========
Convert before-*, broken-* and constructors.html tests to testharness.js

Cleaning up before-*, broken-* and constructors.html tests in media/
to use testharness.js instead of video-test.js.
This will enable to upstream these tests to web-platform-tests.

BUG=588956
==========

to

==========
Convert before-*, broken-* and constructors.html tests to testharness.js

Cleaning up before-*, broken-* and constructors.html tests 
in media/ to use testharness.js instead of video-test.js.
This will enable to upstream these tests to web-platform-tests.

BUG=588956
==========

[Srirama, 2016-05-27 07:47:13]

srirama.m@samsung.com changed reviewers:
+ foolip@chromium.org, fs@opera.com

[Srirama, 2016-05-27 07:47:14]

PTAL

[fs, 2016-05-27 08:10:48]

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
File third_party/WebKit/LayoutTests/media/broken-video.html (left):

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/media/broken-video.html:11:
testExpected("video.error", "[object MediaError]");
Not sure what this was intended to test, but we could keep it and test
toString() (like it does here) or the 'video.error instanceof MediaError'
perhaps? Pick one =)

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
File third_party/WebKit/LayoutTests/media/broken-video.html (right):

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/media/broken-video.html:9: var video =
document.querySelector("video");
Nit: Indentation

[foolip, 2016-05-27 08:22:15]

lgtm % nits (agree with fs's nits too)

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
File third_party/WebKit/LayoutTests/media/before-load-member-access.html
(right):

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/media/before-load-member-access.html:2:
<title>Test that accessing member of a non loaded video works.</title>
This test is a bit silly I think, it's a stretch to think that merely accessing
a member would throw an exception and fail this test. I mean there are such
attributes, but it's a strange thing to test in general.

Can you assert what the actual initial values are instead, so that it will be
obvious later that this can be folded into other tests for the respective
attributes?

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
File third_party/WebKit/LayoutTests/media/broken-video.html (right):

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/media/broken-video.html:12:
assert_equals(video.error.code, MediaError.MEDIA_ERR_SRC_NOT_SUPPORTED);
Add assert_true(video.error instanceof MediaError) as that was kind of tested
(via the stringifier) in the original test.

[fs, 2016-05-27 08:31:13]

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
File third_party/WebKit/LayoutTests/media/before-load-member-access.html
(right):

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/media/before-load-member-access.html:2:
<title>Test that accessing member of a non loaded video works.</title>
On 2016/05/27 at 08:22:15, Philip Jägenstedt wrote:
> This test is a bit silly I think, it's a stretch to think that merely
accessing a member would throw an exception and fail this test. I mean there are
such attributes, but it's a strange thing to test in general.

Well, the new test is at least "pretty silly" (or should that be "silly (but)
pretty") while the old one was just "silly" =P

[Srirama, 2016-05-27 10:57:49]

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
File third_party/WebKit/LayoutTests/media/before-load-member-access.html
(right):

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/media/before-load-member-access.html:2:
<title>Test that accessing member of a non loaded video works.</title>
On 2016/05/27 08:22:15, Philip Jägenstedt wrote:
> This test is a bit silly I think, it's a stretch to think that merely
accessing
> a member would throw an exception and fail this test. I mean there are such
> attributes, but it's a strange thing to test in general.
> 
> Can you assert what the actual initial values are instead, so that it will be
> obvious later that this can be folded into other tests for the respective
> attributes?

Except volume, others are TimeRange objects, so i just asserted the length.

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
File third_party/WebKit/LayoutTests/media/broken-video.html (right):

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/media/broken-video.html:9: var video =
document.querySelector("video");
On 2016/05/27 08:10:48, fs wrote:
> Nit: Indentation

Done.

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/media/broken-video.html:12:
assert_equals(video.error.code, MediaError.MEDIA_ERR_SRC_NOT_SUPPORTED);
On 2016/05/27 08:22:15, Philip Jägenstedt wrote:
> Add assert_true(video.error instanceof MediaError) as that was kind of tested
> (via the stringifier) in the original test.

Done.

[fs, 2016-05-27 11:04:37]

lgtm

[Srirama, 2016-05-27 11:25:53]

The CQ bit was checked by srirama.m@samsung.com

[Srirama, 2016-05-27 11:25:54]

The patchset sent to the CQ was uploaded after l-g-t-m from foolip@chromium.org
Link to the patchset: https://codereview.chromium.org/2019643002/#ps20001
(title: "address comments")

[commit-bot: I haz the power, 2016-05-27 11:26:09]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2019643002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/2019643002/20001

[foolip, 2016-05-27 11:29:25]

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
File third_party/WebKit/LayoutTests/media/before-load-member-access.html
(right):

https://codereview.chromium.org/2019643002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/media/before-load-member-access.html:2:
<title>Test that accessing member of a non loaded video works.</title>
On 2016/05/27 10:57:48, Srirama wrote:
> On 2016/05/27 08:22:15, Philip Jägenstedt wrote:
> > This test is a bit silly I think, it's a stretch to think that merely
> accessing
> > a member would throw an exception and fail this test. I mean there are such
> > attributes, but it's a strange thing to test in general.
> > 
> > Can you assert what the actual initial values are instead, so that it will
be
> > obvious later that this can be folded into other tests for the respective
> > attributes?
> 
> Except volume, others are TimeRange objects, so i just asserted the length.

Sounds good!

[commit-bot: I haz the power, 2016-05-27 12:33:08]

Description was changed from

==========
Convert before-*, broken-* and constructors.html tests to testharness.js

Cleaning up before-*, broken-* and constructors.html tests 
in media/ to use testharness.js instead of video-test.js.
This will enable to upstream these tests to web-platform-tests.

BUG=588956
==========

to

==========
Convert before-*, broken-* and constructors.html tests to testharness.js

Cleaning up before-*, broken-* and constructors.html tests 
in media/ to use testharness.js instead of video-test.js.
This will enable to upstream these tests to web-platform-tests.

BUG=588956
==========

[commit-bot: I haz the power, 2016-05-27 12:33:09]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-05-27 12:34:35]

Description was changed from

==========
Convert before-*, broken-* and constructors.html tests to testharness.js

Cleaning up before-*, broken-* and constructors.html tests 
in media/ to use testharness.js instead of video-test.js.
This will enable to upstream these tests to web-platform-tests.

BUG=588956
==========

to

==========
Convert before-*, broken-* and constructors.html tests to testharness.js

Cleaning up before-*, broken-* and constructors.html tests
in media/ to use testharness.js instead of video-test.js.
This will enable to upstream these tests to web-platform-tests.

BUG=588956

Committed: https://crrev.com/99f16ba7f1252ab27dab926d328421b5cdc55c36
Cr-Commit-Position: refs/heads/master@{#396447}
==========

[commit-bot: I haz the power, 2016-05-27 12:34:36]

Patchset 2 (id:??) landed as
https://crrev.com/99f16ba7f1252ab27dab926d328421b5cdc55c36
Cr-Commit-Position: refs/heads/master@{#396447}