Fixing inconsistency with the media query spec and other browsers

Fixing inconsistency with the media query spec and other browsers
    
With this change we are more interoperable with Opera and Firefox
and we have better test coverage. It fixes some tests which are not
correct according to the specs and other implementations.

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=150435

[eseidel, 2013-05-14 21:55:58]

lgtm

Seems fine.  This code still mallocs a lot. :)  But this looks more readable.

[commit-bot: I haz the power, 2013-05-14 21:56:08]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/kenneth.r.christiansen@intel.com/14578...

[kenneth.r.christiansen, 2013-05-14 22:44:24]

My change should be fine with the spec as an empty media query is equal to "all"
but this requires too much rebasing, but let me change.

[commit-bot: I haz the power, 2013-05-14 22:56:08]

Retried try job too often on linux_layout_rel for step(s) webkit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_layo...

[kenneth.r.christiansen, 2013-05-15 10:32:09]

https://codereview.chromium.org/14578010/diff/25001/LayoutTests/fast/media/mq...
File LayoutTests/fast/media/mq-parsing-expected.txt (right):

https://codereview.chromium.org/14578010/diff/25001/LayoutTests/fast/media/mq...
LayoutTests/fast/media/mq-parsing-expected.txt:6: FAIL: Got "not all, screen,
not all, not all, not all, not all, not all" but was expecting "not all, screen,
not all, not all, not all, not all"
These fails are expected until we fix the splitting algorithm

[rune, 2013-05-15 11:28:55]

https://codereview.chromium.org/14578010/diff/25001/Source/core/css/MediaList...
File Source/core/css/MediaList.cpp (right):

https://codereview.chromium.org/14578010/diff/25001/Source/core/css/MediaList...
Source/core/css/MediaList.cpp:151: // other allowed matching pairs such as (),
[], {}, "", and ''.
Yes, ideally CSSParser should have a parseMediaQueryList() with an
@-webkit-mediaquerylist production, but not until the error recovery in the
parser is better.

[apavlov, 2013-05-15 12:28:36]

https://codereview.chromium.org/14578010/diff/25001/Source/core/css/MediaList...
File Source/core/css/MediaList.cpp (right):

https://codereview.chromium.org/14578010/diff/25001/Source/core/css/MediaList...
Source/core/css/MediaList.cpp:125: ASSERT(!queryString.isEmpty());
I cannot see any code that might guarantee the queryString is not empty
(consider the ",,,," case from the original code comment)

https://codereview.chromium.org/14578010/diff/25001/Source/core/css/MediaList...
Source/core/css/MediaList.cpp:159: if (OwnPtr<MediaQuery> parsedQuery =
parseMediaQuery(queryString))
Shouldn't the code check that queryString is not empty before calling this?

[kenneth.r.christiansen, 2013-05-15 14:25:27]

> Source/core/css/MediaList.cpp:125: ASSERT(!queryString.isEmpty());
> I cannot see any code that might guarantee the queryString is not empty
> (consider the ",,,," case from the original code comment)

I forgot to remove the assert after changing the code. Removed now, thanks for
catching.

[apavlov, 2013-05-15 15:33:32]

lgtm

[commit-bot: I haz the power, 2013-05-15 15:33:40]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/kenneth.r.christiansen@intel.com/14578...

[commit-bot: I haz the power, 2013-05-15 17:20:15]

Change committed as 150435