Use shouldBecomeEqual in asynchronous spellchecking tests.

Use shouldBecomeEqual in asynchronous spellchecking tests.

Every asynchronous spellchecking layout test waits for spelling or grammar
markers using setTimeout. It duplicates the code, moreover the implementation
differs in every test.

Use shouldBecomeEqual helper (introduced in WebKit's r143860) to test
spellchecking asynchronously.

In grammar-paste.html, dump text of destination element instead of source one
to show what is really tested. It stops outputting that the input and textarea
elements can store html tags, for example,
PASS INPUT has a marker on 'I have a<b>n ki</b>wi'.

This is mostly merge of WebKit changeset: http://trac.webkit.org/changeset/156137

In addition, adjust the latest Blink specific spellchecking tests to start using
shouldBecomeEqual apart from spellcheck-editable-on-focus.html due to its flakiness.

Turn on asynchronous spellchecking at the beginning of spellcheck-paste-disabled.html
and spellcheck-paste-continuous-disabled.html. Previously, the feature was enabled in
the second part of the tests and first one used synchronous path of spell checking which
is going to be removed.

BUG=

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=158283

[groby-ooo-7-16, 2013-09-20 15:39:59]

One thing to keep in mind is that (IIRC!) Chromium doesn't actually use the
synchronous path any more. Until it's disabled, we should keep the tests, but
I'd rather not move additional testing onto the sync path.

[Hajime Morrita, 2013-09-20 15:54:04]

If we no longer use synchronous path and the async version is stable enough,
it might be better to just remove the sync path entirely. It will make whole
codebase much cleaner.

[groby-ooo-7-16, 2013-09-20 15:58:54]

I know - but I haven't looked at it in enough detail to know definitively if we
can. Filed http://crbug.com/295722 to track removal.

[tony, 2013-09-20 17:10:24]

LGTM with the following fixes.

https://codereview.chromium.org/23534071/diff/1/LayoutTests/editing/spelling/...
File LayoutTests/editing/spelling/grammar-markers.html (right):

https://codereview.chromium.org/23534071/diff/1/LayoutTests/editing/spelling/...
LayoutTests/editing/spelling/grammar-markers.html:27: return false;
Nit: Looks like the indent is off by one.

https://codereview.chromium.org/23534071/diff/1/LayoutTests/editing/spelling/...
File LayoutTests/editing/spelling/spelling-huge-text.html (right):

https://codereview.chromium.org/23534071/diff/1/LayoutTests/editing/spelling/...
LayoutTests/editing/spelling/spelling-huge-text.html:52:
shouldBecomeEqual('internals.markerCountForNode(testEditable.childNodes[0],
"spelling")', '6', finishJSTest);
Please clear the child node before calling finishJSTest so the expected.txt file
is not so large.

[grzegorz, 2013-09-24 13:42:20]

On 2013/09/20 15:58:54, groby wrote:
> I know - but I haven't looked at it in enough detail to know definitively if
we
> can. Filed http://crbug.com/295722 to track removal.

I could help in using asynchronous spell checking path in layout tests. There
are some left which use synchronous implementation.

[tony, 2013-09-24 16:48:55]

LGTM

[commit-bot: I haz the power, 2013-09-24 16:49:07]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/g.czajkowski@samsung.com/23534071/8001

[commit-bot: I haz the power, 2013-09-24 17:50:54]

Change committed as 158283