Fix inconsistencies wrt whitespaces.

Fix inconsistencies wrt whitespaces.

\u0085 (NEL) is now considered a whitespace in accordance to http://www.unicode.org/Public/6.3.0/ucd/PropList.txt

R=mstarzinger@chromium.org
BUG=v8:3109
LOG=Y

Committed: https://code.google.com/p/v8/source/detail?r=19196

[Michael Starzinger, 2014-02-07 10:02:38]

https://codereview.chromium.org/146983007/diff/1/src/scanner.h
File src/scanner.h (right):

https://codereview.chromium.org/146983007/diff/1/src/scanner.h#newcode143
src/scanner.h:143: return kIsWhiteSpace.get(c) || c == 0xFEFF;
This is kind of brittle! Can we please move this special casing into a
char-predicate (e.g struct WhiteSpace) and use that predicate in the cache
instead of special-casing after the cache lookup?

https://codereview.chromium.org/146983007/diff/1/test/mjsunit/whitespaces.js
File test/mjsunit/whitespaces.js (right):

https://codereview.chromium.org/146983007/diff/1/test/mjsunit/whitespaces.js#...
test/mjsunit/whitespaces.js:1: // Copyright 2012 the V8 project authors. All
rights reserved.
nit: 2014

[Yang, 2014-02-07 12:11:11]

On 2014/02/07 10:02:38, Michael Starzinger wrote:
> https://codereview.chromium.org/146983007/diff/1/src/scanner.h
> File src/scanner.h (right):
> 
> https://codereview.chromium.org/146983007/diff/1/src/scanner.h#newcode143
> src/scanner.h:143: return kIsWhiteSpace.get(c) || c == 0xFEFF;
> This is kind of brittle! Can we please move this special casing into a
> char-predicate (e.g struct WhiteSpace) and use that predicate in the cache
> instead of special-casing after the cache lookup?
> 
> https://codereview.chromium.org/146983007/diff/1/test/mjsunit/whitespaces.js
> File test/mjsunit/whitespaces.js (right):
> 
>
https://codereview.chromium.org/146983007/diff/1/test/mjsunit/whitespaces.js#...
> test/mjsunit/whitespaces.js:1: // Copyright 2012 the V8 project authors. All
> rights reserved.
> nit: 2014

Addressed comments. Had to fix an outdated third-party test.

[Michael Starzinger, 2014-02-07 12:25:49]

LGTM (with one more suggestion).

https://codereview.chromium.org/146983007/diff/140001/test/cctest/test-regexp.cc
File test/cctest/test-regexp.cc (right):

https://codereview.chromium.org/146983007/diff/140001/test/cctest/test-regexp...
test/cctest/test-regexp.cc:448: return
CcTest::i_isolate()->unicode_cache()->IsWhiteSpace(c);
nit: Would it be possible to go directly to the v8::internal::WhiteSpace
predicate here? One less piece than can go wrong when computing the reference
value.

[Yang, 2014-02-07 12:34:24]

https://codereview.chromium.org/146983007/diff/140001/test/cctest/test-regexp.cc
File test/cctest/test-regexp.cc (right):

https://codereview.chromium.org/146983007/diff/140001/test/cctest/test-regexp...
test/cctest/test-regexp.cc:448: return
CcTest::i_isolate()->unicode_cache()->IsWhiteSpace(c);
On 2014/02/07 12:25:49, Michael Starzinger wrote:
> nit: Would it be possible to go directly to the v8::internal::WhiteSpace
> predicate here? One less piece than can go wrong when computing the reference
> value.

Done.

[Yang, 2014-02-07 12:34:55]

Committed patchset #4 manually as r19196 (presubmit successful).