Handle U+3099/309A Kana Voiced Sound Marks in complex path

Handle U+3099/309A Kana Voiced Sound Marks in complex path

U+3099/309A Combining Katakana-Hiragana Voiced/Semi-voiced Sound Marks
are one of few Nonspacing-Mark characters in East Asia. They are handled
as a special case of simple path by normalizing to pre-composed
characters.

This patch eliminates the special casing and use the complex path for
these code point, in order to have less special casing when the simple
path is unified with complex path.

BUG=507351
TEST=fast/text/international/kana-voiced-sound-marks.html

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=199152

[kojii, 2015-07-17 15:00:46]

kojii@chromium.org changed reviewers:
+ drott@chromium.org, eae@chromium.org

[kojii, 2015-07-17 15:00:47]

PTAL.

[drott, 2015-07-17 15:15:58]

Nice, always enjoying a large portion of removed lines in a patch. :-) LGTM with
a minor remark regarding the comment.

https://codereview.chromium.org/1227883003/diff/40001/Source/platform/fonts/U...
File Source/platform/fonts/UTF16TextIterator.h (right):

https://codereview.chromium.org/1227883003/diff/40001/Source/platform/fonts/U...
Source/platform/fonts/UTF16TextIterator.h:49: // TODO(kojii): This looks broken,
we should actually check whether
Minor: I'd suggest to move this comment to a bug report, e.g. "Check
UTF16TextITerator's category mask behavior" and explain the overlap with
HarfBuzz there. In my opinion, here in the code it does not add much without
knowing the context.

[kojii, 2015-07-17 16:32:59]

Thank you for the prompt review. Yeah, I've been wanting to clean this up for a
while, thank you for making complex path better to make the removal possible!

https://codereview.chromium.org/1227883003/diff/40001/Source/platform/fonts/U...
File Source/platform/fonts/UTF16TextIterator.h (right):

https://codereview.chromium.org/1227883003/diff/40001/Source/platform/fonts/U...
Source/platform/fonts/UTF16TextIterator.h:49: // TODO(kojii): This looks broken,
we should actually check whether
On 2015/07/17 15:15:58, drott wrote:
> Minor: I'd suggest to move this comment to a bug report, e.g. "Check
> UTF16TextITerator's category mask behavior" and explain the overlap with
> HarfBuzz there. In my opinion, here in the code it does not add much without
> knowing the context.

Moved to crbug.com/511306, thanks for the advice!

[kojii, 2015-07-18 04:48:03]

The CQ bit was checked by kojii@chromium.org

[kojii, 2015-07-18 04:48:04]

The patchset sent to the CQ was uploaded after l-g-t-m from drott@chromium.org
Link to the patchset: https://codereview.chromium.org/1227883003/#ps60001
(title: "TODO comment moved to a bug")

[commit-bot: I haz the power, 2015-07-18 04:48:11]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1227883003/60001

[commit-bot: I haz the power, 2015-07-18 04:57:01]

Committed patchset #4 (id:60001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=199152

[eae, 2015-07-20 17:48:26]

LGTM (post facto)