[webkit-reviews] review requested: [Bug 15790] Ligatures aren't accounted for when manipulating VisiblePositions : [Attachment 28465] another proposed fix

[bugzilla-daemon at webkit.org]

Hironori Bono <hbono at chromium.org> has asked  for review:
Bug 15790: Ligatures aren't accounted for when manipulating VisiblePositions
https://bugs.webkit.org/show_bug.cgi?id=15790

Attachment 28465: another proposed fix
https://bugs.webkit.org/attachment.cgi?id=28465&action=review

------- Additional Comments from Hironori Bono <hbono at chromium.org>
Sorry for my fix for bug 24342 which caused a conflict with this bug. I added
some rules which prevent a cursor from moving after Indic virama signs.
Nevertheless, I'm a little wondering if this change covers the all possible
combinations for Indic ligatures. I read Unicode spec and investigated Firefox,
IE, TextEdit, and NeoOffice. But their cursor behaviors are not completely the
same. For example, IE prevents a cursor from moving in the middle of a
Devanagari ligature "&#x0915;&#x094D;&#x0915;", written in Table 9-3 of Unicode
specification 5.0, but Firefox, TextEdit, and NeoOffice premit.
Even though this change emulates IE for such cases, please feel free to blame
me if this is not your expected behavior.