[webkit-reviews] review requested: [Bug 48860] U+FEFF is rendered as an empty box in a complex script page on Windows : [Attachment 90154] updated patch addressing Levin's comment

[bugzilla-daemon at webkit.org]

Jungshik Shin <jshin at chromium.org> has asked  for review:
Bug 48860: U+FEFF is rendered as an empty box in a complex script page on
Windows
https://bugs.webkit.org/show_bug.cgi?id=48860

Attachment 90154: updated patch addressing Levin's comment
https://bugs.webkit.org/attachment.cgi?id=90154&action=review

------- Additional Comments from Jungshik Shin <jshin at chromium.org>
Thank you for the review. I addressed all the review comments. 

(9, 10, 13) are excluded because they're turned to an actual space before
reaching the code in question per HTML(5). Well, you may ask why they're
included in treatAsZeroWidthSpace(InComplexScript) in the first place. Perhaps,
they shouldn't be included, but I haven't checked other uses of
treatAsZeroWidthSpace and leaving it alone in this patch. 

As for not testing other characters in treatAsZeroWidthSpaceInComplexScript,
it's an oversight because I made a new layout test by modifying
zero-width-characters.html and didn't notice that the test does not test for
U+007F - U+009F/U+00AD and U+202A - U+202E. 

I guess the latter block is not tested in zero-width-characters.html because
they're BiDi controls and it can mess up the width measurement. I think I can
rewrite zero-width-characters.html to test them as well. 

As for the former, I think it's just an oversight. 

I'll revise that test in a separate CL to address both issues.