[Webkit-unassigned] [Bug 93832] text-combine falls back to none rather than scale when wider than 1.1em

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=93832





--- Comment #8 from Yuki Sekiguchi <yuki.sekiguchi at access-company.com>  2012-08-15 02:03:35 PST ---
Thank you, Ishii-san for good advice.

Fixed typos..


> * ChangeLog: I think you should mention it's not conformant with the current WD, along with the spec URL, rather than just saying "annoying for many Japanese users".
Fixed.

> * *-expected.png: Are they correct? I guess expectations are to text-combine with scale, no?
*-expected.png is old one.
I just delete the png.

"run-webkit-test" don't execute pixel test.
"run-webkit-test -p" execute pixel test.
But the expectation file may not be for Lion because their scrollbar is Aqua style.
I understand that pixel test is not maintained or cannot generate expectation on Lion.
So I just delete the png.

> * You're restoring CTM by multiplying (1 / combinedComplessScale), but I don't think it's reliable nor fast. I think it's better to save/restore.
I think so.
Fixed.

> * Is "m_emWidth = description.computedSize() * textCombineMargin" correct? We should scale to 1em, not to 1.1em, correct?
Thank you!
It is not correct.
I fixed it.
I don't understand why original author added 10% margin from the original ticket, but it may be to avoid fall back to none.
https://bugs.webkit.org/show_bug.cgi?id=50621
I remove the margin.

> * I think you should leave a comment in source file too why you always set m_isCombined to true, and fallback code should be removed. I actually think it might be better to remove them, because it reduces footprint, but I can't be sure how reviewers would say, so you may wait for the real reviewer, up to you.
Add comment.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.