Open Bug 425650 Opened 17 years ago Updated 2 years ago

Combining diacritics misplaced when no precomposed glyph for base+diacritic pair

Categories

(Core :: Graphics: Text, defect)

Product:
Core
Component:
Graphics: Text
Platform:
All
macOS
Type:
defect

Tracking

()

People

(Reporter: mozilla, Unassigned)

References

(
URL
)

Details

Follow-up from bug 404848: Mac has similar problems with Unicode diacritics as demonstrated in attachment 311959 [details].
The Windows bug was blocking, requesting the same here.
Flags: blocking1.9?
Isn't this bug 165881? I'm not on the right Mac atm to check, but that seems to be what the other bug is talking about. (Or is this a regression from the ones that were working?)
Yeah, reading John's bug 404848 comment 18, the last paragraph, is exactly the issue that's holding bug 165881 open (see smontagu's bug 165881 comment 18). We can either dupe this, or use this opportunity to create a relatively clean bug with a summary that specifically identifies the remaining problem with combining diacritics.
i don't think we'll block on this.
Flags: blocking1.9? → blocking1.9-
Going ahead and tidying this up to be a clear bug about this problem (which is the remaining problem from bug 165881; I'll WFM that one). See bug 404848 comment 18 and bug 165881 comment 18 for more analysis of this issue.
Summary: Some Latin diacritics misplaced → Combining diacritics misplaced when no precomposed glyph for base+diacritic pair
Er, I just noticed after WFMing it that bug 165881 had wanted-1.9 on it (added after the diagnosis of this problem as the remaining issue); Stuart, can you add that here (if it's still warranted)?
I'd really like to get this fixed. It's a huge bug in what is really a new feature for 1.9.
This is the mac version of this bug and the problem here has significantly less impact, it only affects unusual usage of these combining characters. This *only* applies to combinations for which a combined codepoints does not exist - so a + umlaut will render but q + umlaut won't. This is going to be very tough to fix because ATS is basically lying to us, it fabricates the cmap table it hands us and sticks in the combining character codepoints. This is the jist of bug 404848 comment 18. So we basically have no way to distinguishing between fonts that have or don't have glyphs for these codepoints, which sucks. I don't see a clear way of fixing this, certainly not in the 1.9 timeframe. But I would love it if someone could prove me wrong...
Severity: normal → S3
Component: Graphics → Graphics: Text
You need to log in before you can comment on or make changes to this bug.