Open Bug 208479 Opened 22 years ago Updated 2 years ago

Add Unicode 'char' to langGroup mapping

Categories

(Core :: Internationalization, defect)

defect

Tracking

()

People

(Reporter: jshin1987, Unassigned)

Details

Currently, GFX:GTK (both X11core, Xft) used fonts configured for the langGroup of the current locale when rendering Unicode documents without lang (html) and xml:lang (xml) specified. For instance, Tamil pages in UTF-8 without lang/xml:lang specified are rendered with fonts configured for Tamil (x-tamil langgroup) under 'ta-IN' locale. However, under 'en-US' locale, Tamil pages in UTF-8 are rendered with fonts configured for x-western langgroup. GFX:Win infers langGroup from the code range so that regardless of the current locale, Tamil pages are always rendered with fonts configured for Tamil (x-tamil) langGroup. (see bug 206123). We need to move up a couple of related files[1] in gfx/src/win to gfx/src/shared (or perhaps intl/locale [2] ) to share the code in GFX:Win used to map the code range to langGroup. (bug 206123) [1] http://lxr.mozilla.org/seamonkey/source/gfx/src/windows/nsUnicodeRange.cpp [2] If mapping code points to langGroup turns out to be useful outside GFX, we'd better put this somewhere in intl. In that case, platform should be 'All'.
Summary: GFX:GTK (X11core, Xft) need to select fonts based on the code range for Unicode documents → GFX:GTK/GFX:Xlib (X11core, Xft) need to select fonts based on the code range for Unicode documents
I vote for [2]. This mapping is pretty generic and has little to do with platform specific stuff.
OS: Linux → All
Hardware: PC → All
I agree with rbs and am changing the summary line. It seems like we have to add a method to nsILanguageAtomService [1] that returns 'langGroup' given a Unicode character (and possibly a Unicode range). [1] http://lxr.mozilla.org/seamonkey/source/intl/locale/idl/nsILanguageAtomService.idl
Status: NEW → ASSIGNED
Summary: GFX:GTK/GFX:Xlib (X11core, Xft) need to select fonts based on the code range for Unicode documents → Add Unicode 'char' to langGroup mapping
This is also useful when NCRs are used to represent characters not covered by the current document encoding. For instance, an ISO-8859-1 document can be used to represent any Unicode character. So, our fallback of inferring 'langGroup' from the page encoding completely breaks down in that case.
QA Contact: amyy → i18n

The bug assignee didn't login in Bugzilla in the last 7 months, so the assignee is being reset.

Assignee: jshin1987 → nobody
Status: ASSIGNED → NEW
Severity: normal → S3
You need to log in before you can comment on or make changes to this bug.