Bug 467982
| Summary: | anaconda doesn't render glyphs with the proper font | ||||||
|---|---|---|---|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Akira TAGOH <tagoh> | ||||
| Component: | anaconda | Assignee: | Anaconda Maintenance Team <anaconda-maint-list> | ||||
| Status: | CLOSED INSUFFICIENT_DATA | QA Contact: | Fedora Extras Quality Assurance <extras-qa> | ||||
| Severity: | medium | Docs Contact: | |||||
| Priority: | medium | ||||||
| Version: | 10 | CC: | anaconda-maint-list, behdad, dcantrell, fonts-bugs, i18n-bugs, nicolas.mailhot, petersen, tcallawa, t.matsuu | ||||
| Target Milestone: | --- | Keywords: | Regression, Reopened | ||||
| Target Release: | --- | ||||||
| Hardware: | i386 | ||||||
| OS: | Linux | ||||||
| Whiteboard: | |||||||
| Fixed In Version: | Doc Type: | Bug Fix | |||||
| Doc Text: | Story Points: | --- | |||||
| Clone Of: | Environment: | ||||||
| Last Closed: | 2009-06-18 17:55:35 UTC | Type: | --- | ||||
| Regression: | --- | Mount Type: | --- | ||||
| Documentation: | --- | CRM: | |||||
| Verified Versions: | Category: | --- | |||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||
| Embargoed: | |||||||
| Bug Depends On: | |||||||
| Bug Blocks: | 469046 | ||||||
| Attachments: |
|
||||||
What fonts should we be using for Japanese now? VL-Gothic-Regular.ttf in VLGothic-fonts package. Hm, that's definitely the font we are including. This happens when running anaconda on a regular system in test mode too. We've merged a lot of Asian (Korean...) fonts lately, maybe it's one of them You really need to have Behdad audit the fontconfig priority level given to all recent CJK fonts (and I include fonts in the pipeline like hanazono). For example, I've seen CJK packages that proposed 59 as priority, when our guidelines (which match upstream's) state the highest priority for a non LGC font should be 65. I hope we'll not repeat the arms race between Chinese/Japanese/Korean users where each one tries to one-up the other and put himself higher in the priority queue. This only seems to affect anaconda, so I don't see why this should be a fontconfig issue. Jens - so, any idea what could be going on here? I'm not exactly a font expert so I'm at a bit of a loss as to what could be the cause of this problem. I guess pango is not seeing the locale information somehow so it doesn't know what CJK font to choose. As I recall this regression started happening (very roughly) a week or more before this bug was filed. Fixed for me in anaconda-11.4.1.57-1. Not sure if I tested well enough, or what happened, but this is back in f10-final unfortunately. :-( Yeah my bad I retested anaconda-11.4.1.57-1 and it was broken too. Actually testing more I realise this is not a new bug: it affects Chinese fonts in F8 and F9 anaconda too. Now in F10 Chinese is ok but Japanese broken instead - presumably the fontconfig default priorities changed that. But this needs to be fixed by some kind of lang tagging or locale setting in anaconda. > it affects Chinese fonts in F8 and F9 anaconda too.
I meant Chinese installs (they get Japanese glyphs appearing when they exist).
Sometimes, font switching in pango is irresponsiblećespecially in alias fonts (serif, sans-serif, monospace). Firefox used to have the similar problem and has been fixed. http://bugzilla.mozilla.gr.jp/show_bug.cgi?id=5180 (Mozilla-gumi bugzilla; in Japanese) https://bugzilla.mozilla.org/show_bug.cgi?id=339513 (bmo) I changed out a lot of font package names for F11. Is this any better now? |
Created attachment 321105 [details] Screenshot Description of problem: When I select Japanese on the language selection with today's rawhide tree, it looks like anaconda is rendering some glyphs with Chinese font. Version-Release number of selected component (if applicable): anaconda-11.4.1.50 How reproducible: always Steps to Reproduce: 1.Select Japanese on the language selection 2.See a notice for beta after pressing Next button twice. 3. Actual results: some glyphs looks like from Chinese font. Expected results: should be from Japanese font. Additional info: This is a regression. it worked before.