Summary: | FcFreeTypeQueryFace fails on postscripts fonts loaded from memory | ||
---|---|---|---|
Product: | fontconfig | Reporter: | nfxjfg |
Component: | library | Assignee: | Akira TAGOH <akira> |
Status: | RESOLVED FIXED | QA Contact: | Behdad Esfahbod <freedesktop> |
Severity: | major | ||
Priority: | medium | CC: | akira, fontconfig-bugs, freedesktop, ismail |
Version: | unspecified | ||
Hardware: | Other | ||
OS: | All | ||
Whiteboard: | |||
i915 platform: | i915 features: |
Description
nfxjfg
2013-04-25 14:21:24 UTC
Ok, I think best way to handle this is to use the FT_Stream API to get the font file data. (In reply to comment #1) > Ok, I think best way to handle this is to use the FT_Stream API to get the > font file data. Why not revert the hash feature entirely? I can't even think of a valid use of this feature, and it's obscure at best. Akira, do you think you can take a look at this quickly? If not, I'll try to make it at least not return NULL for now. Sorry, I was on long vacation and will take a look tomorrow. Sorry for late action. please try http://cgit.freedesktop.org/~tagoh/fontconfig/commit/?h=bz63922 This my repo contains a workaround to skip adding a hash value into the cache and take care of the matching in that case. that wouldn't affects the ordering of fonts though, please let me know if something still goes wrong for you. Unfortunately I can not test, because I took this opportunity to convert the only postscript font file I used to open type. me neither. anyway, after patching out, FcFreeTypeQueryFace() works fine with even a Type42 font, I guess it should works for Type1 too. I've merged the change to master. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.