Behdad Esfahbod wrote: > On 05/10/2010 07:13 PM, Alan Coopersmith wrote: >> >From digging around, it seems like there's a glyph with an advance width >> of 0 in luximr.ttf that fontconfig doesn't expect in this font which is >> confusing it, and making it classify it as not monospaced. >> >> Is this a bug in fontconfig's monospace recognition or a bug in the font? > > I'd say both. Patch welcome. Strangely, when trying to reproduce with a freshly updated git clone, it's now the non-bold variants with spacing=100, and the bold ones failed to match. By inserting a printf into FcFreeTypeCharSetAndSpacingForSize and running fc-cache from git master, it's only finding characters with an advance of 1229 in luximr.ttf & luximri.ttf, but in luximb.ttf & luximbi.ttf it finds 4 each with an advance of 1338, and one with advance of 682: Glyph: 0x 175 Advance: 682 Glyph: 0x c4 Advance: 1338 Glyph: 0x c5 Advance: 1338 Glyph: 0x c2 Advance: 1338 Glyph: 0x c6 Advance: 1338 I'm not sure why these glyphs are different width, or what criteria fontconfig should be using to decide they're not to be counted towards the spacing calculation, so I can't suggest a fontconfig patch right now. I can easily add a fonts.conf file to the X.Org fonts/bh-ttf package to override it there though. -- -Alan Coopersmith- alan.coopersmith@xxxxxxxxxx Oracle Solaris Platform Engineering: X Window System _______________________________________________ Fontconfig mailing list Fontconfig@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/fontconfig
