On Tue, Sep 1, 2015 at 4:01 PM, Fabian Greffrath <fabian@xxxxxxxxxxxxx> wrote:
If you add any patterns about the glyph coverage, you're right and fontconfig will behaves like that. otherwise there should be no different in score between both fonts as your example did.
Hi Akira,
Am Dienstag, den 01.09.2015, 15:36 +0900 schrieb Akira TAGOH:
> p052023l.pfb and uplri8a.pfb should has same score on evaluating.
p052023l.pfb features a wider glyph coverage, so it should get a higher
score.
score. Also, how come that p052023l.pfb is always sorted before
uplri8a.pfb in the initial FcSortList() step, although the latter
appears first in the unsorted set created by FcFontList()?
Assuming you're talking about the case of fontset2, the result is different because both of the pattern for the first FcFontSort() and the second FcFontSetSort() is different.
the result of FcFontSort() is sorted out by the condition of "outline=true" and the result of FcFontSetSort() is sorted out by the condition of "family=URW Palladio L" on the result of the above. if you want to see more details how fontconfig figures out the result on the sort, run your application with FC_DEBUG=3 or compare the result of fc-match -a :outline=true and fc-match -a "URW Palladio L".
- Fabian
Akira TAGOH
_______________________________________________ Fontconfig mailing list Fontconfig@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/fontconfig
