Norbert Preining <preining@logic.at> writes: > Hi Keith, > > On Mi, 25 Jul 2012, Keith Packard wrote: >> > That will *never* happen. DEK has finished TeX and there are only >> > very minor bugfixes in exponentially increasing time frames. >> >> I'll complain next time I see him then :-) > > Good luck! And at the same time ask him about the status of his > iTeX > project, I am shaking for anticipation! Heh. He's got so many projects that we've mostly given up begging for some of them to get done :-) >> Given that these directories contain a pile of links to standard fonts, >> I don't want to just add them wholesale. > > I see, ok, let us drop this approach. Yeah, it consumes memory and time to look at fonts twice. > Is there a pro-con with the location of the real files versus the > links? Fontconfig doesn't care where the fonts are; filenames aren't part of the matching computation. > Do you expect something similar wiht fc-cache? Presumably, but fc-cache only runs as fonts are installed (or fontconfig is updated); the database is built incrementally using directory timestamps, so unchanged directories are 'free'. > But *WHY* then is there a trigger interest in /usr/share/temxf/fonts? > It triggers an fc-cache run without any reason and sense, isn't it? Yup. The trigger was added by Josselin many years ago; it doesn't look like it is correct though. > For the other question: I assume, there are lmodern, tex-gyre, ... > that might be of interest for the rest of the world outside TeX. Essentially anything that uses a standard Unicode encoding should probably be available through fontconfig. > Actually this is one of the most common requests I get: make the fonts > available for fontconfig. The problem is that I don't have the > capacity nor knowledge to wade through all fonts and check whether > they make sense to be findable by fontconfig ... Just making sure their encoding is reasonably sane is all that's really necessary; I don't want people ending up with text which is unreadable because some TeX font was chosen by accident somewhere... -- keith.packard@intel.com
Attachment:
pgpUeNy4wTu8l.pgp
Description: PGP signature