[XeTeX] OpenType fonts in Linux?

Mike Maxwell maxwell at umiacs.umd.edu
Mon Nov 30 04:44:57 CET 2009


Will Robertson wrote:
> I imagine a useful way to present the information would be to list 
> "languages covered"

Ideally, yes.  For most languages this should be relatively easy.  But I 
seem to recall that there are cases where it's not clear what characters 
are "correct."  Like that English language--do you need the 
s-that-looks-like-f, so you can typeset the Declaration of Independence? 
  Lots of languages have these older characters that are seldom used 
today, but which someone will insist ought to be included; I think that 
comes up a lot with Indic languages (but don't quote me).  And then 
there are recently written languages, where the orthography is in flux. 
  Many South and Meso American languages recently went through a process 
of substituting 'k' for 'c/qu' (the latter had been imported from 
Spanish).  And while I know nothing about CJK languages, I suspect there 
is a wide variety of opinions about which Chinese characters are used in 
Japanese and Korean.

Much simpler would be a tabular listing of Unicode code points, probably 
arranged in blocks for simplicity (most fonts don't support both 
Devanagari and Chinese, for example, so no sense listing fonts that 
don't pertain to a particular block).  Ideally, rather than having an X 
in a table cell, you'd have the actual character as rendered by the 
font.  (Obvious problems here for contextually shaped characters and 
diacritics, though, so maybe the X is best...)

I think Unibook supports using a particular font for a given Unicode 
block, which at least allows you to see what one font supports.
-- 
    Mike Maxwell
    What good is a universe without somebody around to look at it?
    --Robert Dicke, Princeton physicist


More information about the XeTeX mailing list