On 02-01-12 00:44:55 MET, Markus Kuhn wrote:
> Robert Joop wrote on 2002-01-11 21:07 UTC:
> > - what (additional) features would you find nice to have?
> 
> Nice would be a best-effort attempt to output as much of Unicode as
> feasible with just the standard Postscript fonts available, as
> demonstrated for instance in
> 
>   http://mail.nl.linux.org/linux-utf8/2001-12/msg00012.html
> 
> That covers already most of MES-1 and is a significant improvement over
> the Latin parts of ISO 8859.

covering MES-1 shouldn't be a problem, psfilt already covered much of it
and more, for example some of the troff/ditroff characters, plus some
more like the telephone answering machine symbol (useful for address
lists), mars/male, venus/female, married and divorced genealogic
symbols.
psfilt for example always printed "->", the pointer operator in
C/C++/perl, using the rightwards arrow (minus greater looks ugly in
Times Roman...).
it simply used everything in standard fonts (including the unencoded
characters) and the symbol and dingbats fonts (532 characters then)
plus the mentioned symbols in its own font.
and i definately will have a look at your overpriting hacks to emulate
characters not existing in PS fonts.

instead of psfilt's monolitic design (recompile to add a new pretty
printer? yuck!) i'll implement something more modular.
perhaps pipelining recode(1) to get the input into UTF-8 form,
some pretty printing filters that will emit XML and
a backend that generates PS from the XML. (the latter will use iconv(3)
for the UTF-8 to UCS2 step.)
i've found unicode to postscript character name mapping files at adobe's
site for 1067 characters (not counting those of their characters they
put into the private plane).
so, even MES-2 shouldn't be out of reach, given the right PS fonts...

i've never printed russian before, but i just did so a few moments ago,
using psfilt.

rj
--
Linux-UTF8:   i18n of Linux on all levels
Archive:      http://mail.nl.linux.org/linux-utf8/

Reply via email to