Re: Preaching DSSSL (Re: using LOUT)

[Valeriy E. Ushakov]

On Wed, Sep 03, 1997 at 01:22:16PM +0100, Sebastian Rahtz wrote:

> what about the worlds most common environments, from the Satan
> of Seattle?

As far as I know the *only* system dependent feature of Lout is
filtered parameters (calling external program to preprocess parameter
value).  The *only* two places where this feature is currently used is
@Verbatim and @CP (printing C programs).  The *only* reason why this
feature is necessary is to save people from manual work of escaping
every special character in @Verbatim and from manually running c2lout
preprocessor in @CP.  For machine generated Lout you don't need this
as all preprocessing can be done in advance.

Everything else is system independent 100% pure ANSI C.  All the file
names are kept withing 8+3 limit.  Just click on the different model
(or how does they call it) before compiling and you will have
Windows-59 Lout or Windows-Not-There Lout or whatever.  It just
simpler to distribute MS-DOS version because it will run under all
these beasts, isn't it.


> PDF deals with the final deliverable, where Unicode doesnt matter;
> true, PDF only deals with 8 bit fonts, but thats just a question of
> mapping from the Unicode internal coding to the right character in an
> output font.

Where can I learn more about font selection schemes deployed in
Unicode TeX.


> if Lout is to be a big-time DSSSL back end, it needs to get into
> Unicode internal processing, or things just go mad.

Yes.  Good or bad but Unicode is inevitable.  I wrote Tibetan
preprocessor for Lout that maps transliteration into font glyph codes
for some particular font.  I'd like to emit Unicode and leave the job
of mapping Unicode to font glyphs up to formatter.

Where can I read more about Unicode TeX.  In particular how it
performs the mapping from codes to glyphs.  Most Tibetan fonts that I
know will need a sequence of codes to be mapped to a single glyph
(vertical consonant cluster).  Most of them also have sometimes up to
a dozen of slightly different shapes for a vowel `u' whith diffrent
shapes used after different consonant clusters for better look.  How
this kind of dependencies are specified with Unicode TeX?

SY, Uwe
-- 
address@hidden                         |       Zu Grunde kommen
http://www.ptc.spbu.ru/~uwe/            |       Ist zu Grunde gehen