Fontinst modifications

Ulrik Vieth vieth@thphy.uni-duesseldorf.de
Tue, 9 Jun 1998 18:45:16 +0200


Thierry wrote:

> i think though that hacks should stay at the mtx level. ETX define
> the encoding & you should conform to it somehow.

> i don't see why you wouldn't like something like 

> %% j8r to 7d&8r (!) 
> \setglpyh{zerooldstyle}\glpyh{zero}{1000}\endsetglpyh
> \setleftrightkerning{zerooldstyle}{zero}

> BTW i believe that null kerns are ignored by fontinst. (because
> there is something like \if\kerning{a}{b}>\minimumkern\then... )

Could be that this is the case for \setlefrightkerning, but AFAIK
it is not for low-level \setkern commands.  

Anyway, I can see that introducing glpyh name reencoding hacks at 
the ETX level is a rather controversial issue, since it is sort 
of an abuse of the original concept of ETX files to describe an 
encoding and nothing else.  Still it would seem useful to me to 
be able to manipulate the way a source encoding is seen not only
when reading fonts \frompl{<font>} but also in \fromafm{<font>}.  
Hacking the AFM files, as you suggested on another occasion, 
certainly doesn't seem to be a proper solution either.

Cheers, Ulrik.