questions & comments

[Hilmar Schlegel]

Thierry Bouche wrote:
> 
> it's nonsense to fake ligatures that are never directly accessed. What
> makes sense is to make active the ligature pointing to ffi ,
> e.g. _only_ when the ffi glyph does exist.

Indeed! And printing black boxes won't help a lot either...

> This design decision in fontinst (well, in latin.mtx and in the ETX
> files) has always puzzled me.

There is the simple solution to check in the definition of the ligatures
if the ligatures really exist (in the encoding, not the avialable
characters). This works fine and makes a complex font-dependent job very
managable via fontinst.
E.g.
\setslot{\lc{C}{c}}
  \if \tty y \else
    \ifisint{\lclig{CT1}{ct1}}\then
      \ligature{LIG}{\lc{T}{t}}{\lclig{CT1}{ct1}}
    \else
    \ifisint{\lclig{CT}{ct}}\then
      \ligature{LIG}{\lc{T}{t}}{\lclig{CT}{ct}} \fi
    \fi
    \ifisint{\lclig{CH}{ch}}\then
      \ligature{LIG}{\lc{H}{h}}{\lclig{CH}{ch}} \fi
    \ifisint{\lclig{CK}{ck}}\then
      \ligature{LIG}{\lc{K}{k}}{\lclig{CK}{ck}} \fi
  \fi
  \comment{The letter `{c}'.}
\endsetslot

Hilmar Schlegel

-- 
---------------------------------------------------------------
mailto:hshlgaii@mailszrz.zrz.TU-Berlin.DE?Subject=Mail response
---------------------------------------------------------------