encoding hack

[Han-Wen Nienhuys]

hi there,

I just realized that there might be an easy answer out of the
font-encoding troubles. We could use the current setup (setting
encodings with \encoding), and solve the problem on the backend by a
two-step process:

* dump all text string using glyphshow. Instead of doing

  (hi) show

we would dump

  /h glyphshow /i glyphshow 

this solves the hairy problem of doing on-the-fly font recoding, and
seems rather trivial to code.


* Compose several fonts of CMR/EC family into a big font containing
  roman, cyrillic, greek, etc.  This would require some hacking of
  mftrace, but I think it would not be hard.  In essence, we would
  automate the process that lead to the cm-super font set.


Ideas, comments? 

(Of course, this is all 2.5 material)

-- 

 Han-Wen Nienhuys   |   address@hidden   |   http://www.xs4all.nl/~hanwen