Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!gatech!bloom-beacon!eru!hagbard!sunic!mcsun!ukc!icdoc!sot-ecs!Sebastian
From: S.P.Q.Rahtz@ecs.soton.ac.uk (Sebastian Rahtz)
Newsgroups: comp.text.tex
Subject: questions about creating virtual font files
Message-ID: <22927.9010121655@manutius.ecs.soton.ac.uk>
Date: 12 Oct 90 16:55:14 GMT
Sender: spqr@ecs.soton.ac.uk
Lines: 45

Is there anyone out there who actually uses virtual font files,
other than, for instance, the automatic ones created by Rokicki's
afm2tfm? I am trying to write a .vpl file to make up the new suggested
256 char TeX font layout from a PostScript font, but my inexperience
is showing up. some questions:

 1) If we copy an example from Knuth's  article on virtual fonts, the
    following code seems reasonable:

    (CHARACTER O 337 (comment just an S)
     ( MAP
       (SETCHAR O 123)))
    which just sets a 123 if I ask for a 337, taking over the
    dimensions of 123. But this:
 (CHARACTER O 337 (comment Germandbls)
 ( MAP
    (SETCHAR O 123)
    (SETCHAR O 123)))
  which should set a double S when 337 is used, duly does that but the
  SS has no width. yes, maybe this is reasonable, but can someone tell
  me when metrics are taken from the base TFM and when they are set to
  nothing (as apparently in this case)?

2) OK, so I want a Z with a dot over it for position 273. so something
   like this answers the needs:
  (CHARACTER O 273 (comment zdotover)
    (CHARWD R 444)
    (CHARHT R 448)
    (MAP
      (SETCHAR O 172)
      (MOVERIGHT R -222)
      (MOVEUP R 20)
      (SETCHAR O 012)
      )
    )

  BUT, is there an easy way, other than visual testing, to determine
  the amount to step backwards and up? how can I derive it
  automatically from the metric for `z'? am I being either naive about
  typography, or missing something everyone else knows already?

3) can one of you character gurus tell me about the `ij' ligature? do
   the letters actually join up? 

Sebastian Rahtz