# How can I modify a fontinst encoding (e.g. t1.etx)?

I am modifying the adobecaslon package to support the Adobe Caslon Expert fonts, basing my work on an old package by Ulrik Vieth for which only some sources are available; I'm also trying to make the package more modular, building on fontinst 1.9.

Adobe Caslon's Expert fonts include st and ct ligatures. To use these, Vieth produced a modified t1.etx encoding file. I would prefer merely to use it an override it. I tried thusly:

``````\relax

\encoding

%% First load the base encoding
\inputetx{t1}

%% Then override some slots
\nextslot{99}
\setslot{\lc{C}{c}}
\ligature{LIG}{\lc{T}{t}}{ct}
\comment{The letter `{c}'.}
\endsetslot

\nextslot{115}
\setslot{\lc{S}{s}}
\ligature{LIG}{\lc{T}{t}}{st}
\comment{The letter `{s}'.}
\endsetslot

\nextslot{141}
\setslot{ct}
\comment{The `old style' ligature ct}
\endsetslot

\nextslot{173}
\setslot{st}
\comment{The `old style' ligature st}
\endsetslot

\endencoding
``````

This seems to work nicely, but I get warnings from fontinst like this:

``````This character already appeared in a LIGTABLE LABEL (line 4312).
(LABEL D 99
) (COMMENT c)
``````

Indeed, if I check the .vpl files output, I find duplicate entries. However, the later entry contains all the same information as the earlier one, plus my modifications, and the output seems to work OK.

What I'd like to know is, am I doing this the right way, and can I ignore the warnings?

-
Have a look at `multislot.sty` (part of `fontinst`). –  Robert Aug 1 '13 at 15:49
Thanks. As far as I can tell, none of the possible `multislot` commands do what I want, as I want to have multiple \setslot...\endslot commands take effect (not just the first) and in the example quoted about changing l to lambda, the "buggy" behavior mentioned is exactly what I want to occur, i.e. retain the original kerns and ligatures. So maybe what I'm doing is right? –  Reuben Thomas Aug 1 '13 at 17:24
For example, in the above, I want to add a ligature for c and for s, while retaining all the existing ligatures and kerns. –  Reuben Thomas Aug 1 '13 at 17:25
Hm, I was especially worried about the double entries for slots 141 and 173, and about the "strange results" that according to `multislot.sty` might ensue, but if `vptovf` does the right thing (for you) by selecting the last entry, then maybe you should indeed just ignore the warning. Or else, could you also provide the driver file for further testing? (my `fontinst` skills have somewhat withered ...) –  Robert Aug 2 '13 at 4:07
Ah yes, of course I want to reset those slots, that is, 141 and 173, but without getting the "first entry only" behavior of `\setslot` that `multislot.sty` gives. –  Reuben Thomas Aug 2 '13 at 9:38