Comments wanted: Directory structure of fontinst/inputs/

[Lars Hellström]

At 23.38 +0200 2000-09-03, Hilmar Schlegel wrote:
[snip]
>There is no such character like Tcedilla! As was kindly corrected in AGL
>1.2 we have only to deal with Tcommaaccent/Scommaaccent which went after
>a few corrective moves into their final Unicode positions (hopefully
>;-). It is just like that amacron is not atilde.
>Even given the case a font provides Tcedilla and not Tcommaaccent, then
>the character is usually wrong. This applies usually to Monotype/M$
>fonts. One the other hand side Linotype fonts are OK. Also the new
>Adobe-Pro aka OpenType fonts are OK. It is much better to construct a T
>with a comma below for the purpose.
>For the same reason there is no g, k, l, n, r cedilla.
>The real problem case for LaTex is Scedilla vs. Scommaaccent since they
>are *both* necessary and to be used for their respective languages.
>LaTex defines for T1 encoding on one hand side the wrong Tcedilla slot
>which must be filled by Tcommaaccent instead and on the other only a
>single slot for Scedilla but not for Scommaaccent (without this you
>could drop also the Tcommaaccent). There was a fix for Unicode and AGL
>meanwhile but there is no real fix for LaTex T1 encoding!

Fixing all of that is one of the major TODOs for latin.mtx (and
corresponding ETXs). I suppose a commaaccent can be faked by shrinking the
comma, and the reverse comma accent (for gcomma) can perhaps be faked by
shrinking quoteleft. Changing things in the ETXs may probably need some
care (and coordination with e.g. the LaTeX project), but latin.mtx should
at least produce all possible latin glyphs.

I doubt I will do those fixes, though. Any volunteers?

Lars Hellström