[9fans] eqn and unicode -- Again

[9fans] eqn and unicode -- Again

[Gregory Pavelcak]

I should probably just drop it because I reveal my ignorance
when I post, and erik is the only one who has shown any interest,
but ...

I guess what I want is for eqn to sort of immediately convert
GAMMA to Γ and then go about its business with Γ, as if it
never saw the word GAMMA, got it?

So, I was looking at eqn -d to get some more feedback
and it seems that a user definition does that. Here's some
output.

cpu% eqn -d
...	eqn version Oct 24, 1991
.lf 1 -
.EQ
.EQ
define Gamma %Γ%
.	lex token = |define|
.	name Gamma defined as Γ
Gamma
.	lex token = |Gamma|
.	found Gamma|=Γ|
.	lex token = |Γ|
.	57346text: S11 <- \f2Γ\fP; b=0,h=1,lf=2,rf=2,ps=10
.ds 11 "\f2Γ\fP


cpu% eqn -d
...	eqn version Oct 24, 1991
.lf 1 -
.EQ
.EQ
Γ
.	lex token = |Γ|
.	57346text: S11 <- \f2Γ\fP; b=0,h=1,lf=2,rf=2,ps=10
.ds 11 "\f2Γ\fP

Of course, I don't know if the treatment is the same in other
contexts, but I'm wondering if a simple solution is to make
gktbl and treat it like deftbl in lex.c instead of like restbl in
text.c.

Greg