Happy new year + proposal to expand set of chars recognized as alphabetic
Subject: Happy new year + proposal to expand set of chars recognized as alphabetic
From: Robert Dodier
Date: Sun, 31 Dec 2006 12:57:45 -0700
Happy new year to everyone!
The Maxima parser recognizes only A-Z and a-z as alphabetic characters
(unless told otherwise by declare(<foo>, alphabetic)).
I propose the alphabetic characters be expanded to include anything
for which ALPHA-CHAR-P yields T.
That will let Maxima recognize symbols which contain non-ascii chars
without having to resort to declare(<foo>, alphabetic).
I believe there is no downside here -- it just makes Maxima more
useful in non-ascii locales.
There may well be non-ascii characters which are not covered by
ALPHA-CHAR-P -- I don't know what a Lisp implementation would
return for syllabaries or ideograms. But covering alphabets at least
would be a good step forward.
Empirically I find SBCL and Clisp yield the expected output (namely T)
for Russian characters (UTF-8). I didn't try any other alphabets.
I don't expect GCL to do the right thing here -- it is not generally
non-ascii aware, from what I know.
Comments?
Robert