[luatex] a question about the names \Umathcharnumdef etc

jfbu jfbu at free.fr
Fri Oct 16 10:35:49 CEST 2015

Le 16 oct. 2015 à 09:47, Ulrike Fischer <luatex at nililand.de> a écrit :

> Am Thu, 15 Oct 2015 18:18:58 +0100 schrieb David Carlisle:
>>> or perhaps \Umathcharnumdef was always ok ?
>> yes, it was, the Umath... commands were always available with their
>> basic luatex names and also prefixed luatex.
> That's not true. In a texlive 2012
> \documentclass{article}
> \begin{document}
> \show\Umathcharnumdef
> \show\luatexUmathcharnumdef
> \end{document}
> gives
>> \Umathcharnumdef=undefined.
> l.36 \show\Umathcharnumdef
> ?
>> \luatexUmathcharnumdef=\Umathcharnumdef.
> l.37 \show\luatexUmathcharnumdef

The file 
says that 
>   % etex and pdftex primitives are enabled without prefixing
>   % as well as extented Unicode math primitives (see below)

but as your test file shows this was not the case for \Umathcharnumdef
(and also \Umathcodenum which I tested).

I recall now that mathastext prior to TeXLive 2013 did not use
\luatexUmathcharnumdef because the right side of the assignment would have
been \luatexUmathcodenum`\-\relax, but \luatexUmathcodenum back then
did not always return a Unicode mathcode.

> Are all these \Umath etc primitives using the same syntax in both
> engines? 

mathastext.sty fortunately does not deal with \Udelcode, \Udelcodenum
from David's answer, it uses (aliases to)
\Umathcharnumdef, \Umathcodenum, \Umathcode
and \Umathchardef with the same syntax for both engines. 

I tested on TL2012 and TL2015 before pushing to CTAN yesterday.

Apart from that I am curious to get explanation as to your TL2012 result
> \Umathcharnumdef=undefined.
l.4 \show\Umathcharnumdef
> \luatexUmathcharnumdef=\Umathcharnumdef.
l.5 \show\luatexUmathcharnumdef

which is strange (and the \luatexUmathcharnumdef does work)


More information about the luatex mailing list