(find-maximanode ...): elisp hyperlinks to the Maxima manual

My main pages about Maxima are this html page, that is in a human-friendly format, and maxima.e, that is made to be used from Emacs, typically like this.

Some of the e-script blocks in maxima.e contain elisp hyperlinks like these ones,

# (find-maximanode "derivabbrev")
# (find-maximanode "depends")
# (find-maximanode "express")
# (find-maximanode "gradef")
# (find-maximanode "gradefs")

that were trivial to implement in Emacs, but were very tricky to htmlize correctly. For the Emacs side I just had to do this:

(code-c-d "maxima" "/usr/share/maxima/5.44.0/" "maxima")
;; Now this:             (find-maximanode "gradef")
;; is a shorthand for: (find-node "(maxima)gradef")
;;                          (info "(maxima)gradef")

The problem is that the Maxima manual uses "@anchor{...}"s - look at Differentiation.texi, for example - and so the string after the `find-maximanode' can be the name of an anchor instead of the name of a node, and we need a way to translate each name of an anchor, like "gradef", to its target...

When we compile Maxima like this (the original is here),

rm -Rf ~/bigsrc/maxima-git-5.47.0/
mkdir  ~/bigsrc/maxima-git-5.47.0/
cd     ~/bigsrc/maxima-git-5.47.0/
git clone ~/bigsrc/maxima .

git checkout 5.47.0
git clean -dfx
git reset --hard

./bootstrap         |& tee ob
./configure         |& tee oc
make                |& tee om

cp -v doc/info/maxima-index-html.lisp ~/blogme3/

it generates a file called maxima-index-html.lisp, that is a translation table in Lisp. The definition of `find-maximanode' in my htmlizer is here; it loads maxima-index-html.lua if it hasn't been loaded yet, and this parses the translation table in Lua with Lisp4.lua and converts it to a Lua table. Then `find-maximanode' uses that table to convert the tag to the a URL, and that's it.