[Wikipedia-l] Re: Parsing TeX

Wed Jul 31 21:10:50 UTC 2002

The Cunctator wrote:

 >On 7/31/02 3:20 PM, "Jens Frank" <JeLuF at gmx.de> wrote:
 >
 >
 >>Building an interpreter as mighty as TeX to process formulas is a big
 >>task, and using PHP would take very long to parse one formula.
 >>
 >>I've hacked the phpwiki code to add a new namespace
 >>[[math: ]] so you can write formulas like [[math:a^2+b^2=c^2]] or
 >>[[math:\sum_{n=0}^{\infty}\frac{1}{n}=\infty]].
 >>
 >>The code will make TeX create an image (PNG) of the formula. Those
 >>images will be cached, they will be created only once and will be
 >>shared between articles. ( [[math:E=m c^2]] might be used on many
 >>pages ).
 >>
That's a _really brilliant_ hack -- the idea of using the TeX code as
the name of a page is the obvious right thing!
We should have a garbage-collector sweep up orphan math: pages every so
often.
We should also be really cautious about TeX doing insecure things. Is
there a subset of TeX syntax we could parse and validate before we pass
it to TeX?

 >>
 >>Processing one formula took about one minute on a very old 486/DX2,
 >>why I didn't submit the code to LDC for review or comments, yet.
 >>
 >>
The new server should run that code in about 0.5 second on one of its
two processors, given typical CPU scaling figures.
See http://field.hypermart.net/CPU/cpu.htm

 >>
 >>