[tex-live] pool_size in texmf.cnf

[Heiko Oberdiek]

On Sun, 26 May 2002, Thomas Esser wrote:

> > mattered. I had assumed that "pool_size.progname" always won
> > over pool_size
>
> It is the general kpathsea rule: first match wins. Same for search
> paths.

Now I am puzzled:

$ grep pool_size /media/texlive7/texmf/web2c/texmf.cnf
pool_size =                     125000
% 25000 less than pool_size, but doesn't need to be nearly that large.
pool_size.hugetex =                     1250000
pool_size =                             1250000
pool_size.mpost =                       500000
pool_size.context        =              1250000
pool_size.metafun        =              1000000
pool_size.cslatex =                     1250000
pool_size.hugelatex =                   1250000
pool_size.latex =                       1250000
pool_size.jadetex =                     1250000
pool_size.pdfjadetex =                  1250000
pool_size.xmltex =                      1250000
pool_size.pdfxmltex =                   1250000
pool_size.pdflatex =                    1250000
pool_size.pdfelatex =                   1250000
pool_size.pdfetex        =              1250000

$ /media/texlive7/bin/i386-linux/kpsewhich -expand-var \$pool_size
125000

$ /media/texlive7/bin/i386-linux/kpsewhich -progname=pdfetex \
  -expand-var \$pool_size
1250000

$ /media/texlive7/bin/i386-linux/kpsewhich -progname=mpost \
  -expand-var \$pool_size
500000

So Sebastian is right?

Viele Gruesse
  Heiko <oberdiek@uni-freiburg.de>