Dear Bran Gough, Frankly, I used a quite dirty solution to generate these pdf files. That is, I first inserted `\nopagenumbers' and changed `\endinput' to `\bye' for each *.tex files generated by randplots.gnp. Then I texed each of them, followed by `dvips', `ps2eps', and `epstopdf' instructions.
To be specific, step 1: modify `rand-gumbel.tex' so that it has the similar structures with other files command 1: sed -i -e '$ a\ \\endinput' rand-gumbel.tex step 2: insert `\nopagenumbers' and chang `\endinput' to `\bye' command 2: for i in `ls rand*.tex`; do sed -i -e '1 a\ \\nopagenumbers' -e '$s/\\endinput/\\bye/' $i; done step 3: tex each file command 3: for i in `ls rand*.tex`; do tex $i; done step 4: dvips each file command 4: for i in `ls rand*.dvi`; do dvips $i; done step 5: ps2eps each file, so that the blank margins can be removed command 5: for i in `ls rand*.ps`; do ps2eps $i; done step 6: generate pdf files command 6: for i in `ls rand*.ps`; do ps2eps $i; done Now, we have all the pdf files needed, including random-walk.pdf. I am not expert on TeX and gnuplot, so the method I used may be far from optimal and looks dirty, and more importantly, error-prone. Any improvements are welcome. On Sun, Mar 29, 2009 at 3:47 PM, Brian Gough <[email protected]> wrote: > At Sun, 22 Mar 2009 11:51:16 +0800, > Hongzheng Wang wrote: >> Thank you for your kind explanation. As a temporary solution, I >> manually converted the figures used in randist.texi into pdf format > > How did you do the conversion? We'd need to be able to regenerate the > figures if the plots change, ideally directly from gnuplot. The > gnuplot file is doc/randplots.gnp. > > -- > Brian Gough > -- HZ _______________________________________________ Help-gsl mailing list [email protected] http://lists.gnu.org/mailman/listinfo/help-gsl
