Samples
nibua-r
Hi,
I'm using gnuplot since I have started to work with LaTeX for 5 years.
Now, I'm working with ConTeXt for 4 months and the context terminal
responds to a real need !
I have opened this topic since I think that it could be interesting to
work on the context terminal with the "improving samples" approach.
This approach is already used, but now, there is a dedicated place to
discuss about "sample experiments". Feel free to give your opinion !
The very first problem I have will seem obvious:
- I'm working on Linux Debian (testing) and I have compiled the CVS
release of gnuplot,
- once the compilation has succeeded, I have tried the context term ->
seems to work,
- I have downloaded http://pub.mojca.org/gnuplot/sample/* and launch
"texexec --pdf document".
The result is not the "two nice mathematical functions"... The log
gives me:
TeXExec 5.4.3 - ConTeXt / PRAGMA ADE 1997-2005
metapost to tex : document-mpgraph
mpto: document-mpgraph.mp:331: Line is too long
metapost : document-mpgraph
format : metafun
This is MetaPost, Version 0.901 (Web2C 7.5.5)
(/usr/local/texlive/2005/texmf/web2c/natural.tcx)
error in metapost run : document-mpgraph.mp:82
I have found in document-mpgraph.log : Unable to make mpx file
What I'm doing wrong ? It looks like I need a lot of work to obtain a
working sample... ;-)
I understand that sample.tex is the output of gnuplot for sample.plt
and that document.tex contains all required command definitions within
\start/\stopMPinclusions. My ConTeXt/METAFUN knowledge isn't sufficient
to understand what happens...
To Mojca: Is the *.pdf in http://pub.mojca.org/gnuplot/demo/ was
generated with the context term ? If yes, is it possible to have the
source ? (could be usefull once I succeed in compiling the "Two nice
mathematical functions")
Mojca Miklavec
> I'm using gnuplot since I have started to work with LaTeX for 5 years.
> Now, I'm working with ConTeXt for 4 months and the context terminal
> responds to a real need !
I'm sorry. I somehow missed this thread. (Don't know how that was
possible, sorry again.)
> I have opened this topic since I think that it could be interesting to
> work on the context terminal with the "improving samples" approach.
> This approach is already used, but now, there is a dedicated place to
> discuss about "sample experiments". Feel free to give your opinion !
What exactly do you mean?
> The very first problem I have will seem obvious:
> - I'm working on Linux Debian (testing) and I have compiled the CVS
> release of gnuplot,
> - once the compilation has succeeded, I have tried the context term ->
> seems to work,
> - I have downloaded http://pub.mojca.org/gnuplot/sample/* and launch
> "texexec --pdf document".
>
...
> I have found in document-mpgraph.log : Unable to make mpx file
>
> What I'm doing wrong ? It looks like I need a lot of work to obtain a
> working sample... ;-)
I suppose that you already noticed from the last thread that there are
general problems with metafun/textext.
> To Mojca: Is the *.pdf in http://pub.mojca.org/gnuplot/demo/ was
> generated with the context term ? If yes, is it possible to have the
> source ? (could be usefull once I succeed in compiling the "Two nice
> mathematical functions")
Yes, the "source" can be found in CVS repository of guplot under
demos. I can upload huge tex output files (which, btw., made pdftex
run out of memory), but I'm unable to (re)compile the samples myself.
I lost so many nerves compiling the samples that I temporary switched
back to postscript terminal in order to be able to finish my reports
in time.
I hope that the problem will be solved as soon as possible.
Mojca