Hi Wilhelm, sorry for my late reply.
At this point it seems like it would be better to leave the Coma
package outside of the official release, at least for the moment.
I suggest that you post it to the forum for user-contributed packages:
https://sourceforge.net/apps/phpbb/maxima/viewforum.php?f=3
Does anyone have an opinion about whether Coma should be
included in a Maxima release, or continue to be hosted at the forum?
You mentioned documentation in your most recent message.
My advice at this point is just to try to change the documentation
to use a generic LaTeX class file so that it's easy for others to
generate the documentation. Converting it to Texinfo would make
the documentation accessible from the command line (I'm waving
my hands while ignoring the problem of how to wedge the Coma
documentation into the on-line reference manual ...) but it is
more laborious.
I've taken the liberty of forwarding your most recent message
so others have some context for this exchange.
All the best, and thanks for your interest in Maxima.
Robert Dodier
PS.
On 7/4/09, Wilhelm Haager <wilhelm.haager at htlstp.ac.at> wrote:
> Hi Robert!
>
> Sorry for the delay, during the last two months I had not much time to
> spend on Maxima. Your announcement for the next release on August 1 was
> the incentive for me to accomplish the duties for the control engineering
> package "COMA":
>
> 1) License statements:
>
>> Yes, the license statement in coma.mac is good.
>> (I guess you mean GNU GPL instead of "Gnuplot license" above.)
>
> of course! (just a typing mistake)
>
>> Each file in the package, including the documentation, should have
>> a license statement in it. The documentation should have the same
>> license statement (GNU GPL) as the programs.
>
> 2) Testscript
>
>> The test script should be organized so that input expressions
>> and the expected results are interspersed --- e.g.
>>
>> x + x;
>> 2*x;
>>
>> sin(0);
>> 0;
>>
>> etc. Then batch(rtest_coma, test) reads each input, computes a
>> result, and compares it to the expected result (same as run_testsuite).
>>
>> If there are some examples for which there is no useful result
>> (e.g. the plotting examples) those should go into a separate file.
>
> I hope I will succeed; at the first try I got different results in
> floating point numbers and lists (roots of polynomials, where the order
> does not matter). I still do not know how to incorporate graphics into the
> test script.
>
> 3) documentation:
>
>> The translated document is good.
>>
>> I guess the .pdf document was generated from LaTeX.
>> If so, probably it's best to commit the .tex and image files.
>> I don't know whether only the .tex and image files should be
>> committed, or both the .tex and images files and .pdf.
>> Perhaps you can ask on the mailing list for opinions.
>
> The source code of the documentation in its present form is not very
> suitable for publication, for the following reasons:
>
> - It uses a quite "experimental" cls-File (derived from KOMA-Script),
> regardless of any copyrights.
>
> - It uses special, non-standard fonts (Bitstream Charter),
> other fonts would upset the present layout.
>
> - Every Maxima code snipped is a particular pdf-file
> (extracted from a file generated with Acrobat distiller),
> that makes more than 100 pdf's.
>
> But I want to provide the entire documentation in Texinfo; I think that is
> the elegant solution.
> It is no problem to take the Maxima texinfo and html files as a reference,
> and adapt my documentation according to that,
> but I still don't know how to start the texinfo process to get the result
> of the whole documentation (in ps, pdf, html or whatever).
>
> How can that be performed? (texinfo is already installed, within MikTeX)
>
> Best regards
> Wilhelm Haager
>
>
>