Re: Explanation of Octave operator overloading

[David Bateman]

Paul Thomas wrote:
> David,
>
> > Note that I intend to delete this code in octave-forge now that the 
> > code to treat positive definite matrices is in octave itself, so 
> > rescuing it and using it as an example in da coda or del segna is 
> > probably a good idea. In fact if you are really motivated, rescuing 
> > both da coda and del segna, and this code and writing a chapter on 
> > oct-files for the octave 3.0 manual might be even better :-)
> Whilst I have gone silent on the maintainers list, I am still reading 
> some of the postings.  I thought that the reaction to del segna was so 
> underwhelming that I was put off from any further work on it.  More 
> importantly, there was an opinion that the API should be improved in 
> such a way as to make oct-files an irrelevance.  This was sufficiently 
> persuasive as to stop me in my tracks!  If there is any call for it, I 
> am prepared to help prepare a manual entry, in any way that I can.
I don't think oct-files will ever be an irrelevancy, as there will 
always be certain codes that must be done in oct-files either for speed 
or because they link to external libraries. Also where will people who 
are whiling to improve the internals of octave come from if they don't 
start with oct-files. Doesn't mean that the API won't change in the 
future to be more consistent or "better". In any case the API in 3.0 is 
suppose to be stable enough to share oct-files between versions of the 
3.0 series, therefore even if its not a perfect API it still needs 
documentation :-).

With Da Coda, Del Segna and the type example from octave-forge there is 
largely enough to start a good manual section on oct-files. What is 
perhaps missing is an example of wrapping an external library for use in 
octave. You might also steal the oct-file sub-section of the sparse 
section of the manual, as it makes more sense for this to be with a 
section on oct-files rather than a section on sparse matrices.

Finally what is really missing is an idea of the structure of the manual 
for 3.0. Will in be a collection of the help sections as present, or 
will it be additional information to the help pages, with the help pages 
presented as a function reference? What will be the consistent way to 
cut up the documentation etc. In any case, getting the information into 
texinfo in one form or another and massaging it to be consist later will 
probably be the only way to attack this.

Note that probably the largest task the remains before 3.0 is released 
is improving the manual, and so I imagine John will be looking for help 
with other sections of the manual as well.

Regards
David

PS. Did you ever run "make check" under OS X on a 2.9.5 release, as I 
remember about a year ago you reported a seg-fault that was never 
specifically addressed in the sparse matrix code?