Re: Octave bindings for GTK+

[muppet <scott asofyet org>]

On Jul 18, 2004, at 7:56 AM, Muthiah Annamalai wrote:

> Muppet wrote
> > You may be able to work around this, if you make some evil 
> > assumptions.  Assume that you have an int large enough to hold a 
> > pointer on all platforms.  Then you could turn the widget pointers 
> > into integers that you'll use as "handles" in octave.
> That was really cool! Abusing an integer ;-)

well, it's neither a great idea nor an original one; both win32 and x11 
use "plain integers" as window handles, and this does not make for a 
very safe interface.  but if you're stuck with it, you're stuck with 
it.

i think matlab also uses integer window ids, but i haven't messed with 
it in over a year (since i turned all my matlab code to octave code).


> > From what you said, I understood that
>
> 1: I must manually write 'C++' wrappers for interfacing Octave with 
> GLib & GOBject.

yes, using octave's extension apis (whatever those might be).

> 2: Then I could use SWIG at some point of time to ceate the binding 
> code for all the GTK+ widgets & GDK.

looking at swig's webpage (http://www.swig.org) i see no mention of 
octave as a target language for its code generator.  you will probably 
have to roll your own solution.

> 3: I could then write Octave wrappers to access the internal 
> dynamically loaded C++ functions of Octave.

all depends how you want to do it.  a common practice is to do simple 
low-level bindings of the C lib to the target language, and then use 
target language-level code to wrap it up in a more idiomatic fashion.  
in some cases, the binding tools can go straight there (e.g. perl's xs) 
and skip the middle man.

i know nothing about this facet of octave, so i leave that as an 
exercise for the reader.  :-)

--
Eight, eight, eight, eight ounces to gooooo, you're gonna be sedated!
  -- Elysse, singing while feeding rice cereal formula to
    fussy infant Yvonne