Re: Plotting status in Octave

[Jordi Gutiérrez Hermoso]

2008/10/22 Jonathan Stickel <address@hidden>:
> That's just it:  a champion is needed to work on the backend of Octaviz,
> whatever the details of the implementation.  I just don't have the c++
> skills to make much progress myself.

Oh, I'm fine with C++. I've coded in that beast of a language for ten
years. And yes, VTKs documentation isn't stellar, but the book and the
source are more or less enough for me.

The problem I have right now with Octaviz is that I don't understand
the flex and bison parsing that is necessary to compile all of VTK in
an Octave-callable way. I'm sure that's where the biggest bugs lie
right now, at least when I compare those files to the ones that VTK
uses to bind to Python, Tcl, and others. And yes, I'm sure that Dragan
just took the sampling parsing files that VTK ships for other
languages and adapted them to Octave, but I don't understand what
those adaptations consist of. If we are to do what Thomas suggests and
use VTK from Octave in a more limited way, we'd need to fix those
files to only compile into .oct the parts of VTK that we want, as well
as the CMake build process, but this latter one could be a bit easier.

VTK is nice, but it is "just" another frontend to OpenGL. MathGL right
now is being maintained, it already has an Octave interface too, and
it seems to already be able to do all the things that Octaviz can do.
I don't know if what Shai and Michael are working does 3d graphics.

I think I will look into MathGL and see if it can be better integrated
into Octave.

- Jordi G. H.