Re: [MayaVi-users] getting mayamat.py and other comments

[Prabhu R. <pr...@ae...>]

Hi Marco,

>>>>> "MH" =3D=3D Marco H=F6fer <ho...@il...> writes:

    MH> Hi Prahbu!  One fast question: Where can I get mayamat.py? I
    MH> was googling around but found nothing.

Well, its not released yet.  I just sent a sneak peek version to the
MayaVi users list.  The sourceforge archives do not save attachments
it seems so there is no way for you to get it off the web. =20

The code is functional but I've been trying to clean up my home page
and put it up there.  I'll try and do that tommorow.  In anycase I'll
mail you a copy off the list.

    MH> And now a few comments:

    MH> First let me say thanks for the wonderful program MayaVi.  At

Thanks for the kind compliments!

    MH> my institute I use python + Numeric and Scipy together with
    MH> some C++ Code for simulating laser resonators. IDLE + dislin
    MH> together with disipyl (for nice 2d plots and sometimes 3d
    MH> plots) + MayaVi (for realy wonderfull data visualisation)
    MH> builts a really wonderfull and powerfull enviroment for doing
    MH> this (If somebody knows something simular in an other (perhabs
    MH> faster) GUI--> Let me know!!)  These three parts fits really
    MH> together (after changing a few lines of Code) . Idle and

Sounds interesting!

    MH> MayaVi are based on TK and in the disipyl package (very nice
    MH> wrapper for dislin) there is the tkdisipyl.Diddle class for
    MH> using dislin interactive in IDLE, so one can really use them
    MH> together in one (IDLE) shell. I tryed many other things but
    MH> found nothing better that gives you an free interactive
    MH> scientific enviroment with "integrated" 2D and 3D plotting
    MH> posibilitis. Perhaps if Scipy is finished (with integreated

You should definitely check Fernando's IPython shell.

  http://www-hep.colorado.edu/~fperez/ipython/

Its a very nice shell for Python.

[snip: interactive data -> vtk files -> MayaVi ]

    MH> If mayamat.surf send the data direct to MayaVi (or better say
    MH> vtk) this would be wonderfull for me.

Well, right now mayamat handles structured points data objects and
sends them directly to VTK and requires no intermediate step for
generating VTK data.  So if you have regularly spaced data sets
mayamat will handle it directly now but nothing more.

My ultimate plan is to extend pyVTK to generate VTK data objects
directly from NumPy arrays rather than generating VTK files.  I think
it should be possible to do except I'm hard pressed for time.  Its in
my TODO list.  The idea is more or less similar.  mayamat does the
necessary magic to convert a Numeric array properly to a vtk array.
One just needs to put these together to generate the corresponding VTK
data objects.  MayaVi already has the code to deal with a VTK data
object passed to it (as against handling a data file).  So then it
should be efficient to script MayaVi with data that you generate.

Thanks again for the comments and compliments!

I'll mail you the mayamat module off-list.

cheers,
prabhu