Hi Zhiwen,
Try pickling the mesh (see below), but performance won't be any better
than recreating the mesh from scratch.
In [1]: from fipy import *
In [2]: m = Grid2D(nx=4, ny=4)
In [3]: dump.write(m, 'm.gz')
In [4]: m0 = dump.read('m.gz')
In [5]: m0.getCellCenters()
Out[5]:
array([[ 0.5, 1.5, 2.5, 3.5, 0.5, 1.5, 2.5, 3.5, 0.5, 1.5, 2.5,
3.5, 0.5, 1.5, 2.5, 3.5],
[ 0.5, 0.5, 0.5, 0.5, 1.5, 1.5, 1.5, 1.5, 2.5, 2.5, 2.5,
2.5, 3.5, 3.5, 3.5, 3.5]])
If performance is an issue then __getstate__ and __setstate__ (see
<https://www.matforge.org/fipy/browser/trunk/fipy/meshes/numMesh/mesh.py#L664>)
would need to be rewritten so that less is done when the object is
unpickled.
I don't think we have ever used GmshExport for anything useful so I
wouldn't trust it, especially if it is not part of the test suite. I
can't think why it is even in there.
On Fri, Apr 4, 2008 at 3:46 PM, Zhiwen Liang <[EMAIL PROTECTED]> wrote:
> Dear Dan and Jon,
>
> I have a problem regarding the GmshExport module in fipy.
>
> What I am trying to do is to save the mesh every time I create it so that I
> can import it the next time I need it without having to create it from
> scratch. So what I did is to export the mesh using gmshExport as the way of
> saving the mesh. Then next time I need the mesh, I just have to use
> gmshImport to import it.
>
> I noticed that there is a test inside the script of gmshExport.py which
> export a Grid3D mesh. I run it and could not go through. I also attached a
> script that I tried to export the unstructured mesh created by GMSH. It
> stops with different errors.
>
> Or maybe you can suggest some other ways of saving the mesh.
>
> We thank you in advance for any advice you are giving.
>
> Regards,
> Zhiwen
>
>
--
Daniel Wheeler
