sieve-dev FIAT import question

[Matthew Knepley]

On Tue, Mar 25, 2008 at 7:22 AM, Jed Brown <jed at 59a2.org> wrote:
> When I try to build the sieve examples, I get errors like the one below.  If I
>  move the `import FIAT.shapes' up to the top of the file, it works just fine.  I
>  don't know if there is something peculiar about my configuration, but this
>  problem hasn't gone away in the last weeks so I thought I'd mention it.  My
>  configuration included the options:

Do you have numpy in a weird place? You are the first to report this
error, and I
am not sure sure how moving an import would matter, but evidently on your
system it does somehow. In order to debug it, I need to know how you installed
numpy.

>  --with-fiat-dir=/home/jed/usr/lib/python2.5/site-packages/FIAT --with-fiat=1
>
>  but the error persists with --download-fiat=yes instead.
>
>  Traceback (most recent call last):
>   File "bratu_quadrature.fiat", line 60, in <module>
>     FEMQuadrature().run()
>   File "bratu_quadrature.fiat", line 46, in run
>     import FIAT.shapes
>   File "/home/jed/usr/petsc-dev/config/BuildSystem/importer.py", line 37, in
>   import_module
>     mod     = self.load_tail(q, tail)
>   File "/home/jed/usr/petsc-dev/config/BuildSystem/importer.py", line 88, in
>   load_tail
>     m = self.import_it(head, mname, m)
>   File "/home/jed/usr/petsc-dev/config/BuildSystem/importer.py", line 130, in
>   import_it
>     mod = self.loader.load_module(fqname, stuff[0])
>   File "/usr/lib/python2.5/ihooks.py", line 326, in load_module
>     exec code in m.__dict__
>   File "/home/jed/usr/lib/python2.5/site-packages/FIAT/shapes.py", line 16, in
>   <module>
>     import numpy, numpy.linalg, exceptions
>   File "/home/jed/usr/petsc-dev/config/BuildSystem/importer.py", line 36, in
>   import_module
>     q, tail = self.find_head_package(parent, name)
>   File "/home/jed/usr/petsc-dev/config/BuildSystem/importer.py", line 79, in
>   find_head_package
>     raise ImportError, "No module named " + qname
>  ImportError: No module named numpy
>  make: *** [bratu_quadrature.h] Error 1
>
>
>  On another note, the bratu example will work much better with the following one
>  character fix.
>
>  diff -r 6cfe5326613c src/dm/mesh/examples/tutorials/bratu.cxx
>  --- a/src/dm/mesh/examples/tutorials/bratu.cxx  Tue Mar 25 06:13:18 2008 -0500
>  +++ b/src/dm/mesh/examples/tutorials/bratu.cxx  Tue Mar 25 13:17:47 2008 +0100
>  @@ -1143,7 +1143,7 @@ PetscErrorCode Jac_Structured_3d_FD(DALo
>            v[1] = -hx*hz/hyb;                                       col[1].k = k;     col[1].j = j - 1; col[1].i = i;
>            v[2] = -hy*hz/hxb;                                       col[2].k = k;     col[2].j = j;     col[2].i = i - 1;
>            v[3] = (hy*hz/hxa + hy*hz/hxb + hx*hz/hya + hx*hz/hyb + hx*hy/hza + hx*hy/hzb); col[3].k = row.k; col[3].j = row.j; col[3].i = row.i;
>  -          v[4] = -hx*hy/hxa;                                       col[4].k = k + 1; col[4].j = j;     col[4].i = i;
>  +          v[4] = -hx*hy/hza;                                       col[4].k = k + 1; col[4].j = j;     col[4].i = i;
>            v[5] = -hx*hz/hya;                                       col[5].k = k;     col[5].j = j + 1; col[5].i = i;
>            v[6] = -hy*hz/hxa;                                       col[6].k = k;     col[6].j = j;     col[6].i = i + 1;
>            v[3] -= lambda*hx*hy*hz*PetscExpScalar(x[k][j][i]);
>

Fixed. I don't run the structured stuff as much as I should.

  Thanks,

     Matt

>  Jed
>



-- 
What most experimenters take for granted before they begin their
experiments is infinitely more interesting than any results to which
their experiments lead.
-- Norbert Wiener