dolfin team mailing list archive

[dolfin@xxxxxxxxxx (DOLFIN)]

One or more new changesets pushed to the primary dolfin repository.
A short summary of the last three changesets is included below.

changeset:   7004:d2f0c7cdc7f7d643ea802d4014abef19f239c462
tag:         tip
user:        "Johan Hake <hake@xxxxxxxxx>"
date:        Thu Sep 10 13:26:40 2009 +0200
files:       dolfin/swig/typemaps.i test/unit/la/python/test.py
description:
Fix for int and dolfin::uint typemaps in PyDOLFIN


changeset:   7003:ccd7d263a0516ec6712136b2365ef64c971cf030
user:        "Johan Hake <hake@xxxxxxxxx>"
date:        Thu Sep 10 10:53:25 2009 +0200
files:       dolfin/swig/numpy_typemaps.i
description:
Fix bug in **double typemap. The old one only worked for mxm blocks.


changeset:   7002:9efdcdff25ee1f8b777919e44ac21e74168143d7
user:        "Johan Hake <hake@xxxxxxxxx>"
date:        Thu Sep 10 10:52:48 2009 +0200
files:       dolfin/swig/la_post.i dolfin/swig/typemaps.i
description:
Fix trouble with numpy.intfoo when indexing.
  - This is a bug introduced in python2.6, where the underlaying PyInt_Check
    was redefined. The typemaps SWIG provides for int uses this. Any numpy
    int type, will fail the PyInt_Check.
  - I have therefore provided our own typemaps for dolfin::uint and now also
    for int
  - v[numpy.long] will still not work...

----------------------------------------------------------------------
For more details, visit http://www.fenics.org/hg/dolfin