On Sun, Nov 02, 2008 at 04:13:36PM +0000, Garth N. Wells wrote:
> I spent some time tracking down a nasty bug related to cont_cast in 
> Mesh::init. The flag Mesh::_ordered was being reset despite the mesh 
> being const. Can we avoid this const_cast, or at least move the 
> const_cast up from the internals of Mesh so it is somewhat clearer that 
> const-ness is being cast away?
> 
> Garth

The init() function is the only place where we should have a
const_cast in the Mesh.

It should not reset the _ordered flag. The init() function may
potentially compute new connectivity (which existed before but was not
computed) but the new connectivity should always be ordered. I'm not
sure it is but we should make sure it is.

-- 
Anders


> 
> DOLFIN wrote:
> > One or more new changesets pushed to the primary dolfin repository.
> > A short summary of the last three changesets is included below.
> > 
> > changeset:   5065:292230d5fffec9c4939b08865de84a312db07ec5
> > tag:         tip
> > user:        "Garth N. Wells <[EMAIL PROTECTED]>"
> > date:        Sun Nov 02 12:58:07 2008 +0000
> > files:       dolfin/mesh/TopologyComputation.cpp
> > description:
> > Reorder mesh if it was odered before topology computation.
> > 
> > 
> > changeset:   5064:1d2d1d74992cccd6d275b4c14f99917bc58a635d
> > user:        "Garth N. Wells <[EMAIL PROTECTED]>"
> > date:        Sun Nov 02 12:39:03 2008 +0000
> > files:       demo/function/eval/cpp/main.cpp dolfin/mesh/MeshOrdering.cpp 
> > dolfin/mesh/TopologyComputation.cpp
> > description:
> > Print warning if ordered mesh is invalidated.
> > 
> > Work in getting Function demo to run.
> > 
> > 
> > changeset:   5063:ccf83d7e17b1db1f5c9a0d0e896f317914c23de5
> > user:        "Garth N. Wells <[EMAIL PROTECTED]>"
> > date:        Sun Nov 02 10:39:25 2008 +0000
> > files:       demo/pde/dg/poisson/cpp/main.cpp 
> > dolfin/function/SpecialFunctions.cpp dolfin/function/SpecialFunctions.h
> > description:
> > Update SpecialFunction constructors and get DG Poisson demo working.
> > 
> > ----------------------------------------------------------------------
> > For more details, visit http://www.fenics.org/hg/dolfin
> > _______________________________________________
> > DOLFIN-dev mailing list
> > [email protected]
> > http://www.fenics.org/mailman/listinfo/dolfin-dev
> _______________________________________________
> DOLFIN-dev mailing list
> [email protected]
> http://www.fenics.org/mailman/listinfo/dolfin-dev

Attachment: signature.asc
Description: Digital signature

_______________________________________________
DOLFIN-dev mailing list
[email protected]
http://www.fenics.org/mailman/listinfo/dolfin-dev

Reply via email to