On Thu, 2 Aug 2012, Paul T. Bauman wrote:

The patch posted
hereĀ http://users.ices.utexas.edu/~pbauman/tensor_tools.patch creates a
TensorTools namespace (within the libMesh namespace) which collects various
operators on VectorValue, VectorType, TensorValue, TensorType, and
TypeNTensor. This includes the IncrementRank and DecrementRank bits as well
as libmesh_dot and libmesh_norm. The majority of the changes in the patch
were appending TensorTools to all occurrences of libmesh_norm. The impetus
is some other handy methods are coming in the Nedelec patch and Roy
suggested we collect all this in a namespace.
OK for trunk?

I hadn't thought about libmesh_dot and libmesh_norm... there's no
point in putting something in TensorTools except to keep the global
(well, the libMesh::) namespace uncluttered, and for backwards
compatibility reasons we might want to keep a libmesh_norm() in the
global namespace.  libmesh_dot() is probably recent enough to move,
but if it's actually getting a serious namespace then we can probably
drop the libmesh_ pseudo-namespace and just call it TensorTools::dot()
(or TensorTools::inner_product?)

Anyone else have any preferences?  I could be talked into
TensorTools::norm_sq() if nobody else is using libmesh_norm() in user
code; that would have the bonus of being less confusing for people who
are unfamiliar with std::norm but familiar with the *correct*
definition of a norm.  Other suggestions for the namespace would also
be welcome; TensorTools was the best we could come up with but the
time to set it in stone is now...
---
Roy
------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Libmesh-devel mailing list
Libmesh-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/libmesh-devel

Reply via email to