[ublas] Consistent implementation of nnz() for sparse_tag trait

From: Russ (c.r.coggrave_at_[hidden])
Date: 2006-04-19 15:39:07

I have noticed that only some of the vector and matrix classes using the 
sparse_tag container_type trait support the nnz() method. I am currently trying 
to write a specialisation for vectors and matrices with the sparse_tag, but this 
inconsitency prevents me from calling the nnz() method, and instead I must count 
the non-zero elements by iterating over the container, which is hurting 
performance.

Is there another trait that can be used to check for support of nnz(), or 
alternatively, would it be possible to add nnz() to all vector and matrix 
classes that use the sparse_tag container_type trait?

An example would be ublas::unit_vector<> class, which has the sparse_tag 
container_type trait, but does not support nnz(). However, implementing nnz() 
would of course be very simple.

Russ