[PEP draft 2] Adding new math operators

[Robin Becker]

In article <m3snsejzrs.fsf at chinon.cnrs-orleans.fr>, Konrad Hinsen
<hinsen at cnrs-orleans.fr> writes
...
>).
>
>This scheme should work find as long as conversions between matrices
>and arrays occur infrequently. This is certainly the case in all of my
>own applications, which typically construct an array using elementwise
>operations and at some point convert it to a matrix that is used in
>matrix operations.
>
>So the question is: for which real-life cases would this approach become
>too cumbersome to be useful?

I thought that's what you must be implying. The implementation part is
the problem ie can you 'fake' the conversions without too much work.
Can't numpy already present the same data in strange ways ie isn't the
header data separate from the real data so that I can set B=A[1,:]
without creating a new copy of the first row (ie B is a slice).
-- 
Robin Becker