Re: Standard Matrix interfaces

From: helmut.zeisel_at_[hidden]
Date: 2001-08-20 08:58:02

--- In boost_at_y..., Hayati Ayguen <Ayguen_at_b...> wrote:
> helmut.zeisel_at_a... schrieb:
> > 

> 
> 
> > IMHO, what is needed most is a simple
> > interface for rectangular, dense matrices,
> > covering C-style layout.
> > In a first step no operators, no expression templates,
> > no transpose etc.
> 
> 
> especially sparse matrices need access methods/iterators to the 
elements
> of a matrix. 

Yes.

> declaring a too simple interface supporting only
> rectangular, dense matrices is wrong. 

I don't think so.

> such a standard interface 
would
> lead many people to write higher level code (matrix solver ...) for 
such
> an interface. but you'll have to rewrite / throw away such code if 
other
> issues require a new more sophisticated interface solving element 
access
> in a more uniform manner.

Usually you will take a different algorithm/matrix solver
for, say, sparse or band matrices.
So for special matrices, one has to write special code anyway.

In addition, I personally need some OO wrappers for
existing librarires (like IMSL)
more urgently than some new OO-from-scratch libraries

> all current issues have to be taken into account for specifying such 
an
> interface.
>

Again this would mean that we will never have a standard interface.
 
> i agree that we don't have to implement anything like operators,
> expression templates, transpose, ... but we need proper interfaces 
for
> such functions. take transpose() for example:
> 
> a) should it be a global function where
> b) a static member function
> c) a member function modifying itself

Exactly.
From my point of view, these interfaces
are almost equivalent.
Specifying these interfaces is exactly one
important step in the right direction.

Helmut