spdiags function algorithm in matlab

Hi everybody, in MATLAB there is a spdiags function, creates sparse matrix by taking diagonal matrices, for example A = spdiags(B,d,m,n) creates an m-by-n sparse matrix by taking the columns of B and placing them along the diagonals specified by d. I want to know what algorithm is behind it, if I want to transfer my MATLAB code to c++, Is there any C++ codes has this function, or is any C++ library or class to do the same thing as this function do?

Answer 1

[John D'Errico]

What is the algorithm? You know the various diagonal elements, and thus the set of indexes for those diagonals and various super/sub diagonals. Stuff the values into those entries.

In MATLAB, this would involve a final call to sparse. If you are writing the code in C, you know (in theory, if you understand sparse matrices) what order to stuff the entries into the array to be efficient.

If you want to know the EXACT algorithm specifics of what spdiags does, then just do this:

type spdiags

and read the code.

[Cedric]

And for the way MATLAB handles/implements sparse matrices, the tech note is here.