CSparse is a package of direct methods for sparse linear systems written
to complement the book "Direct Methods for Sparse Linear Systems", by
Timothy A. Davis. The algorithms in CSparse have been chosen with
five goals in mind:

(1) they must embody much of the theory behind sparse matrix algorithms,
(2) they must be either asymptotically optimal in their run time and memory
usage or be fast in practice,
(3) they must be concise so as to be easily understood and short enough to
print in the book,
(4) they must cover a wide spectrum of matrix operations, and
(5) they must be accurate and robust.

The focus is on direct methods; iterative methods and solvers for
eigenvalue problems are beyond the scope of this package.

CXSparse is a version of CSparse that operates on both real and complex
matrices, using either int or UF_long integers.