Generic cholesky() fails on the trivial matrix

[orlitzky]

sage: A = matrix(QQ,[])
sage: A.cholesky()
...
IndexError: matrix index out of range

If we simply return A itself in this case, the definition of the decomposition is satisfied.

Component: linear algebra

Author: Michael Orlitzky

Branch/Commit: 64d9028

Reviewer: Markus Wageringel

Issue created by migration from https://trac.sagemath.org/ticket/33107

[orlitzky]

Branch: u/mjo/ticket/33107

[orlitzky]

Author: Michael Orlitzky

[orlitzky]

New commits:

08bbd3c

Trac #33107: cholesky() for trivial matrices.

[orlitzky]

Commit: 08bbd3c

[sagetrac-git]

Changed commit from 08bbd3c to 89c46c6

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1. This was a forced push. New commits:

89c46c6

Trac #33107: cholesky() for trivial matrices.

[mwageringel]

comment:3

sage: matrix(QQ, [[1]]).cholesky().is_immutable()
True

Usually, .cholesky() returns an immutable matrix. Should we do that in the 0×0 case as well, for consistency? For example, the subdivisions could be mutated, otherwise.

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1. New commits:

64d9028

Trac #33107: return immutable trivial Cholesky factors.

[sagetrac-git]

Changed commit from 89c46c6 to 64d9028

[mwageringel]

Reviewer: Markus Wageringel

[mwageringel]

comment:5

Thanks.

[slel]

comment:6

Setting milestone to 9.6 now that 9.5 is out.

[vbraun]

Changed branch from u/mjo/ticket/33107 to 64d9028