27 Jul
2020
27 Jul
'20
9:26 p.m.
On 24.07.20 11:19, Vladimir V. Kisil wrote:
I have attempted to add some more tests for sqrfree_parfrac() and it turned out that GiNaC crashes on x*(x^2-1)^(-1) with
terminate called after throwing an instance of 'std::runtime_error' what(): matrix::solve(): inconsistent linear system
Thus some further investigation is required in this routine.
...this routine being sqrfree_parfrac(), because the linear system it tries to solve is indeed inconsistent. Anyway, notice what the commit message of commit 199b64938 says about sqrfree_parfrac(). Since 2001, this routine has been untouched. Your fix about the leading coefficient in Yun's algorithm was the first change since then. I'm not sure I can fix this. -richy.