Problem with algsys and realonly -& gt; false solution from solve_rat_ineq.
Subject: Problem with algsys and realonly -& gt; false solution from solve_rat_ineq.
From: Leo Butler
Date: Fri, 30 Apr 2010 22:36:54 +0100 (BST)
On Fri, 30 Apr 2010, Volker van Nek wrote:
< Hello Robert,
<
< I want you to get noticed that the bug described below is fixed in cvs
< but of course present in 5.21.1. algsys doesn't give any solution if
< realonly is set to true. I can imagine that as a result of this some
< code is broken.
<
< I succeeded in implementing this stupid bug into versions 5.21.0. and
< 5.21.1. I run the src-test-suite and it passed!
< I just noticed that there are only two tests for algsys in rtest8.mac
< and none of them with the realonly flag. Rather poor.
Volker, I've added a few additional tests to rtest8 to prevent
this particular regression, but it would be good to add a few more.
I've also added tests to the share testsuite for solve_rat_ineq.
Leo
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.