axiom-developer
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Axiom-developer] #325 Severe bug in normalize$EFSTRUC


From: root
Subject: [Axiom-developer] #325 Severe bug in normalize$EFSTRUC
Date: Thu, 20 Dec 2007 16:13:39 -0500

>> Status: fix proposed 


I've spent the last 2 hours dragging thru the fricas change list
trying to reverse-engineer #325 Severe bug in normalize$EFSTRUC

I'm giving up on this task until someone posts a patch.

A "fix proposed" status needs to say what the bug was and what the fix
is. The fact that it runs correctly in Fricas provides no useful
information about what the fix might be and, therefore, "fix proposed"
is not the proper tag. 

Where Axiom has fixes I cite the actual diff-Naur patch number so
someone can look it up in the mailing list. The mailing list post
provides an explanation for the fix.




If we ever hope to keep the algebra from diverging, the only possible
solution seems to be posting diff-Naur patches, possibly in a central
location. Without that level of documentation it becomes impossible
to find what changed. And the algebra should have a converged test
suite so we know we all get the same answers. I posted my regression
test code and there are over 100 input files with known answers that
can be automatically checked.

Tim




reply via email to

[Prev in Thread] Current Thread [Next in Thread]