-----maxima-bounces at math.utexas.edu wrote: -----
>I believe this came up in your work on the test suite. In the context of
>testing, this shouldn't matter. The expected output should simply be
>constructed to be precisely what is output by the code....
Yes, this issue came up in my work with the test suite. I think that
most terms of the form -p/q in the expected output need to be changed
to -(p/q). When I'm (mostly) satisfied with my test suite equality
test, I'll make it available so you all can help make the test suite
better.
I don't think that our test suite allows terrible bugs to slip through,
but it does allow some glitches to go unnoticed. Also, too much of the
test suite is willy-nilly. We need to expand it an give the basics a
really good workout...
Barton