"trigrat" caught in infinite loop - depending on how you call it

On 2012-11-15, Robert Dodier <robert.dodier at gmail.com> wrote:

> I think I've fixed the bug which trigrat exposed -- actually the problem
> is that some functions related to rational expressions don't bind
> VARLIST, so it leaks from one call into the next. I found this bug in
> rat, ratexpand, and the simplifications for plus and times (all exposed
> by the examples posted in this thread), but there may be others; I
> didn't carry out an exhaustive search.
>
> Patch shown below and some test cases were committed as c36715e95ed.

I've reverted that commit, as it caused some test cases (mostly
integrals) to return different results. I tried to fix up the situation
by calling the old versions of $RAT and $RATEXPAND internally, but I
wasn't able to resolve all the differences. So I put things back they
way they were before. Maybe at some point I'll have another go at it.

best

Robert Dodier