> On Wed, Jan 21, 2009 at 6:12 AM, Barton Willis <willisb at unk.edu> wrote:
> > I was looking into bug 2465066 "unsimplified result from
integrate."...
> > Changing simplify to ($expand ... 0 0) fixes the bug. I'm not sure
why
> > ($expand ... 0 0) works and simplify doesn't. Maybe the code needs to
be
> > fixed elsewhere. Advice? Is ($expand ... 0 0) an OK fix?
>
> Well, ($expand ... 0 0) is possibly an acceptable interim patch until
> you find the source of the problem, but the problem should really be
> fixed at the source. Some piece of code is returning as simplified
> without going through the simplifier, which may be causing other
> problems as well.
Thanks for the advice.
I'll keep looking---I think I need to find the code that applies
additivity (or
outativity) for antiderivatives. If somebody knows where this code is, let
me know ($1 for tapping a switch with a mallot and $9,999 for knowing
which
switch to tap). My fix only changes (simplify ... ) to ($expand ... 0 0),
so I'd guess the code was already cynically patched by its author :(
Barton