Barton Willis <willisb at unk.edu> writes:
> Rupert,
>
> I tried the tdzero / tdpn route. Are you sure the tdpn mechanism works? I get an infinite loop with
> integrate(a^(p*x) * b^(q*x) * c^(r*x),x).
>
> OK:
>
> (%i2) integrate(a^(p*x) * b^(q*x),x);
> Is log(b)*q/(log(a)*p) equal to -1?
> no;
>
> No OK:
>
> (%o2) (a^(p*x))^(log(b)*q/(log(a)*p)+1)/(log(a)*p*(log(b)*q/(log(a)*p)+1))
> (%i3) integrate(a^(p*x) * b^(q*x) * c^(r*x),x);
>
> Is log(c)*r/(log(a)*p)+log(b)*q/(log(a)*p) equal to -1?
> no;
>
> Is log(c)*r/(log(a)*p)+log(b)*q/(log(a)*p) equal to -1?
> no;
>
> Is log(c)*r/(log(a)*p)+log(b)*q/(log(a)*p) equal to -1?
> no;
>
> /* Give up and answer yes: */
>
> Is log(c)*r/(log(a)*p)+log(b)*q/(log(a)*p) equal to -1?
> yes;
>
> (%o3) x
Hmm. I confess that I didn't actually try it out. It's a bit confusing
though since, looking at the code for meqp, I would expect to end up at
the call to dcompare at the bottom (and then meqp would hopefully return
nil when it got $pn back out).
I don't have time to look at it now, but maybe try tracing dcompare to
see what's going on?
Rupert
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 315 bytes
Desc: not available
URL: <http://www.math.utexas.edu/pipermail/maxima/attachments/20130523/c750da7b/attachment.pgp>;