Hi,
I read the conversation about changing how limit deals with integrate()
etc. and I completely agree that it's the right thing to
do. However... I've got some code that I wrote this morning and worked
until I pulled the latest version of maxima, so I was kind of surprised!
Basically, it looks something like
approx_for_eps: 'integrate( something_nasty, s, A-eps, A );
and then you get a Taylor series for the approximation by
differentiating the above and taking the limit as eps -> 0+. Problem is,
the limits now all include noun forms :)
Is there a way to tell Maxima that "Yes, I know what I'm doing. This
integral is continuous." ?
I think changing the limit code to be more mathematically correct is
good. But I also think that one needs to be able ask for the more
specific result if the integrand is known to be smooth (say) but is far
From being integrable.
Rupert
P.S. Sorry if two copies of this appear - I sent the first one via gmane
and it appears to have silently got eaten.
-------------- 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/20090916/a5d0de9e/attachment.pgp