Ugh! I'm guessing that my hypergeometric code has this bug for some
time. Previously, rtest_hg had a funny scheme that did something like
load, compile, load, so I didn't see the bug. I removed this scheme
from rtest_hg because it may have caused trouble for some non-CCL
Lisps. Since make clean doesn't delete compiled code that is in
/share, you have to be careful.
Usually, non-mathematical bugs flummox me, but I'll see what I can
do. Thanks for testing with SBCL + Clisp.
Barton
-----maxima-bounces at math.utexas.edu wrote: -----
>On?7/27/09,?Barton?Willis?<willisb at unk.edu>?wrote:
>
>>?(%i2)?hypergeometric([1/2,1],[3/2],2807.0);
>>
>>?BIGFLOAT:??Unable?to?convert?false?to?a?CL?or?BIGFLOAT?number
>>??--?an?error.??To?debug?this?try?debugmode(true);
>>
>>?(%i3)?load("hypergeometric")$
>>
>>?(%i4)??hypergeometric([1/2,1],[3/2],2807.0);
>>
>>?(%o4)????????????3.5629454083782787E-4?-?0.0296482231479244?%i
>
>I?see?the?same?behavior?w/?SBCL.
>Works?the?first?time?w/?Clisp.
>
>Robert