Maxima -- GPL CAS based on DOE-MACSYMA

Thanks for pointing out the second issue. Both of these are fixed in
plot.lisp, rev 1.144.

Closing bug report.

Thanks for looking at this,

draw also works fine with this change.

I see the same piece of code in coerce-lisp-function-or-lisp-lambda, but I
don't know of possible implications of changing this function too.

Found the problem. It's in coerce-maxima-function-or-maxima-lambda. In
this routine, the function does:

(if ($numberp result)
($float result)
result)))

But in this case, result is '$%pi, and ($numberp '$%pi) returns NIL.
Perhaps this is a misunderstanding of what $numberp does, but I see no
reason why we can't just get rid of the test and always call $float. With
this fix, the graph is pi between -1 and 1.