#3081 abs(...) -> "bfloat: attempted conversion of floating-point infinity."

[David Scherfgen]

I just came across something very strange: When doing abs(%e^(10000/3)+1);, I get the error "bfloat: attempted conversion of floating-point infinity."

OK, you might think, it's just a huge number. But when removing the /3, it works! It also works when removing the +1.

Maxima branch_5_37_base_407_g2af0c38 http://maxima.sourceforge.net
using Lisp GNU Common Lisp (GCL) GCL 2.6.12
Distributed under the GNU Public License. See the file COPYING.
Dedicated to the memory of William Schelter.
The function bug_report() provides bug reporting information.
(%i1) abs(%e^(10000/3)+1);

bfloat: attempted conversion of floating-point infinity.
 -- an error. To debug this try: debugmode(true);
(%i2) abs(%e^10000+1);
                                    10000
(%o2)                             %e      + 1
(%i3) abs(%e^(10000/3));
                                     10000/3
(%o3)                              %e

[David Scherfgen]

This seems to be the cause:

Maxima branch_5_37_base_407_g2af0c38 http://maxima.sourceforge.net
using Lisp GNU Common Lisp (GCL) GCL 2.6.12
Distributed under the GNU Public License. See the file COPYING.
Dedicated to the memory of William Schelter.
The function bug_report() provides bug reporting information.
(%i1) float(%e^(10000/3));
(%o1)                             i.nfE-16128
(%i2) float(%e^10000);
(%o2)                       8.806818225658306b4342

For some reason, Maxima doesn't "promote" the first result to a big float, but it does do so with the second one.

[David Scherfgen]

Fixed by https://sourceforge.net/p/maxima/code/ci/2c7bf01b64711c4823fa025995fda9dd9794c18c/

[Robert Dodier]

Note that since the float-to-bigfloat promotion code has been moved to branch promote-float-to-bigfloat and reverted on master, this bug report is no longer applicable. I've marked it "wont-fix" to distinguish it from "closed".