[Larry Bates <bates@tiki.math.ucalgary.ca>] Bug#151416: maxima integrates incorrectly



tags 151416 forwarded
thanks


Greetings!  I've received this bug report from a Debian user.  Has
this been fixed in the latest CVS?  If not, could someone please look
into it?

Many thanks!


-- 
Camm Maguire			     			camm@enhanced.com
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah
------- Start of forwarded message -------
Subject: Bug#151416: maxima integrates incorrectly
Date: Sat, 29 Jun 2002 17:21:11 -0600
From: Larry Bates <bates at tiki>

Package: maxima
Version: 5.6

Maxima integrates incorrectly in user defined functions.  Here is a 
sample output from a run:

bates@symplectic:~$ maxima
GCL (GNU Common Lisp)  Version(2.4.0) Thu Mar 28 08:08:45 EST 2002
Licensed under GNU Library General Public License
Contains Enhancements by W. Schelter
Maxima 5.6 Thu Mar 28 08:08:39 EST 2002 (with enhancements by W. Schelter).
Licensed under the GNU Public License (see file COPYING)
(C1) f(a):=integrate(sin(a)/(1-cos(a)^2*sin(t)^2), t,0,2*%Pi);

                                      SIN(a)
(D1)           f(a) := INTEGRATE(-------------------, t, 0, 2 %PI)
                                        2       2
                                 1 - COS (a) SIN (t)
(C2) f(%Pi/2);

(D2)                      2 %PI
(C3) f(3.14159/2);


RAT replaced -1.76038469784954E-12 by -1//568057653092 = 
-1.76038469785046E-12

RAT replaced -1.76038469784954E-12 by -1//568057653092 = 
-1.76038469785046E-12

RAT replaced -1.76038469784954E-12 by -1//568057653092 = 
-1.76038469785046E-12

RAT replaced -1.76038469784954E-12 by -1//568057653092 = 
-1.76038469785046E-12

RAT replaced -1.76038469784954E-12 by -1//568057653092 = 
-1.76038469785046E-12
(D3)                        0



The correct answer is 2Pi, and in fact does not depend on the choice of 
the variable a if 0<a<Pi.

Sorry, but I have no skill to locate the error in the source code, let 
alone provide a patch.

Cheers,

Larry Bates

-- 
*********************************************************************

Larry Bates                  Telephone: (403) 220-3942
Department of Mathematics    Facsimile: (403) 282-5150
University of Calgary        Electronic mail: bates@tiki.math.ucalgary.ca
Calgary, Alberta
Canada    T2N 1N4

*********************************************************************

------- End of forwarded message -------