BUG: Strange behavior in diff with depends?



Am Dienstag, den 01.02.2011, 21:16 -0800 schrieb dlakelan:
> 
> Maxima 5.22.1 http://maxima.sourceforge.net
> using Lisp SBCL 1.0.34.0.debian
> 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) depends(mu,T);
> (%o1)                               [mu(T)]
> (%i2) depends(T,[z,t]);
> (%o2)                              [T(z, t)]
> (%i3) diff(mu,z);
>                                      dmu dT
> (%o3)                               --- --
>                                      dT  dz
> (%i4) diff(1/mu,z);
> (%o4)                                  0
> (%i5) diff(1/mu,mu)*diff(mu,T)*diff(T,z);
>                                       dmu dT
>                                       --- --
>                                       dT  dz
> (%o5)                              - ------
>                                          2
>                                        mu
> 
> -------------
> 
> shouldn't %o4 be the same as  %o5 be the same?

We had a bug in Maxima 5.22 (I have introduced it, when correcting
another bug). In Maxima 5.23.2 the bug is no longer present. Perhaps,
you can download and install the newest version Maxima 5.23.2.

(%i2) depends(mu,T);
(%o2) [mu(T)]

(%i3) depends(T,[z,t]);
(%o3) [T(z,t)]

(%i4) diff(mu,z);
(%o4) 'diff(mu,T,1)*'diff(T,z,1)

(%i5) diff(1/mu,z);
(%o5) -'diff(mu,T,1)*'diff(T,z,1)/mu^2

(%i6) diff(1/mu,mu)*diff(mu,T)*diff(T,z);
(%o6) -'diff(mu,T,1)*'diff(T,z,1)/mu^2

Dieter Kaiser