Robert, Valery, et al:
I just submitted several fixes to CVS. The tensor packages are now again
fully functional (assuming I didn't miss something.) All tests are
successful, and all demos run correctly, too.
Thanks for making my job easy by actually finding the cause (changes to
sum()) of many of the failures!
Viktor
-----Original Message-----
From: Robert Dodier [mailto:robert.dodier at gmail.com]
Sent: Wednesday, May 24, 2006 12:23 PM
To: Viktor T. Toth
Cc: Valery Pipin; maxima at math.utexas.edu
Subject: Re: [Maxima] itensor is broken, partly
Hello,
At least for itensor, changing the defn of lc2kdt from
lc2kdt(exp):=block([exp:expand(exp),n:nterms(exp)],
sum(_lc2kdt0(if n > 1 then part(exp,i) else exp),i,1,n))$
to
lc2kdt(exp):=block([exp:expand(exp),n:nterms(exp)],
sum(_lc2kdt0(if n > 1 then part(exp,i%) else exp),i%,1,n))$
i.e. renaming the summation index to avoid a name collision,
causes problems 26 and 27 to yield the expected results.
I guess we can have a discussion about sum/product evaluation now.
Dunno about ctensor or atensor.
Robert