[GiNaC-devel] Matrices vs indexed [Was: Fixes w.r.t. indexed objects]

[Chris Dams]

Dear Alexei,

On Sun, 30 Jul 2006, Sheplyakov Alexei wrote:

> But even GiNaC developers *do* write such things, see:
> http://thep.physik.uni-mainz.de/pipermail/ginac-list/2005-April/000638.html

Yes, unfortonately they do :-(. Also in code that went into GiNaC :-(. See
http://www.ginac.de/pipermail/ginac-devel/2006-July/000985.html . IIRC it
was also Vladimir who submitted the first patch to start doing automatic
index renaming. However, if we start doing that we also should be careful
that our indexed expressions make sense. This means that symbols in
indices should occur once or twice in a product of indexed objects and if
they occur twice in a varidx it should be in an up/down pair. For
everything else there can be no guarantee whatsoever. It doesn't mean
anyting. We shouldn't be showing that kind of stuff on our webpage.

Best,
Chris