Dear Christian, On Mon, 2 Aug 2004, Christian Bauer wrote:
(ncmul(gamma~mu,gamma~nu))^dagger = ncmul((gamma~nu)^dagger,(gamma~mu)^dagger).
Shouldn't this work in the same way for matrix, color, and possibly for user-defined classes too?
That is a good question. At the moment complex conjugation of matrices complex conjugates their entries but does not do transposition. When I wrote complex conjugation I only took into account my own experience with complex conjugation of strings of Dirac matrices and this told me that doing hermitian conjugation on them would be useful. However, I almost (?) never use matrices in GiNaC. Same for the color algebra. If you think that hermitian conjugation is the most usefull way of complex conjugating them, I could write a patch to do it. Personally I do not care very much because I hardly use GiNaCs matrices and color objects. Best, Chris