"Vadim V. Zhytnikov" <vvzhy@mail.ru> writes:
> As far as I understand there is no any internal special function for
> representing indexed tensor objects. They just mean that for example
> for itensor
> ZZ([A,B],[X,Y,Z])
> is a tensor with two covariant and three contravariant indices.
> But for Maxima in general this is nothing but function ZZ with
> two arguments.
OK, but on the Lisp-level the Maxima-function is represented as a list. And
that is exactly my problem with _components_. If one uses the "workaround"
for components, an expression like ZZ(L1,L2):=BLOCK ... defines a function on
lisp-level which replaces each occurence of ZZ(..) with the rhs of the
function. The $components function in contrast generates a property-list with
properties like TEXPRS and there is (as far as I can see) no possibilty to
evaluate this structure ...
Wolfgang Zocher