Further considerations:
- soon there will be a PR to add auto-matrix indices capabilities. In the construction of a tensor, indices marked by True instead of an index symbol will be considered auto-matrix indices, they behave upon multiplication with other tensors as matrices or vectors (there can be 0, 1 or 2 automatrix indices for each TensorIndexType, 1 => vector behavior, 2=> matrix behavior). A tensor can have many automatrix indices, it is important they are applied on different TensorIndexType. - gamma matrices will have default auto-matrix behavior on their spinor indices. - tensor indices are conceptually strictly related to representations of Lie groups, in the future we could generate TensorIndexType instances from an object representing a representation of a Lie group. - Lie groups are the reason why it is better to represent gamma matrices as (Lorentz, Spinor, Spinor), the (Lorentz)-matrix representation loses the spinor-type transformation properties. - What happens on numerical indices? My guess is that a tensor gets reduced in its rank. E.g. *GammaMatrix(U(3))*, this means that in a certain basis we are representing gamma^3 covariant. Operators U( ) and D( ) (up and down) need to be introduced. -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/sympy. For more options, visit https://groups.google.com/groups/opt_out.
