Tommy Hofmann
Tommy Hofmann
On our case the type does not determine the return value. Think of `GF(2)` and `GF(3)`.
Don't think they care. Base is full of `zero(T)` and its usage is increasing.
I am happy to add it for `fmpq` and `fmpz` and all element types with unique parents and throw errors for everything else :)
Hecke uses `Group` for the abelian groups. Alternatively the group Interface stuff just goes here without further discussions.
If you use the Nemo version (which computes the nullspace over Z), you will get a basis of the reactions, any other possible reaction is a linear combination of those....
Since `VectorSpace(F, n)` always produces a free modulo of rank `n`, we have a fixed basis and there is no ambiguity. We can just make it work for any ring.
> Is this supporting my comment or Claus's? Sorry, our posts crossed. This was meant to support your proposal. I don't think there is a good reason to not support...
The result will be in `Q`. There is a canonical action of `Mat(F, n, n)` on every `F^n`. I don't think this is a problem. It is the same as...
Yes, we should probably document this for the time being. Now I know what you mean @fieker. You should have mentioned the non-square case right at the beginning. Does Magma...
@wbhart Is there a clean way to do this with the current interface?