grhkm21

I can overload the constructor of `SchemeMorphism_sum` and check the correctness of types there, I guess. Let me try it first. Just a heads up, this PR is my first...

I don't understand how multi-hierarchy should be done in Sage. Here I want the `Homset_sum` (let's say `X`) to be both a FormalSum and a SchemeMorphism. If I put `FormalSum`...

Update: I don't think this is possible, since not everything is hashable. For example, consider hashing an endomorphism of an elliptic curve over $\mathbb{Q}_3$, where you map $x$ to $ax$,...

Okay I spent an unnecessary large amount of time on this and here are some thoughts for future people who come across this 1. EllipticCurveHom cannot inherit from SchemeMorphism (for...

Something about "Test modularized distributions" is failing, and not just for me

@nbruin > Concerning the motivation in the ticket: it would look to me you'd first need endomorphism rings of abelian varieties and then homomorphisms to/from them can be modules over...

> A relevant old ticket: > > * [Coercion for homsets #14279](https://github.com/sagemath/sage/issues/14279) I will have a look at that. This line seems interesting: > One important change: I let `sage.categories.map.Map`...

Very nice to see that all examples in #14279 have been sped up by ~5x though, either by my CPU or by Sage

I will fix this after my exams are done, but I think a more productive and better way to approach this is to fix the issue with SchemeMorphism and Morphism...

draft until I can think of a sane way to implement this that does not copy and pasting more code and maintaining the same code on two different places