further typing annotations for methods has_something #40196
Conversation
|
Documentation preview for this PR (built with commit 6680d1f; changes) is ready! 🎉 |
|
@vincentmacri @tobiasdiez could you please have a look ? |
| @@ -1273,7 +1273,7 @@ def tamagawa_number(self, P, proof=None): | |||
|
|
|||
| return self.local_data(P, proof).tamagawa_number() | |||
|
|
|||
| def tamagawa_numbers(self): | |||
| def tamagawa_numbers(self) -> list: | |||
There was a problem hiding this comment.
I think this can be list[Integer] but if you want to leave specifying list types for future work that's fine too.
There was a problem hiding this comment.
I would prefer not. One point is that imho we should discuss and decide before moving forward : whether or not to distinguish between python integer and Sage Integer or just use a new abstract t_Integer class for typing. We should also talk about what kind of type we should use for polynomials.
There was a problem hiding this comment.
If the return/input type is clear, then you can use the specific Python int or sage Integer class. For the case where either can be returned, I would say something like
type Intlike = int | sage.Integer
should work.
There was a problem hiding this comment.
Doesn't Sage use its own Integer type everywhere (or at least it should)? I think the only place where we would return a python integer is in magic methods like __len__ or __hash__.
This is a more complicated matter for function parameters where many Sage functions (but not all) accept python ints.
There was a problem hiding this comment.
Doesn't Sage use its own Integer type everywhere (or at least it should)? I think the only place where we would return a python integer is in magic methods like len or hash.
It's a somewhat stupid test, but there are over 1000 instances of return 0. But yeah, most methods should return sage integer.
as a sequel of #40123
📝 Checklist