Repository navigation
attrs plugin error: "The erased type of self '...*' is not a supertype of its class" on class definition with a generic using a constrained type variable, but then seems to work fine after definition #5542
Description
Consider the following example:
from typing import TypeVar, Generic
import attr
class V:
pass
class W:
pass
T = TypeVar("T", V, W)
@attr.s(auto_attribs=True)
class A(Generic[T]):
v: T
reveal_type(A(W()))My expectation here would be a revealed type of A[W*], which I do get.
But I also get an error on the @attr.s definition:
The erased type of self 'V*' is not a supertype of its class 'A[T`1]'
The erased type of self 'W*' is not a supertype of its class 'A[T`1]'
if I remove the constraints on T, (i.e. T = TypeVar("T")) then the errors go away.
Activity
gvanrossum commentedon Aug 30, 2018
Maybe the attrs plugin author (@euresti) can shed some insight?
ilevkivskyi commentedon Aug 30, 2018
This may be related to #5263 (if not a duplicate).
ilevkivskyi commentedon Aug 30, 2018
Just to clarify, both issues may be caused by bad interaction of patched
self/clswithcheckmember.bind_self().euresti commentedon Aug 31, 2018
Actually. I can repro with dataclasses too.
So it's related to the added
__cmp-ish__methodsilevkivskyi commentedon Aug 31, 2018
OK, the problem is here and here where we unconditionally grab
id=1for the self-type type variable thus causing a type variable clash (the same probably happens forattrs). I think the solution would be to check somehow the next available id instead.Also btw by agreement ids for generic functions/methods should have negative numbers.
euresti commentedon Aug 31, 2018
Would the solution just be to make the id
-1? It seems to me like -1 should never be used at this point because negative numbers are used for methods/functions and I'm at this point creating it right? Or are the id's global in scope?ilevkivskyi commentedon Sep 1, 2018
After some thinking it seems to me just using
-1should be fine since the methods are auto-generated anyway. Btw, I have found thatsenamal_namedtuplealso uses1instead of-1. Could you please try replacing it by-1in both plugins and in named tuple code and play with it trying to break (try some nested classes within generic functions, maybe some scenarios with type inference)?Avoid type-var collisions by using -1 instead of 1.
Avoid type-var collisions by using -1 instead of 1.
Avoid type-var collisions in attrs and dataclasses (#5552)