PEP 483 has invalid variance example

[cowfoodbagman]

PEP 483 explains variance and provides this example:

T_co = TypeVar('T_co', Employee, Manager, covariant=True)
T_contra = TypeVar('T_contra', Employee, Manager, contravariant=True)

class Base(Generic[T_contra]):
    ...

class Derived(Base[T_co]):
    ...

Mypy gives this error when I try this example out:
error: Variance of TypeVar "T_co" incompatible with variance in parent type [type-var]

[StanFromIreland]

This belongs in the peps repo, note that PEPs are not documentation and may be out of date, 483 is finalized.

[cowfoodbagman]

is the documentation issue or a mypy issue?

[StanFromIreland]

This is a PEPs are not documentation issue, and so this is probably fine unless it is some weird regression.

[cowfoodbagman]

Sorry I missed the comment about it not being a documentation issue. If this is an issue, presumably it would be in mypy. I just don't know if pep documentation doesn't match actual behavior whether that is considered acceptable is some cases. I guess maybe opening as issue for mypy and getting feedback there would be the logical next step?

[cowfoodbagman]

I brought this question to the mypy questions chat

[terryjreedy]

Sometimes when PEPs are implemented, and more ofter thereafter, behavior is sometimes intentionally (with approval) different from the PEP spec. Thus PEPs can become obsolete ( and yes, this can be confusing). The current docs should correspond to the current behavior.