Stanislav Terliakov
Stanislav Terliakov
This wording is almost the same as the message of the corresponding `RuntimeError`, so I think it's good enough. This behavior is a bit inconsistent, though, your FR to reject...
Was this implemented in #40250? See discussion in the linked PR #40731.
Ough, sorry, my bad - that is indeed limited to CloudWatch.
> Affected versions of django-stubs are potentially vulnerable to Security Misconfiguration. The inclusion of type stubs for deprecated and insecure password hashers (MD5PasswordHasher, SHA1PasswordHasher, and CryptPasswordHasher) may inadvertently encourage their...
I played a bit with this implementation, and noticed an odd discrepancy. I'm not sure if this can/should be fixed here, though. Cf. ``` [case testTypeIsListObjectFromList] # flags: --warn-unreachable from...
I don't consider Never vs unreachable issue critical at all: these cases differ only in error messages in fact, correctly rejecting unexpected checks. This might be a problem for users...
Missed this ticket - also resolved by #18465 and passes on current master.
Note that this isn't enum-specific. ```python class Test: a: int | None = 0 def set_a(self, value: int | None) -> None: self.a = value a = Test() reveal_type(a.a) #...
That's yet another case where applying return type context and args context separately causes overly broad inference.
Yes, sure! I'll update this tomorrow to keep (the only) non-autogen change in this PR. Re mypy: actually after fixing autogen there were only ~10 non-strict mypy errors, all either...