Fixes #1424
Fixes https://github.com/mypyc/mypyc/issues/939

Not that I really like `hasattr(...

Fix #11795

The fix is straightforward (but ideally generic callables should be normalized in...

There isn't really a good reason to have a default here, feels like it
mainly just increases th...

Fixes #2569

Fixes #13537

I also discovered another similar false negative, when a property in protocol w...

Fixes #13536

Fixes #5018
Fixes #5439
Fixes #10850

The implementation is simple but not the most beauti...

This pull request:

1. Fixes #8334. Overriding a dataclass attribute with a method or
pr...

During the error handling transform we have access to always
defined attributes, which allows u...

Fixes #10090

Unfortunately we cannot fully support this use case. Mypy requires explicit typ...

Per-module error codes were added in #13502, let's recommend using them.

The existing type ig...

Fixes #6801
Fixes #10577
Fixes #12642
Fixes #12337
Fixes #10639
Fixes #13390

All these...

This might slightly improve performance when using native ints,
and it's consistent with how ot...

Fixes #13521

Basic support for arithmetic, bitwise and comparison ops using native
int types i32 and i64. Al...

These are somewhat buggy and modern per-module options should be
preferred. I can't find any us...

I forgot to add this to yesterdays PR #13501

Fixes #9440

This is a bit non-trivial because I decided to make per-module code act as overri...

Fixes #12241

The code involved is really old. The mentioned test case does not
trigger any errors. I cannot ...

Blocking errors are a bad user experience and there's no reason for this
one to be one / there ...

Fixes #4536

This use case is specified by PEP 544 but was not implemented. Only instances (no...

Fixes #13499

Today this code reads like "stubs should all target 3.7" and this
is indeed how...

Fixes #6393

This is unspecified behavior in terms of PEP 544, so we just try to
do somethin...

Fixes #13226

Fixes #9282

It looks like `super()` checking is too strict for methods with self-types. Mypy...

The same as ErrorWatcher constructor.

Co-authored-by: Bas van Beek <[email protected]>

* Fix daemon crashes related to ParamSpec and TypeVarTuple

Fix daemon crash when using fine-g...

Fixes #4441

This uses a different approach than the initial attempt, but I re-used some of t...

Fixes #4125

Previously the code compared the original signatures for properties. Now we compa...

`subject='',` -> `subject: str = '',`

Fixes #13456

The fix required me to refactor the Constraint class to preserve the original ty...

The issue was exposed by merge of subtype visitors. Fix is actually trivial, but the diff is big...

Source commit:
https://github.com/python/typeshed/commit/5435ed76ef9346dfc3309d6ee21e1791c4b4ecd7

Co-authored-by: Nikita Sobolev <[email protected]>

* Allow unpacking from TypeVars by resolving bounds

TypeVars aren't iterable, but their bound...

Source commit:
https://github.com/python/typeshed/commit/a92da58328aa259184ff45bb7408e82426fb56...

Progress on mypyc/mypyc#868.

Also fix returning a value from inside a try block when the finally block
does a yield. (Which ...

More generally, we shouldn't be advising use of --no-strict-optional

Co-authored-by: Shantanu <[email protected]>

This now falls back to being handled due to bool being typed as final

Fixes #3863

This builds on top of some infra I added for recursive types (Ref #13297). Implem...

Fixes #685

This builds on top of some infra I added for recursive types (Ref https://github.c...

Typeshed currently has 6 allowlist entries relating to `__vectorcalloffset__` attributes in 3.10...

When building a module for web assembly with pygame-web ready to
use toolchain (clang 15 based...

`script_path` was entirely useless since `bin_dir` had been dropped long time
ago by #5690. Thi...

It is a minor change to fix the example in https://github.com/python/typeshed/issues/7904

* Add support for Final in literal_hash()
* Add test for tuple

Fixes #1362

I delete the old cryptic error message and instead:
* Enable the situations that...

With #13391 (adding `from __future__ import annotations` everywhere), it's now possible to remov...

Fixes https://github.com/python/mypy/issues/708
Fixes https://github.com/python/mypy/issues/548...

Resolves #5876

Co-authored-by: hauntsaninja <>
Co-authored-by: Ivan Levkivskyi <levkivskyi@g...

Fixes #5374

As discussed in the issue, we should allow this common use case, although it may ...

Implement some constraints tests for typevar tuples.

This will be nice for if anyone wants to use slicker new syntax for mypy internals without
wor...