Fix/protocol explicit self

[vgvr0]

Summary

Fixes #5089

Protocol matching could incorrectly accept bound methods whose explicit self: annotation was incompatible with the concrete receiver.

The issue occurred because bound-method matching strips the first parameter before callable comparison, so an explicit receiver constraint such as self: ElementOpsMixin[int] could be lost. This allowed cases like Series[bool] to satisfy a protocol through normal bool <: int argument compatibility even though ElementOpsMixin[bool] is not compatible with invariant ElementOpsMixin[int].

This change:

• checks explicit self: compatibility before Protocol callable matching discards the receiver parameter;

• reuses the same compatibility logic for single functions and overloaded methods;

• distinguishes unchanged, partially filtered, and fully incompatible overload sets;

• rejects Protocol matching when every overload has an incompatible explicit receiver;

• preserves the existing recursion guard for protocol-typed self: annotations and recursive Protocol conformance;

• leaves ordinary bool <: int callable argument compatibility unchanged.

• I am a human, and I fully understand the code I am submitting.

Test Plan

Added regression coverage for:

• the original Series[bool]() - True false negative;
• valid Series[int]() - True;
• direct bound-method calls with an incompatible explicit self:;
• valid and invalid Protocol conformance;
• overloaded methods where every explicit receiver is incompatible;
• existing partial/all-compatible overload behavior;
• recursive and mutually recursive Protocols;
• generic explicit self: handling;
• ordinary bool <: int argument compatibility.

Validation:

cargo test -p pyrefly protocol
183 passed

cargo test -p pyrefly class_subtyping
29 passed

cargo test -p pyrefly callable
284 passed

cargo fmt --all -- --check
passed

git diff --check
passed

[meta-codesync]

This pull request has been imported. If you are a Meta employee, you can view this in D123197668. (Because this pull request was imported automatically, there will not be any future comments.)

[github-actions]

Diff from mypy_primer, showing the effect of this PR on open source code:

============================================================
SUMMARY
============================================================
Total: +37 new errors, -25 fixed errors
By preset: +36/-10 (default), +36/-11 (strict)

Projects with changes (6):
  scikit-learn: +0 -1
  pandas-stubs: +12 -0
  hydpy: +1 -0
  pandas: +23 -22
  static-frame: +0 -1
  numpy-stl: +1 -1
============================================================

FULL DIFF DETAILS
------------------------------------------------------------

scikit-learn (https://github.com/scikit-learn/scikit-learn)
- ERROR sklearn/inspection/tests/test_partial_dependence.py:769:16-25: The type of this argument is unknown [unknown-argument-type]

pandas-stubs (https://github.com/pandas-dev/pandas-stubs)
+ ERROR tests/indexes/bool/test_truediv.py:30:15-23: `/` is not supported between `Index[bool]` and `Literal[True]` [unsupported-operation]
+ ERROR tests/indexes/bool/test_truediv.py:37:15-23: `/` is not supported between `Literal[True]` and `Index[bool]` [unsupported-operation]
+ ERROR tests/indexes/bool/test_truediv.py:49:15-23: `/` is not supported between `Index[bool]` and `list[bool]` [unsupported-operation]
+ ERROR tests/indexes/bool/test_truediv.py:56:15-23: `/` is not supported between `list[bool]` and `Index[bool]` [unsupported-operation]
+ ERROR tests/series/bool/test_truediv.py:30:15-23: `/` is not supported between `Series[bool]` and `Literal[True]` [unsupported-operation]
+ ERROR tests/series/bool/test_truediv.py:37:15-23: `/` is not supported between `Literal[True]` and `Series[bool]` [unsupported-operation]
+ ERROR tests/series/bool/test_truediv.py:44:21-24: No matching overload found for function `pandas.core.series.Series.truediv` called with arguments: (Literal[True]) [no-matching-overload]
+ ERROR tests/series/bool/test_truediv.py:55:17-20: No matching overload found for function `pandas.core.series.Series.truediv` called with arguments: (Literal[True]) [no-matching-overload]
+ ERROR tests/series/bool/test_truediv.py:62:22-25: No matching overload found for function `pandas.core.series.Series.rtruediv` called with arguments: (Literal[True]) [no-matching-overload]
+ ERROR tests/series/bool/test_truediv.py:73:18-21: No matching overload found for function `pandas.core.series.Series.rtruediv` called with arguments: (Literal[True]) [no-matching-overload]
+ ERROR tests/series/bool/test_truediv.py:87:15-23: `/` is not supported between `Series[bool]` and `list[bool]` [unsupported-operation]
+ ERROR tests/series/bool/test_truediv.py:94:15-23: `/` is not supported between `list[bool]` and `Series[bool]` [unsupported-operation]

hydpy (https://github.com/hydpy-dev/hydpy)
+ ERROR hydpy/auxs/iuhtools.py:516:13-20: `ndarray[tuple[Any, ...], dtype[float64]] | Unknown` is not assignable to variable `t` with type `VectorFloatFlex` [bad-assignment]

pandas (https://github.com/pandas-dev/pandas)
- ERROR pandas/core/arrays/sparse/array.py:1440:17-29: The type of this argument is unknown [unknown-argument-type]
- ERROR pandas/core/dtypes/missing.py:653:16-31: Returning implicit Any from function declared to return "_ArrLikeT" [no-any-return-implicit]
+ ERROR pandas/core/dtypes/missing.py:653:16-31: Returned type `Index | ndarray | Unknown` is not assignable to declared return type `_ArrLikeT` [bad-return]
- ERROR pandas/core/dtypes/missing.py:655:16-43: Returned type `ndarray | Unknown` is not assignable to declared return type `_ArrLikeT` [bad-return]
+ ERROR pandas/core/dtypes/missing.py:655:16-43: Returned type `Index | ndarray | Unknown` is not assignable to declared return type `_ArrLikeT` [bad-return]
+ ERROR pandas/core/frame.py:2267:42-51: Cannot index into `list[ArrayLike]` [bad-index]
+ ERROR pandas/core/frame.py:2267:42-51: Cannot index into `list[Unknown]` [bad-index]
+ ERROR pandas/core/frame.py:2272:35-69: Cannot index into `list[ArrayLike]` [bad-index]
+ ERROR pandas/core/frame.py:2272:35-69: Cannot index into `list[Unknown]` [bad-index]
- ERROR pandas/core/indexes/base.py:6113:16-25: Returning Any from function declared to return "Hashable" [no-any-return-explicit]
+ ERROR pandas/core/indexes/base.py:6113:16-25: Returned type `Index | Any` is not assignable to declared return type `Hashable` [bad-return]
- ERROR pandas/core/internals/construction.py:414:16-48: Implicit conversion of `Literal[False] | ndarray[tuple[Any, ...], dtype[numpy.bool]] | Any` to `bool` is not allowed [implicit-bool]
+ ERROR pandas/core/internals/construction.py:412:21-37: Cannot index into `list[Any]` [bad-index]
+ ERROR pandas/core/resample.py:2878:14-23: Object of class `Index` has no attribute `asi8` [missing-attribute]
+ ERROR pandas/core/resample.py:2889:33-36: Expected 0 positional arguments, got 1 in function `object.__init__` [bad-argument-count]
+ ERROR pandas/core/resample.py:2889:38-43: Unexpected keyword argument `dtype` in function `object.__init__` [unexpected-keyword]
+ ERROR pandas/core/resample.py:2889:44-54: Argument `ExtensionDtype | PeriodDtype | dtype` is not assignable to parameter `dtype` with type `_HasDType[dtype] | _HasNumPyDType[dtype] | dtype | list[Any] | str | _DTypeDict | tuple[Any, Any] | type[Any] | None` in function `numpy.ndarray.__new__` [bad-argument-type]
- ERROR pandas/core/reshape/concat.py:756:23-69: The type of `overlap` is unknown; it is inferred as an implicit `Any` [unknown-variable-type]
- ERROR pandas/core/reshape/concat.py:808:19-65: The type of `overlap` is unknown; it is inferred as an implicit `Any` [unknown-variable-type]
- ERROR pandas/core/reshape/merge.py:3375:12-14: Cannot infer type of empty container; it will be treated as containing `Any` [implicit-any-empty-container]
- ERROR pandas/core/reshape/merge.py:3385:8-12: Implicit conversion of `list[Unknown]` to `bool` is not allowed [implicit-bool]
+ ERROR pandas/core/reshape/merge.py:3385:8-12: Implicit conversion of `list[Any]` to `bool` is not allowed [implicit-bool]
- ERROR pandas/io/stata.py:1286:14-22: Fixpoint iteration did not converge. Inferred result `list[LiteralString | int] | list[int | list[LiteralString | int] | list[LiteralString | int | list[LiteralString | int] | list[LiteralString | int | list[LiteralString | int] | list[Unknown] | Unknown] | list[Unknown] | Unknown] | list[LiteralString | int | list[LiteralString | int] | list[Unknown] | Unknown] | list[int | list[LiteralString | int] | list[LiteralString | int | list[LiteralString | int] | list[LiteralString | int | list[LiteralString | int] | list[Unknown] | Unknown] | list[Unknown] | Unknown] | list[LiteralString | int | list[LiteralString | int] | list[Unknown] | Unknown] | list[int | str] | list[Unknown] | str | Unknown] | list[int | str] | list[Unknown] | str | Unknown] | list[int | str] (instance attribute)`. Adding annotations may help. [non-convergent-recursion]
- ERROR pandas/io/stata.py:1286:29-38: Fixpoint iteration did not converge. Inferred result `list[dtype | list[dtype | list[dtype | list[dtype | list[dtype] | list[Unknown] | Unknown] | list[dtype] | list[Unknown] | Unknown] | list[dtype | list[dtype] | list[Unknown] | Unknown] | list[dtype | str] | list[dtype] | list[Unknown] | str | Unknown] | list[dtype | list[dtype | list[dtype] | list[Unknown] | Unknown] | list[dtype] | list[Unknown] | Unknown] | list[dtype | list[dtype] | list[Unknown] | Unknown] | list[dtype | str] | list[dtype] | list[Unknown] | str | Unknown] | list[dtype | str] | list[dtype] (instance attribute)`. Adding annotations may help. [non-convergent-recursion]
- ERROR pandas/io/stata.py:1295:14-22: Fixpoint iteration did not converge. Inferred result `list[list[list[list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[str] (instance attribute)`. Adding annotations may help. [non-convergent-recursion]
- ERROR pandas/io/stata.py:1298:14-22: Fixpoint iteration did not converge. Inferred result `list[list[list[list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[str] (instance attribute)`. Adding annotations may help. [non-convergent-recursion]
- ERROR pandas/io/stata.py:1498:16-19: The type of this argument is unknown [unknown-argument-type]
- ERROR pandas/io/stata.py:1806:24-38: Object of class `list` has no attribute `startswith` [missing-attribute]
- ERROR pandas/io/stata.py:1808:81-84: Argument `list[list[list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown` is not assignable to parameter `fmt` with type `str` in function `_stata_elapsed_date_to_datetime_vec` [bad-argument-type]
- ERROR pandas/io/stata.py:1813:47-60: Argument `list[list[list[list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[str]` is not assignable to parameter `lbllist` with type `Sequence[str]` in function `StataReader._do_convert_categoricals` [bad-argument-type]
- ERROR pandas/io/stata.py:1855:20-23: The type of this argument is unknown [unknown-argument-type]
- ERROR pandas/io/stata.py:1860:20-23: The type of this argument is unknown [unknown-argument-type]
+ ERROR pandas/io/stata.py:1930:33-50: Cannot index into `list[dtype | str]` [bad-index]
+ ERROR pandas/io/stata.py:1930:33-50: Cannot index into `list[dtype]` [bad-index]
+ ERROR pandas/io/stata.py:1930:33-50: Cannot index into `list[Unknown]` [bad-index]
+ ERROR pandas/io/stata.py:1931:32-48: Cannot index into `list[LiteralString | int]` [bad-index]
+ ERROR pandas/io/stata.py:1931:32-48: Cannot index into `list[int | str]` [bad-index]
+ ERROR pandas/io/stata.py:1931:32-48: Cannot index into `list[Unknown]` [bad-index]
+ ERROR pandas/io/stata.py:1932:32-48: Cannot index into `list[str]` [bad-index]
+ ERROR pandas/io/stata.py:1932:32-48: Cannot index into `list[Unknown]` [bad-index]
+ ERROR pandas/io/stata.py:1933:32-48: Cannot index into `list[str]` [bad-index]
+ ERROR pandas/io/stata.py:1933:32-48: Cannot index into `list[Unknown]` [bad-index]
- ERROR pandas/plotting/_matplotlib/hist.py:102:18-39: The type of `values` is unknown; it is inferred as an implicit `Any` [unknown-variable-type]
- ERROR pandas/plotting/_matplotlib/hist.py:104:39-45: The type of this argument is unknown [unknown-argument-type]
- ERROR pandas/tests/io/test_stata.py:875:49-58: Argument `list[list[list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown] | list[list[str] | list[Unknown] | str | Unknown] | list[str] | list[Unknown] | str | Unknown` is not assignable to parameter `x` with type `Buffer | SupportsIndex | SupportsInt | SupportsTrunc | str` in function `int.__new__` [bad-argument-type]

static-frame (https://github.com/static-frame/static-frame)
- ERROR static_frame/core/bus.py:1086:33-47: Argument `IndexBase | Iterable[TLabel]` is not assignable to parameter `obj` with type `Sized` in function `len` [bad-argument-type]

numpy-stl (https://github.com/WoLpH/numpy-stl)
- ERROR stl/base.py:763:13-34: The type of this argument is unknown [unknown-argument-type]
+ ERROR stl/base.py:1092:9-17: Class member `BaseMesh.__iter__` overrides parent class `Mapping` in an inconsistent manner [bad-override]