python / python/mypy

`copy.replace` type inference regression: breaks Self-returning __replace__ with bound TypeVar

Open
#21,672 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug topic-self-types topic-type-variables
Dominant language
Python
Stars
20.6k
Forks
3.3k
PR merge metrics
PR metrics pending

Description

Bug Report

typeshed PR python/typeshed#14786 (synced into mypy) changed copy.replace from a bound-TypeVar signature to a covariant protocol. This breaks copy.replace for __replace__ methods that return Self (dataclasses, namedtuples, etc.) when the argument is a bounded TypeVar.

To Reproduce

https://mypy-play.net/?mypy=master&python=3.14&gist=32a94d730b07faeb680192a3fdad1b0f

import copy
from dataclasses import dataclass
from typing import TypeVar, assert_type

@dataclass(frozen=True)
class BaseConfig:
    pg_ssl_key: str | None = None

@dataclass(frozen=True)
class SubConfig(BaseConfig):
    pg_host: str = "localhost"

T = TypeVar("T", bound=BaseConfig)

def f(config: T) -> T:
    result = copy.replace(config, pg_ssl_key="replaced")
    assert_type(result, T)
    return result

Expected Behavior

assert_type(result, T) should pass — copy.replace should return the same type as the argument.

Actual Behavior

error: Expression is of type "BaseConfig", not "T"  [assert-type]
error: Incompatible return value type (got "BaseConfig", expected "T")  [return-value]

mypy resolves _RT_co from _SupportsReplace.__replace__ return type, which for a Self-returning method resolves to the upper bound BaseConfig, not T.

Note: concrete subclass works fine — copy.replace(SubConfig(), ...) correctly returns SubConfig. Only the TypeVar-bounded case breaks.

Root Cause

typeshed PR python/typeshed#14786 changed copy.pyi from:

_SR = TypeVar("_SR", bound=_SupportsReplace)
class _SupportsReplace(Protocol):
    def __replace__(self, ...) -> Self: ...
def replace(obj: _SR, /, ...) -> _SR: ...

to:

_RT_co = TypeVar("_RT_co", covariant=True)
class _SupportsReplace(Protocol[_RT_co]):
    def __replace__(self, ...) -> _RT_co: ...
def replace(obj: _SupportsReplace[_RT_co], /, ...) -> _RT_co: ...

The bound-TypeVar approach preserved the argument type; the covariant protocol loses it because _RT_co is inferred from the protocol method return type which, for Self, resolves at the upper bound level.

Note

The motivation for python/typeshed#14786 was mypy issue #19694 which is about dataclasses.replace (not copy.replace), so the PR did not actually address its intended target.

Your Environment

  • Mypy version used: 2.2.0+dev
  • Mypy command-line flags: --python-version 3.13
  • Python version used: 3.13+

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the mypy-play reproducer and compare the two copy.pyi signatures described in the issue. Trace how the covariant _SupportsReplace protocol infers its return type for a bounded TypeVar; done means the assert_type(result, T) and return checks pass without regressing concrete subclass inference.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
devtools
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
52/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.