python / python/mypy

__new__ on abstract classes produces false positive "Cannot instantiate abstract class"

Open
#14,106 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug
Dominant language
Python
Stars
20.6k
Forks
3.3k
PR merge metrics
PR metrics pending

Description

Bug Report

When an abstract class defines __new__ (note, not __init__) method, mypy produces error when that class is instantiated, even though __new__() does not imply instantiation.

To Reproduce

from abc import ABCMeta, abstractmethod


class Abstract(metaclass=ABCMeta):
    def __new__(cls) -> "Abstract":
        return super().__new__(Concrete)

    @abstractmethod
    def hello(self) -> int:
        ...


class Concrete(Abstract):
    def hello(self) -> int:
        return 0


x = Abstract()
print(x.hello())

Expected Behavior

No errors reported when checked with mypy. Calling a class with __new__ method shuld be effectively treated as just calling a simple method, like

def make_abstract() -> "Abstract":
    return Concrete()

Actual Behavior

error message is printed:

 error: Cannot instantiate abstract class "Abstract" with abstract attribute "hello"  [abstract]

Your Environment

  • Mypy version used: mypy-0.991
  • Mypy command-line flags: none
  • Mypy configuration options from mypy.ini (and other config files): none
  • Python version used: 3.10.6

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

No repository file or test is named in the issue. Start with the supplied Python reproducer and trace mypy's abstract-class instantiation check when new is defined; done means the false [abstract] error is absent while the intended abstract-class diagnostic remains covered by a regression test.

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
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.