python / python/mypy

Rust style force handling of return value

Open
#6,936 27 comments 18 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

feature priority-2-low
Dominant language
Python
Stars
20.6k
Forks
3.3k
PR merge metrics
PR metrics pending

Description

I tried to implement something like Rust's Result type. The interesting thing is that it has a linting attribute called #[must_use] which makes the linter enforce that the value is not ignored when used as a return type.

I think this could be very helpful when implementing functions that returns status and you want to ensure that it is not unhandled.

Consider the following toy implementation:

from typing import cast, Generic, Optional, TypeVar, Union
from whereever import MustUse

T = TypeVar("T")
E = TypeVar("E")

# We inherit from MustUse to ensure this type isn't ignore
class Result(Generic[T, E], MustUse):
    def __init__(self, value: Union[T, E], is_ok: bool):
        self._value = value
        self.is_ok = is_ok

    @classmethod
    def ok(cls, value: T) -> "Result[T, E]":
        return cls(value, is_ok=True)

    @classmethod
    def error(cls, value: E) -> "Result[T, E]":
        return cls(value, is_ok=True)

    @property
    def is_error(self) -> bool:
        return not self.is_ok

    def get_value(self) -> Optional[T]:
        if self.is_ok:
            return cast(T, self._value)
        return None

    def get_error(self) -> Optional[E]:
        if not self.is_ok:
            return cast(E, self._value)
        return None

def f(success: bool) -> Result[str, int]:
    if not success:
        return Result[str, int].error(123)
    return Result[str, int].ok("Hello")


# No warning is raised since the value is saved to a variable
a = f(True)

# No warning is raised since we convert the Result into an Optional
f(False).get_value()

# This will raise a warning since we never do anything with the Result like:
#     error: unused return value with attribute MustUse
# or
#     error: return value of function call must be used
f(True)

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

The issue names no mypy files, tests, or entry points. Start by reviewing mypy's type-checking and linting architecture, then define how a Result-like return value is distinguished from an intentionally consumed value; done means ignored status returns are diagnosed without warning for assigned or explicitly consumed results.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, rust
Domain
tooling
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
30/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.