python / python/mypy

TypedDict 'in' narrowing w/o @final

Open
#15,697 4 comments 4 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

feature topic-final topic-type-narrowing topic-typed-dict
Dominant language
Python
Stars
20.6k
Forks
3.3k
PR merge metrics
PR metrics pending

Description

We shouldn't require @final decoration for TypedDicts to narrow them based on the 'in' operator.

Why?

Basically @erictraut's comment.

In #13838, we've added "key in Union[TypedDict, ...]" narrowing for TypedDicts that are marked @final. The reason was to prevent this:

class Mammal(TypedDict):
  mammary_glands: int

class Bird(TypedDict):
  eggs: int

class Echidna(Mammal):
  eggs: int

animal: Mammal | Bird
if 'eggs' in animal:
  assert_type(animal, Bird)  # WRONG! Could still be a Mammal (a Echidna)
if 'eggs' in animal and 'mammary_glands' in animal:
  assert_never(animal)  # WRONG! Could still be a Mammal (a Echidna)

However, per @erictraut's comment, due to TypedDict being a structural type, we shouldn't consider the class hierarchy when type-matching.

This will be consistent with pyright and TypeScript (Playground Link).

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 files or tests; start by reviewing the existing key in Union[TypedDict, ...] narrowing added in #13838 and reproduce the examples above. Define done as allowing narrowing without @final while treating TypedDicts structurally, then add regression coverage for the shown cases and verify the incorrect assertions are rejected.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
devtools
Issue type
Feature
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.