python / python/mypy

Poor inference of dict literal when dict type is unioned with certain types

Open
#15,217 0 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

bug topic-join-v-union
Dominant language
Python
Stars
20.6k
Forks
3.3k
PR merge metrics
PR metrics pending

Description

Bug Report

To Reproduce

https://mypy-play.net/?mypy=master&python=3.11&gist=5cf1ab002a6dc63c5013fa0f6a00ba4d

Expected Behavior
I expect the dict literal to be inferred as dict[str, str |int] when passed to both foo and bar

Actual Behavior
It's inferred as dict[str, object] when passed to foo, which causes a false positive.

Note that the inference depends on the type you union the dict with. E.g. if you replace Iterable[tuple[str] with list[tuple[str]] you get the expected inference.

Your Environment

  • Mypy version used: Tried 1.0.0, 1.0.1, 1.1.1, 1.2.0 and master in the playground, all have the same behavior
  • Mypy command-line flags: See playground
  • Mypy configuration options from mypy.ini (and other config files): See playground
  • Python version used: 3.11

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 linked mypy playground example using Python 3.11 and the reported versions to reproduce the differing inference for foo and bar. Trace the dict-literal inference behavior for unions involving Iterable[tuple[str]] and list[tuple[str]]. Done means the dict is inferred as dict[str, str | int] in both calls and a regression test covers the case.

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
Needs clarification
Newbie friendliness
38/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.