Skip to content

Conversation

@sterliakov
Copy link
Collaborator

Fixes #19920. The removed assert was incorrect as demonstrated by the referenced issue - using additional inference context may change overload resolution outcomes, so no reasonable assumptions can be made about the reinferred type

@sterliakov sterliakov force-pushed the bugfix/gh-19920-tuple-inference-crash branch from ca1a2fc to 1bbdc0a Compare September 24, 2025 16:44
@github-actions

This comment has been minimized.

@sterliakov
Copy link
Collaborator Author

@hauntsaninja please don't merge this one - as @ilevkivskyi suggested in the linked ticket, this might not be the right thing to do, and maybe overload selection process should ignore the outer context entirely. I tried to implement that - but the result was rather unsatisfactory (#20047), this probably needs some deeper investigation.

@sterliakov sterliakov removed the upnext label Jan 18, 2026
@sterliakov sterliakov marked this pull request as draft January 18, 2026 22:56
@hauntsaninja
Copy link
Collaborator

Ah, that comment makes sense

@github-actions
Copy link
Contributor

According to mypy_primer, this change doesn't affect type check results on a corpus of open source code. ✅

@ilevkivskyi
Copy link
Member

Btw I likely will not have time to work on this soon, sorry.

@ilevkivskyi
Copy link
Member

(This means if anyone wanted to try a solution along the lines I suggested, please go ahead! :-))

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Mypy crashes on async tuple to tuple assignment

3 participants