[pyupgrade] Allow shadowing non-builtin bindings (UP029)#22749
Merged
Conversation
Summary
--
I thought the fix unsafety example in the rule docs looked a bit suspicious
while I was going through more potential default rules today:
```py
def str(x):
return x
from builtins import str
str(1) # `"1"` with the import, `1` without
```
Changing the behavior in this way seemed to go beyond fix unsafety and into bug
territory. Sure enough, there was an existing bug report in #16182.
This PR fixes #16182 (and the fix safety example) by checking that the builtin
import that the rule flags is actually shadowing a builtin binding. I also left
an exception for `from builtins import *`, but we could consider ignoring that
case too.
I initially tried reusing `SemanticModel::resolve_name` and `only_binding`, but
they are specific to `ExprName`s because that seems to be all that is inserted
into the `SemanticModel::resolved_names` map.
Test Plan
--
New tests based on #16182 and the fix safety docs
|
MichaReiser
approved these changes
Jan 20, 2026
crates/ruff_linter/src/rules/pyupgrade/rules/unnecessary_builtin_import.rs
Outdated
Show resolved
Hide resolved
carljm
added a commit
that referenced
this pull request
Jan 20, 2026
* main: (149 commits) Preserve required parentheses in lambda bodies (#22747) [`flake8-simplify`] Make fix unsafe if it deletes comments (`SIM911`) (#22661) [`refurb`] Make fix unsafe if it deletes comments (`FURB145`) (#22670) [`ruff`] Make fix unsafe if it deletes comments (`RUF020`) (#22664) [ty] Add README for `ty_completion_bench` CLI tool [ty] Update completion eval [ty] Collect completions into a max-heap [ty] Truncate imports and qualifications derived from completions [ty] Get rid of high-level completion deduplication [ty] Refactor how we sort completions [ty] Optimize collection of "all symbols" [ty] Speed up completions by tweaking sorts [ty] Add new `ty_completion_bench` for ad hoc benchmarking gitignore: ignore scratch directories in completion eval truth directory [`pyupgrade`] Allow shadowing non-builtin bindings (`UP029`) (#22749) [ty] Emit invalid type form for stringified annotations (#22752) [ty] Allow `if type(x) is Y` narrowing for types other than class-literal types (#22729) [ty] Add basic support for overloads in `ParamSpec` (#21946) [`ruff`] Make fix unsafe if it deletes comments (`RUF019`) (#22663) [`flake8-bugbear`] Make fix unsafe if it deletes comments (`B014`) (#22659) ...
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
I thought the fix unsafety example in the rule docs looked a bit suspicious
while I was going through more potential default rules today (playground):
Changing the behavior in this way seemed to go beyond fix unsafety and into bug
territory. Sure enough, there was an existing bug report in #16182.
This PR fixes #16182 (and the fix safety example) by checking that the builtin
import that the rule flags is actually shadowing a builtin binding. I also left
an exception for
from builtins import *, but we could consider ignoring thatcase too.
I initially tried reusing
SemanticModel::resolve_nameandonly_binding, butthey are specific to
ExprNames because that seems to be all that is insertedinto the
SemanticModel::resolved_namesmap.Test Plan
New tests based on #16182 and the fix safety docs