[red-knot] Add initial support for * imports
#16923
Merged
Conversation
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
…`DefinitionKind` for star imports
AlexWaygood
force-pushed
the
alex/star-imports-redux
branch
from
829b35c to
d87cd7c
Compare
|
This comment was marked as resolved.
This comment was marked as resolved.
AlexWaygood
commented
Mar 22, 2025
crates/red_knot_python_semantic/src/semantic_index/definition.rs
Outdated
Show resolved
Hide resolved
AlexWaygood
force-pushed
the
alex/star-imports-redux
branch
from
9752faa to
4e6f758
Compare
I figured out the bug here and fixed it. Unfortunately, fixing the bug completely broke all our special casing for the builtin There were a huge amount of test failures in |
|
AlexWaygood
force-pushed
the
alex/star-imports-redux
branch
4 times, most recently
from
9479f00 to
550a937
Compare
carljm
approved these changes
Mar 23, 2025
crates/red_knot_python_semantic/src/semantic_index/re_exports.rs
Outdated
Show resolved
Hide resolved
crates/red_knot_python_semantic/src/semantic_index/re_exports.rs
Outdated
Show resolved
Hide resolved
MichaReiser
reviewed
Mar 24, 2025
crates/red_knot_python_semantic/src/semantic_index/re_exports.rs
Outdated
Show resolved
Hide resolved
crates/red_knot_python_semantic/src/semantic_index/re_exports.rs
Outdated
Show resolved
Hide resolved
crates/red_knot_python_semantic/src/semantic_index/re_exports.rs
Outdated
Show resolved
Hide resolved
crates/red_knot_python_semantic/src/semantic_index/re_exports.rs
Outdated
Show resolved
Hide resolved
|
(haven't finished addressing/responding to review comments yet, will continue tomorrow!) |
AlexWaygood
commented
Mar 24, 2025
Comment on lines
+1039
to
+1041
| let Some(module) = resolve_module(self.db, &module_name) else { | ||
| continue; | ||
| }; |
There was a problem hiding this comment.
I'm wondering if we should cache the resolved module in the DefinitionKind for * import definitions? It feels slightly silly to re-resolve the module name and module for these definitions in infer.rs, when we know that we wouldn't have created any definitions for the node during semantic indexing unless the module name and module were both resolvable.
There was a problem hiding this comment.
Hrm, this involves quite an annoying amount of refactoring unless we also cache the resolved ModuleName as well as the File on StarImportDefinitionKind. Leaning towards leaning this as-is for now, even though the duplication of logic between this file and infer.rs is somewhat annoying.
AlexWaygood
force-pushed
the
alex/star-imports-redux
branch
3 times, most recently
from
7a6bc73 to
92d50f1
Compare
AlexWaygood
force-pushed
the
alex/star-imports-redux
branch
from
92d50f1 to
ca5184f
Compare
AlexWaygood
commented
Mar 24, 2025
crates/red_knot_python_semantic/resources/mdtest/import/star.md
Outdated
Show resolved
Hide resolved
…rts should only have 0 or 1 definition(s) associated with them
carljm
approved these changes
Mar 24, 2025
crates/red_knot_python_semantic/resources/mdtest/import/star.md
Outdated
Show resolved
Hide resolved
AlexWaygood
added a commit
that referenced
this pull request
Mar 24, 2025
## Summary This PR separates out the entirely new tests from #16923 into a standalone PR. I'll rebase #16923 on top of this branch. The reasons for separating it out are: - It should make it clearer to see in <#16923> exactly how the functionality is changing (we can see the assertions in the tests _change_, which isn't so obvious if the tests are entirely new) - The diff on <#16923> is getting pretty big; this should reduce the diff on that PR somewhat - These tests seem useful in and of themselves, so even if we need to do a wholesale revert of <#16923> for whatever reason, it'll be nice to keep the tests ## Test Plan `cargo test -p red_knot_python_semantic`
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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
This PR adds initial support for
*imports to red-knot. The approach is to implement a standalone query, called from semantic indexing, that visits the module referenced by the*import and collects all global-scope public names that will be imported by the*import. TheSemanticIndexBuilderthen adds separate definitions for each of these names, all keyed to the sameast::Aliasnode that represents the*import.There are many pieces of
*-import semantics that are still yet to be done, even with this PR:This PR does not attempt to implement any of the semantics to do with
__all__. (If a module defines__all__, then only the symbols included in__all__are imported, not all public global-scope symbols.With the logic implemented in this PR as it currently stands, we sometimes incorrectly consider a symbol bound even though it is defined in a branch that is statically known to be dead code, e.g. (assuming the target Python version is set to 3.11):
Implementing these features is important, but is for now deferred to followup PRs.
Many thanks to @ntBre, who contributed to this PR in a pairing session on Friday!
Test Plan
Assertions in existing mdtests are adjusted, and several new ones are added.
TODO: the(I figured out the bug here and fixed it!)collections.abcintegration test at the bottom ofstar.mdis still failing. Not sure why...