Rust: Implement type inference for closures and calls to closures #20130
+479
−78
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
There was a problem hiding this comment.
Pull Request Overview
This PR implements type inference for closure expressions and calls to closures in Rust, including proper handling of FnOnce trait bounds. The implementation reuses tuple types for function parameters and represents closures as dyn FnOnce types.
- Adds type inference support for closure expressions, their parameters, and return types
- Implements type propagation for calls to closures and higher-order functions
- Introduces handling of the
FnOncetrait with syntactic sugar for function types
Reviewed Changes
Copilot reviewed 6 out of 6 changed files in this pull request and generated 2 comments.
Show a summary per file
| File | Description |
|---|---|
| closure.rs | New test file with comprehensive closure type inference test cases |
| main.rs | Removes old closure test module and adds reference to new test file |
| TypeMention.qll | Adds support for parenthesized argument lists and FnOnce trait handling |
| TypeInference.qll | Implements core closure type inference logic and call expression handling |
| Stdlib.qll | Defines the FnOnce trait class for type system integration |
| type-inference.expected | Expected test output showing inferred types for closure expressions |
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Outdated suggestions cannot be applied.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.
This PR implements type inference for closure expressions, calls to closures, and correctly handles
FnOncetrait bounds.A few notes on the implementation and the PR:
FnOncetrait has a type parameterArgswhich is a tuple of all the arguments. That is reflected in our implementation as well, i.e., we re-use our tuple types for parameters.FnOnce,FnMut, andFntraits. They form a trait hierarchy withFnOnceat the top.FnOnceuses an associated type for the return type and since we don't support associated types inherited in subtraits (tests for this was added in Rust: Fix type inference for trait objects for traits with associated types #20122) we can right now only really understand theFnOncetrait and not the others.dyn FnOncetype. That is, instead of creating some new type for closures we just reusedyn. The Rust language itself says that closures have some internal type that can't be written by users, and all that known about it is that it implements some of theFnOnce/Fn/FnMuttraits.dyngives us this effect. rust-analyzer shows closures as having inimpl FnOncetype. I think that makes a bit more sense thandyn, but usingdynwas simpler.Fntrait for closures, as it is a subtrait of the others, but, again, we can't understandFnyet.DCA seems fine, though we only get a 0.04% point increase in resolved calls, which is less than I had hoped.