fix(codegen): gate String-method dispatch on string receiver — user method like internals.trim(v,s) (#5271)#5272
Open
proggeramlug wants to merge 1 commit into
Open
Conversation
internals.trim(value, schema) on a non-string receiver was mis-lowered to String.prototype.trim and failed compile on arity. Thread receiver-type info so String-method fast-path only fires for string receivers; object/any route to dynamic method dispatch.
📝 WalkthroughWalkthroughFixes issue ChangesString.prototype fast-path fix for non-string receivers
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Possibly related issues
Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
crates/perry-codegen/src/stmt/let_stmt.rs (1)
66-79:⚠️ Potential issue | 🟠 Major | ⚡ Quick winObject-literal receiver tracking is currently too narrow.
ctx.object_literal_localsis only updated inside the!mutablebranch and only for direct object-literal-shaped initializers. That misses common shapes likelet o = { ... }and aliasing (const b = a), so a laterb.trim()/o.split(...)can still be claimed by the String fast-path when arity matches.Suggested patch shape
- if !mutable { - if let Some(init_expr) = init { - ... - if is_object_literal_init(init_expr) { - ctx.object_literal_locals.insert(id); - } - } - } + if let Some(init_expr) = init { + if !mutable { + if let Some(props) = crate::lower_call::extract_options_fields(ctx, init_expr) { + ctx.option_object_locals.insert(id, props); + } + } + if is_object_literal_init(init_expr) { + ctx.object_literal_locals.insert(id); + } else if let perry_hir::Expr::LocalGet(src_id) = init_expr { + if ctx.object_literal_locals.contains(src_id) { + ctx.object_literal_locals.insert(id); + } + } + }🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@crates/perry-codegen/src/stmt/let_stmt.rs` around lines 66 - 79, The object-literal receiver tracking in the let_stmt.rs file is too restrictive because it only updates ctx.object_literal_locals inside the !mutable branch. This misses cases where mutable bindings (like let o = { ... }) should also track object-literal locals, causing later method calls on those variables to be incorrectly claimed by the String fast-path. Move the object-literal tracking logic (the is_object_literal_init check and ctx.object_literal_locals.insert call) outside of the !mutable condition so that both mutable and immutable variable declarations with object-literal initializers are properly tracked, ensuring that method calls like o.trim() or b.split() resolve to the object's own method rather than the static String-method fast path.
🧹 Nitpick comments (1)
crates/perry/tests/issue_5271_string_method_nonstring_receiver.rs (1)
78-133: ⚡ Quick winAdd regressions for mutable and inline object-literal receivers.
Current tests are strong, but they don’t pin
let o = { ... }; o.trim()or inline method-bearing object literals (IIFE-lowered shape), which are the remaining collision-prone receiver forms.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@crates/perry/tests/issue_5271_string_method_nonstring_receiver.rs` around lines 78 - 133, Add two new regression test functions following the same pattern as the existing tests in this file. First, add a test for mutable object-literal receivers (using let o = { trim() { ... } }; o.trim() style) to verify that user-defined methods on mutable object variables resolve correctly and don't incorrectly dispatch to String builtins. Second, add a test for inline object-literal receivers passed directly or in IIFE-lowered forms (like ({ trim() { ... } }).trim()) to verify the same collision-avoidance behavior. Both tests should use compile_and_run and assert that user-defined methods are called, not the String builtins, to ensure these edge-case receiver forms don't regress.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@crates/perry-codegen/src/lower_call/property_get.rs`:
- Around line 459-466: The receiver_is_object_literal check in the condition
block is incomplete because it only detects LocalGet references to object
literal locals and direct Object expressions, but misses IIFE-lowered object
literals. Extend the receiver_is_object_literal guard to also detect the
object-building IIFE pattern (a Call expression with a Function body that
returns an object). This will prevent incorrectly routing such patterns to the
String fast-path when they should not be treated as simple object literals.
---
Outside diff comments:
In `@crates/perry-codegen/src/stmt/let_stmt.rs`:
- Around line 66-79: The object-literal receiver tracking in the let_stmt.rs
file is too restrictive because it only updates ctx.object_literal_locals inside
the !mutable branch. This misses cases where mutable bindings (like let o = {
... }) should also track object-literal locals, causing later method calls on
those variables to be incorrectly claimed by the String fast-path. Move the
object-literal tracking logic (the is_object_literal_init check and
ctx.object_literal_locals.insert call) outside of the !mutable condition so that
both mutable and immutable variable declarations with object-literal
initializers are properly tracked, ensuring that method calls like o.trim() or
b.split() resolve to the object's own method rather than the static
String-method fast path.
---
Nitpick comments:
In `@crates/perry/tests/issue_5271_string_method_nonstring_receiver.rs`:
- Around line 78-133: Add two new regression test functions following the same
pattern as the existing tests in this file. First, add a test for mutable
object-literal receivers (using let o = { trim() { ... } }; o.trim() style) to
verify that user-defined methods on mutable object variables resolve correctly
and don't incorrectly dispatch to String builtins. Second, add a test for inline
object-literal receivers passed directly or in IIFE-lowered forms (like ({
trim() { ... } }).trim()) to verify the same collision-avoidance behavior. Both
tests should use compile_and_run and assert that user-defined methods are
called, not the String builtins, to ensure these edge-case receiver forms don't
regress.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro Plus
Run ID: c10c3d9a-e786-4745-a3fc-47f287b8c1db
📒 Files selected for processing (8)
crates/perry-codegen/src/codegen/closure.rscrates/perry-codegen/src/codegen/entry.rscrates/perry-codegen/src/codegen/function.rscrates/perry-codegen/src/codegen/method.rscrates/perry-codegen/src/expr/mod.rscrates/perry-codegen/src/lower_call/property_get.rscrates/perry-codegen/src/stmt/let_stmt.rscrates/perry/tests/issue_5271_string_method_nonstring_receiver.rs
Comment on lines
+459
to
466
| let receiver_is_object_literal = matches!( | ||
| &**object, | ||
| Expr::LocalGet(id) if ctx.object_literal_locals.contains(id) | ||
| ) || matches!(&**object, Expr::Object(_)); | ||
| if is_string_only_method | ||
| && string_only_method_arity_ok(property, args.len()) | ||
| && !receiver_is_object_literal | ||
| && !is_array_expr(ctx, object) |
There was a problem hiding this comment.
Receiver-shape guard misses IIFE-lowered object literals.
receiver_is_object_literal does not include the object-building IIFE shape, so inline method-bearing literals can still hit the String fast-path for matching-arity builtin names.
Suggested guard extension
let receiver_is_object_literal = matches!(
&**object,
Expr::LocalGet(id) if ctx.object_literal_locals.contains(id)
-) || matches!(&**object, Expr::Object(_));
+) || matches!(&**object, Expr::Object(_))
+ || matches!(
+ &**object,
+ Expr::Call { callee, args, .. }
+ if matches!(args.first(), Some(Expr::Object(_)))
+ && matches!(
+ callee.as_ref(),
+ Expr::Closure { params, .. }
+ if params.first().map_or(false, |p| p.name == "__perry_obj_iife")
+ )
+ );📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| let receiver_is_object_literal = matches!( | |
| &**object, | |
| Expr::LocalGet(id) if ctx.object_literal_locals.contains(id) | |
| ) || matches!(&**object, Expr::Object(_)); | |
| if is_string_only_method | |
| && string_only_method_arity_ok(property, args.len()) | |
| && !receiver_is_object_literal | |
| && !is_array_expr(ctx, object) | |
| let receiver_is_object_literal = matches!( | |
| &**object, | |
| Expr::LocalGet(id) if ctx.object_literal_locals.contains(id) | |
| ) || matches!(&**object, Expr::Object(_)) | |
| || matches!( | |
| &**object, | |
| Expr::Call { callee, args, .. } | |
| if matches!(args.first(), Some(Expr::Object(_))) | |
| && matches!( | |
| callee.as_ref(), | |
| Expr::Closure { params, .. } | |
| if params.first().map_or(false, |p| p.name == "__perry_obj_iife") | |
| ) | |
| ); | |
| if is_string_only_method | |
| && string_only_method_arity_ok(property, args.len()) | |
| && !receiver_is_object_literal | |
| && !is_array_expr(ctx, object) |
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@crates/perry-codegen/src/lower_call/property_get.rs` around lines 459 - 466,
The receiver_is_object_literal check in the condition block is incomplete
because it only detects LocalGet references to object literal locals and direct
Object expressions, but misses IIFE-lowered object literals. Extend the
receiver_is_object_literal guard to also detect the object-building IIFE pattern
(a Call expression with a Function body that returns an object). This will
prevent incorrectly routing such patterns to the String fast-path when they
should not be treated as simple object literals.
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
1 participant
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.
Fixes #5271.
internals.trim(value, schema)(a user method on a non-string receiver) was mis-lowered toString.prototype.trimand failed the compile on arity (String.trim takes no args, got 2), blocking joi.Fix: added an arity-aware gate (
string_only_method_arity_ok) inlower_call/property_get.rs— a builtin-named method on a receiver that is NOT provably a string, whose arg count can't match the String builtin's signature, falls through to normal dynamic method dispatch instead of forcingString.prototype.<m>. Genuine string receivers still fast-path.Verified:
internals.trim("a","b")→a:b; string regressions (" hi ".trim(),split,toLowerCase,repeat,padStart) match Node; new testissue_5271_string_method_nonstring_receiver. joi's String.trim error is gone (it now advances to a separate regex-incompatibility wall — Rustregexcrate lacks lookahead).No version/CHANGELOG/Cargo.lock edits (for maintainer to fold in at merge).
Summary by CodeRabbit
String.prototypebuiltins (e.g.,trim,split,slice). User-defined methods are now correctly called instead of incorrectly attempting String method dispatch on non-string receivers.