fix(hir): preserve computed field key scope

[proggeramlug]

Closes #10889

Root cause

Class capture synthesis treated computed instance-field key expressions as constructor-time work. In nested and CommonJS-wrapped classes it remapped those expressions to synthetic constructor capture parameters, but ClassDefinitionEvaluation executes the key snapshots outside the constructor, where those parameters are unbound.

In undici's PoolBase, [kClients] = [] was consequently installed under the wrong key. Pool later read the real kClients symbol and failed at this[kClients].length (pool.js:101).

Fix

• keep computed field-key expressions in their class-definition scope
• exclude definition-time keys from constructor capture synthesis
• retain constructor capture rewriting for field initializer values, which do run during construction
• add a regression where an unrelated method capture still synthesizes a constructor without capturing/remapping the computed key

Validation

• cargo test -p perry-hir lower::tests --lib (102 passed, 1 ignored)
• python3 scripts/check_test_registration.py
• git diff --check
• symbolized real undici@8.9.0 public-entry fixture: the reported Pool[kClients] failure at pool.js:101 is gone

The fixture advances to a separate Node stream contract failure: Socket.read() returns undefined instead of null, causing a later chunk.length failure at client-h1.js:323. That next blocker is tracked independently in #10908.

Summary by CodeRabbit

• Bug Fixes

  • Fixed computed instance-field keys in nested and CommonJS-wrapped classes so they are resolved correctly when the class is defined.
  • Symbol-keyed fields now retain their resolved keys when creating class instances, preventing initialization failures in affected request-handling scenarios.

• Tests

  • Added coverage to ensure computed field keys are not incorrectly treated as constructor captures.

[coderabbitai]

Understand this PR’s impact

Explore downstream dependencies and potential security impact with Blast Radius.

View blast radius →

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Advanced

Run ID: b6db2441-4ada-48ec-920f-aa1cf365bdd9

📥 Commits

Reviewing files that changed from the base of the PR and between 841b605 and 7a0d63a.

📒 Files selected for processing (4)

• changelog.d/10889-computed-field-key-captures.md
• crates/perry-hir/src/ir/decl.rs
• crates/perry-hir/src/lower/tests.rs
• crates/perry-hir/src/lower_decl/class_captures.rs

Included review availability: Your plan provides up to 8 included reviews per hour; 6 remain after this review.

---

📝 Walkthrough

Walkthrough

The change prevents computed instance-field keys from being collected or rewritten as constructor captures. The HIR documentation describes definition-time key storage. A regression test verifies capture behavior for nested classes with symbol-keyed fields.

Changes

Computed Field Capture Handling

Layer / File(s)	Summary
Computed key evaluation contract crates/perry-hir/src/ir/decl.rs, crates/perry-hir/src/lower_decl/class_captures.rs	The HIR documentation states that computed keys resolve during class definition evaluation. Capture analysis no longer collects references from computed keys.
Constructor rewrite and regression coverage crates/perry-hir/src/lower_decl/class_captures.rs, crates/perry-hir/src/lower/tests.rs, changelog.d/10889-computed-field-key-captures.md	Constructor rewriting remaps only field initializers. The regression test verifies that a computed key does not create a constructor capture while an unrelated method capture remains. The changelog records the fix.

Priority: ➖ Normal

Estimated code review effort: 2 (Simple) | ~15 minutes

Change: Bug fix · Severity of issue fixed: Medium

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 33.33% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 3 functions across 3 files. (1 skipped: 1…	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and concisely describes the primary change: preserving computed field key scope during HIR lowering.
Description check	✅ Passed	The description provides the root cause, fix, related issue, validation commands, regression-test coverage, and the remaining independent fixture failure. It does not use every template heading or che…
Linked Issues check	✅ Passed	The PR addresses #10889's reported computed-instance-field failure. synthesize_class_captures no longer collects or constructor-remaps key_expr, while field initializer expressions still receive c…
Out of Scope Changes check	✅ Passed	The changes stay within #10889. The implementation changes computed-field capture synthesis, the test covers the regression, the IR comment documents the corrected scope, and the changelog records the…

Full details: Docstring Coverage

Explanation

Docstring coverage is 33.33% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 3 functions across 3 files. (1 skipped: 1 unsupported.)

• Fix all pre-merge checks with AI

✨ Finishing Touches 💡 2

📝 Generate docstrings 💡

• Commit to this branch
• Create a new PR

🧪 Generate unit tests (beta)

• Commit to this branch
• Create a new PR

🛠️ Fix failing CI checks 💡

• Commit to this branch
• Create a new PR

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[proggeramlug]

Landed via merge train 253 (#10918) as v0.5.1633 — merge commit 0fa3915293.

Expedited at the owner's request. Carried with nine other PRs; the stacked ones (#10899/#10900 on #10886, #10901 on #10885) had only their unique commits taken.

Evidence on the assembled tree: perry-runtime full suite 4233 passed / 0 failed / 0 SIGABRT, perry-hir 471 / 0, cargo check --workspace --all-targets under -D warnings clean, cargo fmt --check clean, and all eight ratchets rc=0. The gap sweep and compiler-output suites were not run.

Closing here rather than merging — a train lands the commits directly.