fix(codegen): preserve renamed namespace constructor args

[proggeramlug]

Summary

Restore constructor argument forwarding for classes reached as new ns.VisibleAlias(...) through a renamed re-export. The namespace import now keeps the class's origin metadata under its visible scoped key, and static class resolution consumes that collision-free key instead of falling through to a function wrapper.

Changes

• Resolve renamed namespace class metadata through the defining export name in both namespace-import paths.
• Route namespace-qualified new through the scoped (namespace, member) class entry introduced by fix(compile): support full OpenCode source builds #9133.
• Preserve the existing dynamic-constructor fallback for namespace members that are functions or closure-valued variables rather than classes.

Related issue

Fixes #9285

Test plan

Run on root@perrymaster.skelpo.net with coherent compiler/runtime archives:

• ./scripts/pre-tag-check.sh --quick
• cargo build -p perry -p perry-runtime-static -p perry-stdlib-static
• PERRY_NO_AUTO_OPTIMIZE=1 RUST_TEST_THREADS=1 cargo test -p perry --test functional_batch2_regressions imported_class_namespace_reexport_uses_visible_alias --quiet
• cargo test -p perry-codegen --lib --quiet (1,378 passed; 1 ignored)
• PERRY_NO_AUTO_OPTIMIZE=1 RUST_TEST_THREADS=1 cargo test -p perry --test functional_batch2_regressions --quiet (7 passed)
• PERRY_NO_AUTO_OPTIMIZE=1 RUST_TEST_THREADS=1 cargo test -p perry --test issue_5437_cross_module_member_new_capture --quiet (1 passed)
• PERRY_NO_AUTO_OPTIMIZE=1 RUST_TEST_THREADS=1 cargo test -p perry --test module_import_forms --quiet (6 passed)

The issue fixture was also compiled with --trace llvm: before the fix it called js_new_function_construct; after the fix it directly calls model_ts__Child_constructor with the user argument.

• cargo build --release clean
• cargo test --workspace --exclude perry-ui-ios --exclude perry-ui-tvos --exclude perry-ui-watchos --exclude perry-ui-gtk4 --exclude perry-ui-android --exclude perry-ui-windows passes
• Existing regression test covers the user-visible failure
• (if CLI / stdlib / runtime API changed) Updated docs/src/
• (if touching a platform UI backend) Built -p perry-ui-<backend> locally on that platform

Screenshots / output

Before:

value: undefined
instanceof: true

After:

value: namespace
instanceof: true

Checklist

• I have NOT bumped the workspace version or edited CLAUDE.md / CHANGELOG.md (maintainer handles these at merge)
• My commits follow the loose feat: / fix: / docs: / chore: prefix convention used in the log
• I've read CONTRIBUTING.md and agree to the Code of Conduct

Summary by CodeRabbit

• Bug Fixes
  • Fixed an issue where classes accessed through renamed namespace exports could fail to receive constructor arguments correctly.
  • Constructor parameter properties and other initialization now run as expected when creating instances through a renamed namespace alias.
  • Improved class resolution for namespace imports and re-exports while preserving reliable behavior when names overlap.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: b57da9e1-a484-40ee-a75c-22b899b077cb

📥 Commits

Reviewing files that changed from the base of the PR and between 7a6d5a1 and 14773ef.

📒 Files selected for processing (5)

• changelog.d/9296-renamed-namespace-constructor.md
• crates/perry-codegen/src/expr/new_dynamic.rs
• crates/perry-codegen/src/expr/v8_interop.rs
• crates/perry-codegen/src/type_analysis/predicates.rs
• crates/perry/src/commands/compile/run_pipeline.rs

Included review availability: Your plan provides up to 8 included reviews per hour; 2 remain after this review.

---

📝 Walkthrough

Walkthrough

The compiler now preserves origin class metadata and resolves renamed namespace re-exports through collision-free registry keys, allowing constructor arguments and parameter properties to initialize correctly.

Changes

Renamed namespace class resolution

Layer / File(s)	Summary
Class metadata registration crates/perry/src/commands/compile/run_pipeline.rs, changelog.d/9296-renamed-namespace-constructor.md	Namespace-like and namespace re-export imports fall back to origin-name class metadata when the visible export name is not registered directly. The changelog records the constructor fix.
Static class lookup and ownership crates/perry-codegen/src/expr/v8_interop.rs, crates/perry-codegen/src/expr/new_dynamic.rs, crates/perry-codegen/src/type_analysis/predicates.rs	try_static_class_name returns borrowed or owned names and resolves namespace members through collision-free registry keys. Callers borrow the result where required, and dynamic construction passes the member name by reference.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Merge Risk: ⚪ Minimal · up to 14773

Renamed namespace constructors now forward arguments correctly while non-class namespace members retain their existing behavior; no actionable merge-blocking risk remains beyond normal checks and review.

Suggested reviewers: thehypnoo

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 40.00% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 5 functions across 4 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 summarizes the primary fix: preserving constructor arguments for classes accessed through renamed namespace re-exports.
Description check	✅ Passed	The description includes the required Summary, Changes, Related issue, Test plan, Screenshots/output, and Checklist sections. It documents the implementation, targeted verification, regression behavio…
Linked Issues check	✅ Passed	The changes satisfy issue #9285. They fix constructor argument forwarding for renamed namespace re-exports, preserve namespace class identity, and retain correct instanceof behavior without modifying …
Out of Scope Changes check	✅ Passed	The changes are within scope for issue #9285. The compiler updates, namespace metadata resolution, regression documentation, and targeted tests all support the reported re-export constructor bug. No u…

Full details: Description check

Explanation

The description includes the required Summary, Changes, Related issue, Test plan, Screenshots/output, and Checklist sections. It documents the implementation, targeted verification, regression behavior, and unchecked full-suite commands.

Full details: Linked Issues check

Explanation

The changes satisfy issue #9285. They fix constructor argument forwarding for renamed namespace re-exports, preserve namespace class identity, and retain correct instanceof behavior without modifying the fixture.

Full details: Out of Scope Changes check

Explanation

The changes are within scope for issue #9285. The compiler updates, namespace metadata resolution, regression documentation, and targeted tests all support the reported re-export constructor bug. No unrelated code changes are identified.

Full details: Docstring Coverage

Explanation

Docstring coverage is 40.00% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 5 functions across 4 files. (1 skipped: 1 unsupported.)

• Fix all pre-merge checks with AI

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

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]

Audited the diff — root-cause fix on both sides, and it explains the symptom exactly. I'd merge.

The reported failure was odd in a specific way: value: undefined but instanceof: true. This diff accounts for both halves.

Codegen side. try_static_class_name returned the bare property for namespace member access, relying on ctx.classes holding imported classes under their original name. That holds for export { Child } and breaks for export { Child as PublicChild }: the property is PublicChild while the class is registered as Child. Returning the collision-free (namespace, member) registry key removes the assumption rather than special-casing the rename. Widening the return to Cow<'a, str> is the minimum needed to return a composed key without forcing allocation on the common borrowed paths.

Pipeline side. The registration path had the same origin-vs-alias mismatch, and the comment names the consequence precisely:

new ns.PublicChild(arg) falls through to a function wrapper and never runs the class constructor

That is the missing half of the explanation. A function wrapper still satisfies instanceof through the class registry, which is why identity looked correct while the parameter property was never initialized — the constructor simply never ran. Falling back to the origin key here, matching what the var-classification path above already did, makes the two paths agree.

Fixing the lookup key and the registration is the right scope: fixing only one would leave the other resolving a different entity for the same source expression.

Small note

The two fallback blocks in run_pipeline.rs are near-identical get(&key).or_else(|| origin_key…) chains. If a third site ever needs the same resolution, it is worth naming — resolve_exported_class(&exported_classes, &key, &origin_key_under_origin_name) or similar — so the origin-vs-alias rule lives in one place. Not worth churning this PR for; a release blocker should land.