Skip to content

Remove dead code from react_compiler and the node:http2 frame parser - #37272

Closed
robobun wants to merge 9 commits into
mainfrom
claude/farm/d4de2af7/dead-code-react-compiler-h2
Closed

Remove dead code from react_compiler and the node:http2 frame parser#37272
robobun wants to merge 9 commits into
mainfrom
claude/farm/d4de2af7/dead-code-react-compiler-h2

Move the removed _exp pragmas to the harness's ignored bucket

df39aa2
Select commit
Loading
Failed to load commit list.
Claude / Claude Code Review completed Aug 9, 2026 in 21m 25s

Code review found 2 potential issues

Found 5 candidates, confirmed 2. See review comments for details.

Details

Severity Count
🔴 Important 0
🟡 Nit 2
🟣 Pre-existing 0
Severity File:Line Issue
🟡 Nit src/react_compiler/hir/mod.rs:71 hir::is_use_state_type is now zero-caller (only callers were in the removed _exp validation)
🟡 Nit src/runtime/api/bun/h2_frame_parser.rs:392-397 More h2_frame_parser dead-code survivors: stale write() doc + wire-layout artifacts, stale 'legacy frame state being ret

Annotations

Check warning on line 71 in src/react_compiler/hir/mod.rs

See this annotation in the file changed.

@claude claude / Claude Code Review

hir::is_use_state_type is now zero-caller (only callers were in the removed _exp validation)

`pub fn is_use_state_type` at hir/mod.rs:1722 is now zero-caller — its only two call sites (and its import) were inside `record_instruction_derivations`, one of the 22 exclusive helpers of the removed `_exp` validation. Same `pub`-item escape hatch as the other survivors already swept in 66cce130/705bdeb4; hir/mod.rs is already touched by this PR (HirBox), and the guard test already scans it, so this is a 3-line delete plus `["src/react_compiler/hir/mod.rs", /\bis_use_state_type\b/]` in the reac

Check warning on line 397 in src/runtime/api/bun/h2_frame_parser.rs

See this annotation in the file changed.

@claude claude / Claude Code Review

More h2_frame_parser dead-code survivors: stale write() doc + wire-layout artifacts, stale 'legacy frame state being retired' comment, orphaned Error::UnableToDecode

A few more survivors of the same dead-code-blind-spot class already swept in deb52418f0/66cce130a7/705bdeb48b: the `FullSettingsPayload` doc at L396-398 references "the per-field swap_bytes() in write()" (removed here) — and with `write()` and the bytemuck impls gone the seven `_*_type: u16` fields, `#[repr(C, packed)]`, `BYTE_SIZE`, and the const assert have no remaining byte-layout consumer (the earlier review's "independently guards `serialize_local_settings`" rationale named a nonexistent fu