napi: tolerate pending VM exception in napi_create_string_* (debug/asan abort) - #36091
Closed
robobun wants to merge 3 commits into
Closed
napi: tolerate pending VM exception in napi_create_string_* (debug/asan abort)#36091robobun wants to merge 3 commits into
robobun wants to merge 3 commits into
Claude / Claude Code Review
completed
Jul 27, 2026 in 19m 27s
Code review found 1 important issue
Found 1 candidates, confirmed 1. See review comments for details.
Details
| Severity | Count |
|---|---|
| 🔴 Important | 1 |
| 🟡 Nit | 0 |
| 🟣 Pre-existing | 0 |
| Severity | File:Line | Issue |
|---|---|---|
| 🔴 Important | src/runtime/napi/napi_body.rs:659-665 |
napi_create_array{,_with_length} share the same abort path but are not fixed |
Annotations
Check failure on line 665 in src/runtime/napi/napi_body.rs
claude / Claude Code Review
napi_create_array{,_with_length} share the same abort path but are not fixed
`napi_create_array` and `napi_create_array_with_length` (napi_body.rs:530-571) share the identical abort path this PR fixes: `get_env!` (no pending-exception gate) → `JSValue::create_empty_array` → `call_zero_is_throw` → `ExceptionValidationScope::assert_exception_presence_matches(false)` → `releaseAssertNoException` → SIGABRT in debug/asan when a VM exception is already pending. Per REVIEW.md's whole-class rule, either route these through the same `SuspendExceptionScope` treatment (e.g. a `napi
Loading