fix(oxc-runtime): route require() to CJS helper variant (#9263) by IWANABETHATGUY · Pull Request #9526 · rolldown/rolldown

IWANABETHATGUY · 2026-05-23T15:10:10Z

The on-disk @oxc-project/runtime package ships two variants of every helper: a CJS form at src/helpers/<name>.js that sets module.exports = fn, and an ESM form at src/helpers/esm/<name>.js that does export default fn. The plugin previously embedded only the ESM variant and reused it for every caller, which meant require("@oxc-project/runtime/helpers/X") returned the ESM namespace { default: fn } rather than the function itself.

This PR embeds both variants and dispatches resolution by ImportKind: require() gets the CJS variant (callable directly), import / dynamic import() get the ESM variant. Relative imports between virtual helpers preserve the importer's directory so an ESM helper's ./typeof.js stays in esm/. Fixes #9263, where _defineProperty resolved to the ESM namespace inside CJS-classified modules and threw TypeError: _defineProperty is not a function at init.

IWANABETHATGUY · 2026-05-23T15:10:24Z

fix(oxc-runtime): route require() to CJS helper variant (#9263) #9526 👈 (View in Graphite)
main

How to use the Graphite Merge Queue

Add the label graphite: merge-when-ready to this PR to add it to the merge queue.

You must have a Graphite account in order to use the merge queue. Sign up using this link.

_{An organization admin has enabled the Graphite Merge Queue in this repository.} _{Please do not merge from GitHub as this will restart CI on PRs being processed by the merge queue.}

This stack of pull requests is managed by Graphite. Learn more about stacking.

netlify · 2026-05-23T15:10:40Z

✅ Deploy Preview for rolldown-rs canceled.

Name	Link
🔨 Latest commit	`f06c2a0`
🔍 Latest deploy log	https://app.netlify.com/projects/rolldown-rs/deploys/6a13a7b421074300089b47cb

codspeed-hq · 2026-05-23T16:54:29Z

Merging this PR will not alter performance

✅ 4 untouched benchmarks
⏩ 10 skipped benchmarks¹

_{Comparing 05-09-fix_9263 (8d4e742) with main (d82d2ff)}

10 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

hyf0 · 2026-05-24T02:08:31Z

I'm not sure this is a good direction. I'm worred that this might cause the runtime code have dual package hazard.

What will happen if the runtime code got both import and require?

This PR embeds both variants and dispatches resolution by ImportKind: require() gets the CJS variant (callable directly), import / dynamic import() get the ESM variant. Relative imports between virtual helpers preserve the importer's directory so an ESM helper's ./typeof.js stays in esm/.

From the PR description, I assume the anwser is that rolldown will ship both esm and cjs version code.

I think we need to refer to other bundlers and see how does they deal with the situation. @sapphi-red cc

IWANABETHATGUY · 2026-05-24T03:16:06Z

I'm not sure this is a good direction. I'm worred that this might cause the runtime code have dual package hazard.

What will happen if the runtime code got both import and require?

This PR embeds both variants and dispatches resolution by ImportKind: require() gets the CJS variant (callable directly), import / dynamic import() get the ESM variant. Relative imports between virtual helpers preserve the importer's directory so an ESM helper's ./typeof.js stays in esm/.

From the PR description, I assume the anwser is that rolldown will ship both esm and cjs version code.

I think we need to refer to other bundlers and see how does they deal with the situation. @sapphi-red cc

1. Almost all helpers are pure stateless functions

I scanned all 116 ESM helpers. The overwhelming majority are pure functions like:

function _defineProperty(e, r, t) { /* …pure… */ }
function _typeof(o) { /* …pure… */ }

For those, "two copies in one bundle" just means two equivalent function objects — no helper is compared by identity, only called, so duplication is at worst a few wasted bytes.

The exceptions worth naming explicitly:

Helper	Why it isn't strictly pure	Risk if duplicated
`OverloadYield`	Constructor used as a brand. `wrapAsyncGenerator` does `o instanceof OverloadYield` to detect values produced by `awaitAsyncGenerator` / `asyncGeneratorDelegate`.	The `instanceof` check could miss values if `wrapAsyncGenerator` and `awaitAsyncGenerator` resolve to different variants and an async generator from one classification delegates to one from the other.
`AwaitValue`	Constructor (`this.wrapped = t`). No other helper checks `instanceof AwaitValue`, but user-emitted regenerator code might.	Same shape as `OverloadYield`, but only matters if user code does cross-variant identity checks on the wrapped value.
`wrapNativeSuper`	Holds a closure-local `Map` cache mapping native ctor → `Wrapper`.	Two variants → two separate caches → wrapping the same native twice yields two `Wrapper`s, so `obj instanceof Wrapper` across variants would fail. Almost never reached in practice — each transformed `extends NativeCtor` site goes through one specific call.
`classPrivateFieldLooseKey`	Module-level `var id = 0` counter producing `__private_<id>_<name>` keys.	Each duplicate copy restarts at 0, so the IDs are no longer globally unique. Harmless in practice because keys are scoped to the class instance they're written on.

Within a single user module, the helper and its transitive ./Foo.js imports go through the same ImportKind (the PR's relative-import resolver preserves the importer's directory — esm/wrapAsyncGenerator.js resolves ./OverloadYield.js to esm/OverloadYield.js, and the CJS sibling stays in CJS). So the only way to trip the OverloadYield brand check is to pass a value produced by helpers from one variant into helpers from the other variant — i.e. async-generator delegation across the CJS/ESM module boundary. Niche, but worth flagging.

2. The on-disk package already does exactly this dispatch

Look at @oxc-project/runtime's own package.json exports:

"./helpers/defineProperty": [
  {
    "node":    "./src/helpers/defineProperty.js",       // CJS variant
    "import":  "./src/helpers/esm/defineProperty.js",   // ESM variant
    "default": "./src/helpers/defineProperty.js"
  },
  "./src/helpers/defineProperty.js"
]

So if oxc-resolver (or Node, or webpack) were resolving @oxc-project/runtime/helpers/defineProperty against the real package, require() would already give the CJS file and import would already give the ESM file. The PR isn't inventing a new dual-shipping policy — it's making the embedded plugin honor the policy the upstream package already encodes via exports conditions. Until now we were always returning the ESM variant regardless of caller, which is what causes #9263.

3. The CJS variant is interop-safe by construction

The on-disk CJS helper ends with:

module.exports = _defineProperty;
module.exports.__esModule = true;
module.exports["default"] = module.exports;

So require("…"), require("…").default, and import X from "…" all yield the function. Two copies of this in one bundle still both behave as fn. There's no shape divergence to expose.

4. Other bundlers in this niche behave the same way

@babel/runtime ships the identical dual-export shape (it's where this package's structure comes from), and webpack/Rollup pick the variant via exports conditions on a per-caller basis — same outcome as this PR. @swc/helpers does the same. So even if a real app reaches one helper from both import and require() callers, it'd already end up with two copies under those bundlers.

5. The alternatives are worse

The other ways to fix #9263 are:

(a) Keep ESM-only and fix the CJS classification adapter so var X = (init_X(), __toCommonJS(X_exports)) binds to the function instead of the namespace. This is doable but tangled — the helper is a virtual \0… module, the CJS-emitter path needs to special-case it, and we'd be growing logic in the linking/codegen stages rather than where the helper lives.
(b) Always emit the CJS variant. Then every ESM helper import in user code becomes a __toESM(require_…) shape, which the rest of tree-shaking and chunking treats as a "CJS classified" module. Helpers would no longer be statically-inlinable function decls. That regresses ESM output noticeably.
(c) This PR. Match what the package's own exports map says. Matches Node and matches what every other bundler does when resolving this package off disk.

IWANABETHATGUY · 2026-05-24T03:23:07Z

Another reason I don't prefer the way Keep ESM-only and fix the CJS classification adapter is written in https://github.com/IWANABETHATGUY-reproduction/9263-comparison/blob/main/README.md

hyf0

Webpack + babel runtime has the same behavior as this PR. LGTM.

IWANABETHATGUY · 2026-05-25T01:33:05Z

Merge activity

May 25, 1:33 AM UTC: The merge label 'graphite: merge-when-ready' was detected. This PR will be added to the Graphite merge queue once it meets the requirements.
May 25, 1:36 AM UTC: IWANABETHATGUY added this pull request to the Graphite merge queue.
May 25, 1:41 AM UTC: Merged by the Graphite merge queue.

IWANABETHATGUY · 2026-05-25T01:35:40Z

I have fixed the issues(refactor) that @shulaoda mentioned, so I merged.

The on-disk `@oxc-project/runtime` package ships two variants of every helper: a CJS form at `src/helpers/<name>.js` that sets `module.exports = fn`, and an ESM form at `src/helpers/esm/<name>.js` that does `export default fn`. The plugin previously embedded only the ESM variant and reused it for every caller, which meant `require("@oxc-project/runtime/helpers/X")` returned the ESM namespace `{ default: fn }` rather than the function itself. This PR embeds both variants and dispatches resolution by `ImportKind`: `require()` gets the CJS variant (callable directly), `import` / dynamic `import()` get the ESM variant. Relative imports between virtual helpers preserve the importer's directory so an ESM helper's `./typeof.js` stays in `esm/`. Fixes #9263, where `_defineProperty` resolved to the ESM namespace inside CJS-classified modules and threw `TypeError: _defineProperty is not a function` at init.

sapphi-red · 2026-05-25T02:59:02Z

I guess we can add module condition to the runtime package to reduce the output size. I think it's not urgent though.

… (rolldown#9526) The on-disk `@oxc-project/runtime` package ships two variants of every helper: a CJS form at `src/helpers/<name>.js` that sets `module.exports = fn`, and an ESM form at `src/helpers/esm/<name>.js` that does `export default fn`. The plugin previously embedded only the ESM variant and reused it for every caller, which meant `require("@oxc-project/runtime/helpers/X")` returned the ESM namespace `{ default: fn }` rather than the function itself. This PR embeds both variants and dispatches resolution by `ImportKind`: `require()` gets the CJS variant (callable directly), `import` / dynamic `import()` get the ESM variant. Relative imports between virtual helpers preserve the importer's directory so an ESM helper's `./typeof.js` stays in `esm/`. Fixes rolldown#9263, where `_defineProperty` resolved to the ESM namespace inside CJS-classified modules and threw `TypeError: _defineProperty is not a function` at init.

@kylecannon

## [1.0.3] - 2026-05-27 ### 🚀 Features - transform: respect decorator strictNullChecks option (#9580) by @kylecannon - drop `defer` keyword (#9503) by @TheAlexLichter ### 🐛 Bug Fixes - ci: create target dir before cargo release-oxc update (#9584) by @shulaoda - ci: reorder prepare-release steps to avoid dirty git check failure (#9583) by @shulaoda - testing: canonicalize temp dir early and use platform-specific separator in test262 (#9582) by @shulaoda - testing: resolve symlinked temp dir in test262 snapshot normalization (#9581) by @shulaoda - testing: canonicalize temp dir path in test262 snapshot normalization (#9579) by @shulaoda - dev: `onOutput` called twice when initial build fails (#9552) by @hyf0 - dev: make `ensureCurrentBuildFinish` not returning error when engine closes (#9564) by @h-a-n-a - oxc-runtime: route require() to CJS helper variant (#9263) (#9526) by @IWANABETHATGUY - generator: use exporter chunk's export mode for CJS default re-exports (#9299) (#9529) by @IWANABETHATGUY - rolldown: always run reduced-atom static cycle check (#9441) (#9514) by @IWANABETHATGUY - apply transform.dropLabels before scanning (#9521) (#9522) by @IWANABETHATGUY - rolldown_watcher: take `rolldown` dep through the workspace (#9510) by @Boshen - cache: keep the scan-stage cache consistent when a build fails (#9495) by @h-a-n-a - skip JSON default-import namespace optimization for write targets (#9484) (#9489) by @IWANABETHATGUY - deps: skip pnpm frozen-lockfile on Netlify to dodge catalog mismatch bug (#9471) by @Boshen ### 🚜 Refactor - oxc-runtime: use Cow for helper path construction (#9538) by @IWANABETHATGUY - fold import defer phase drop into PreProcessor (#9524) by @IWANABETHATGUY - distinguish `map: null` vs `map: undefined` in transform hook output (#9497) by @sapphi-red ### 📚 Documentation - explain the policy for Rust crates (#9547) by @sapphi-red - cache: add design doc for cache (#9544) by @h-a-n-a - guide/troubleshooting: add TDZ error section (#9537) by @sapphi-red - dev-engine: add design doc for dev-engine (#9479) by @h-a-n-a - lazy-barrel: tweak some words (#9483) by @shulaoda - lazy-barrel: expand reasoning behind LARGE_BARREL_MODULES advice (#9477) by @shulaoda ### ⚡ Performance - generate: thread ast_table by value into codegen consumer (#9555) by @Boshen - finalizers: replace `_reExport` construction with a direct call to avoid calling `clone_in` (#9501) by @Dunqing - reorder hot-path boolean checks to short-circuit on cheap predicates first (#9523) by @Boshen ### 🧪 Testing - rolldown: regression fixture for #9401 (#9418) by @IWANABETHATGUY - failing test for #9441 (#9504) by @TheAlexLichter ### ⚙️ Miscellaneous Tasks - deps: upgrade oxc to 0.133.0 (#9563) by @Dunqing - deps: update crate-ci/typos action to v1.46.3 (#9576) by @renovate[bot] - deps: update mimalloc-safe to 0.1.62 (#9577) by @shulaoda - mimalloc-safe: update to a bug-fix branch for verification (#9569) by @shulaoda - deps: update test262 submodule for tests (#9551) by @rolldown-guard[bot] - point published crates' readme to root README.md (#9553) by @Boshen - replace actions-cool/issues-helper with gh CLI (#9543) by @Boshen - deps: update cargo-shear to 1.12.4 (#9541) by @Boshen - deps: update taiki-e/install-action action to v2.79.4 (#9535) by @renovate[bot] - deps: update github actions (#9532) by @renovate[bot] - deps: update rust crates (#9534) by @renovate[bot] - deps: update npm packages (#9533) by @renovate[bot] - gate experimental/testing-only items to silence dead_code in publish builds (#9517) by @Boshen - docs: deploy to Void (#9509) by @Boshen - release: set up cargo-release-oxc for publishing crates (#9476) by @Boshen - rolldown_plugin_lazy_compilation: add missing description (#9507) by @Boshen - mimalloc-safe: update to a bug-fix branch for verification (#9506) by @shulaoda - deps: update crate-ci/typos action to v1.46.2 (#9468) by @renovate[bot] ### ❤️ New Contributors * @kylecannon made their first contribution in [#9580](#9580)

@kylecannon

## [1.0.3] - 2026-05-27 ### 🚀 Features - transform: respect decorator strictNullChecks option (#9580) by @kylecannon - drop `defer` keyword (#9503) by @TheAlexLichter ### 🐛 Bug Fixes - ci: create target dir before cargo release-oxc update (#9584) by @shulaoda - ci: reorder prepare-release steps to avoid dirty git check failure (#9583) by @shulaoda - testing: canonicalize temp dir early and use platform-specific separator in test262 (#9582) by @shulaoda - testing: resolve symlinked temp dir in test262 snapshot normalization (#9581) by @shulaoda - testing: canonicalize temp dir path in test262 snapshot normalization (#9579) by @shulaoda - dev: `onOutput` called twice when initial build fails (#9552) by @hyf0 - dev: make `ensureCurrentBuildFinish` not returning error when engine closes (#9564) by @h-a-n-a - oxc-runtime: route require() to CJS helper variant (#9263) (#9526) by @IWANABETHATGUY - generator: use exporter chunk's export mode for CJS default re-exports (#9299) (#9529) by @IWANABETHATGUY - rolldown: always run reduced-atom static cycle check (#9441) (#9514) by @IWANABETHATGUY - apply transform.dropLabels before scanning (#9521) (#9522) by @IWANABETHATGUY - rolldown_watcher: take `rolldown` dep through the workspace (#9510) by @Boshen - cache: keep the scan-stage cache consistent when a build fails (#9495) by @h-a-n-a - skip JSON default-import namespace optimization for write targets (#9484) (#9489) by @IWANABETHATGUY - deps: skip pnpm frozen-lockfile on Netlify to dodge catalog mismatch bug (#9471) by @Boshen ### 🚜 Refactor - oxc-runtime: use Cow for helper path construction (#9538) by @IWANABETHATGUY - fold import defer phase drop into PreProcessor (#9524) by @IWANABETHATGUY - distinguish `map: null` vs `map: undefined` in transform hook output (#9497) by @sapphi-red ### 📚 Documentation - explain the policy for Rust crates (#9547) by @sapphi-red - cache: add design doc for cache (#9544) by @h-a-n-a - guide/troubleshooting: add TDZ error section (#9537) by @sapphi-red - dev-engine: add design doc for dev-engine (#9479) by @h-a-n-a - lazy-barrel: tweak some words (#9483) by @shulaoda - lazy-barrel: expand reasoning behind LARGE_BARREL_MODULES advice (#9477) by @shulaoda ### ⚡ Performance - generate: thread ast_table by value into codegen consumer (#9555) by @Boshen - finalizers: replace `_reExport` construction with a direct call to avoid calling `clone_in` (#9501) by @Dunqing - reorder hot-path boolean checks to short-circuit on cheap predicates first (#9523) by @Boshen ### 🧪 Testing - rolldown: regression fixture for #9401 (#9418) by @IWANABETHATGUY - failing test for #9441 (#9504) by @TheAlexLichter ### ⚙️ Miscellaneous Tasks - deps: upgrade oxc to 0.133.0 (#9563) by @Dunqing - deps: update crate-ci/typos action to v1.46.3 (#9576) by @renovate[bot] - deps: update mimalloc-safe to 0.1.62 (#9577) by @shulaoda - mimalloc-safe: update to a bug-fix branch for verification (#9569) by @shulaoda - deps: update test262 submodule for tests (#9551) by @rolldown-guard[bot] - point published crates' readme to root README.md (#9553) by @Boshen - replace actions-cool/issues-helper with gh CLI (#9543) by @Boshen - deps: update cargo-shear to 1.12.4 (#9541) by @Boshen - deps: update taiki-e/install-action action to v2.79.4 (#9535) by @renovate[bot] - deps: update github actions (#9532) by @renovate[bot] - deps: update rust crates (#9534) by @renovate[bot] - deps: update npm packages (#9533) by @renovate[bot] - gate experimental/testing-only items to silence dead_code in publish builds (#9517) by @Boshen - docs: deploy to Void (#9509) by @Boshen - release: set up cargo-release-oxc for publishing crates (#9476) by @Boshen - rolldown_plugin_lazy_compilation: add missing description (#9507) by @Boshen - mimalloc-safe: update to a bug-fix branch for verification (#9506) by @shulaoda - deps: update crate-ci/typos action to v1.46.2 (#9468) by @renovate[bot] ### ❤️ New Contributors * @kylecannon made their first contribution in [#9580](#9580)

IWANABETHATGUY force-pushed the 05-09-fix_9263 branch 2 times, most recently from 36ef49e to 8d4e742 Compare May 23, 2026 15:27

IWANABETHATGUY marked this pull request as ready for review May 23, 2026 16:48

IWANABETHATGUY requested review from h-a-n-a, hyf0, sapphi-red and shulaoda May 23, 2026 16:48

hyf0 approved these changes May 24, 2026

View reviewed changes

h-a-n-a approved these changes May 24, 2026

View reviewed changes

shulaoda self-assigned this May 24, 2026

IWANABETHATGUY force-pushed the 05-09-fix_9263 branch from 8d4e742 to 2842ec4 Compare May 25, 2026 01:32

IWANABETHATGUY added the graphite: merge-when-ready label May 25, 2026

graphite-app Bot force-pushed the 05-09-fix_9263 branch from 2842ec4 to f06c2a0 Compare May 25, 2026 01:36

graphite-app Bot merged commit f06c2a0 into main May 25, 2026
32 of 33 checks passed

graphite-app Bot removed the graphite: merge-when-ready label May 25, 2026

graphite-app Bot deleted the 05-09-fix_9263 branch May 25, 2026 01:41

graphite-app Bot temporarily deployed to void May 25, 2026 01:42 Inactive

rolldown-guard Bot mentioned this pull request May 27, 2026

release: v1.0.3 for rust crates #9585

Merged

shulaoda mentioned this pull request May 27, 2026

release: v1.0.3 #9586

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

fix(oxc-runtime): route require() to CJS helper variant (#9263)#9526

fix(oxc-runtime): route require() to CJS helper variant (#9263)#9526
graphite-app[bot] merged 1 commit into
mainfrom
05-09-fix_9263

IWANABETHATGUY commented May 23, 2026 •

edited

Loading

Uh oh!

IWANABETHATGUY commented May 23, 2026

Uh oh!

netlify Bot commented May 23, 2026 •

edited

Loading

Uh oh!

codspeed-hq Bot commented May 23, 2026

Uh oh!

hyf0 commented May 24, 2026

Uh oh!

IWANABETHATGUY commented May 24, 2026

Uh oh!

IWANABETHATGUY commented May 24, 2026

Uh oh!

hyf0 left a comment

Uh oh!

IWANABETHATGUY commented May 25, 2026 •

edited by graphite-app Bot

Loading

Uh oh!

IWANABETHATGUY commented May 25, 2026

Uh oh!

Uh oh!

sapphi-red commented May 25, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

Conversation

IWANABETHATGUY commented May 23, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

IWANABETHATGUY commented May 23, 2026

How to use the Graphite Merge Queue

Uh oh!

netlify Bot commented May 23, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

✅ Deploy Preview for rolldown-rs canceled.

Uh oh!

codspeed-hq Bot commented May 23, 2026

Merging this PR will not alter performance

Footnotes

Uh oh!

hyf0 commented May 24, 2026

Uh oh!

IWANABETHATGUY commented May 24, 2026

1. Almost all helpers are pure stateless functions

2. The on-disk package already does exactly this dispatch

3. The CJS variant is interop-safe by construction

4. Other bundlers in this niche behave the same way

5. The alternatives are worse

Uh oh!

IWANABETHATGUY commented May 24, 2026

Uh oh!

hyf0 left a comment

Choose a reason for hiding this comment

Uh oh!

IWANABETHATGUY commented May 25, 2026 • edited by graphite-app Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Merge activity

Uh oh!

IWANABETHATGUY commented May 25, 2026

Uh oh!

Uh oh!

sapphi-red commented May 25, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

IWANABETHATGUY commented May 23, 2026 •

edited

Loading

netlify Bot commented May 23, 2026 •

edited

Loading

IWANABETHATGUY commented May 25, 2026 •

edited by graphite-app Bot

Loading