fix(resolver): honor package.json#type for .jsx/.tsx extensions (#9690)#9699
Merged
Merged
Conversation
Merging this PR will not alter performance
Comparing Footnotes
|
✅ Deploy Preview for rolldown-rs canceled.
|
1d83558 to
e340989
Compare
sapphi-red
reviewed
Jun 10, 2026
57cb539 to
7923b0d
Compare
Member
Author
How to use the Graphite Merge QueueAdd 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. |
sapphi-red
approved these changes
Jun 10, 2026
Member
Author
Merge activity
|
… (#9699) `oxc_resolver` only resolves `module_type` from `package.json#type` for `.js`/`.ts` extensions, following the Node.js ESM spec which has no native `.jsx`/`.tsx`. rolldown relied on that `module_type`, so a `.tsx` entry under `"type": "module"` fell back to `Unknown` instead of `EsmPackageJson`. This made identical `.ts` and `.tsx` sources emit different CJS interop for a default import of an external CJS module: the `.ts` file got node-mode `__toESM(mod, 1)` while the `.tsx` file got `__toESM(mod)`, breaking `mod.default` at runtime. As a bundler, rolldown already treats `.jsx`/`.tsx` as js-like extensions, so fall back to reading the nearest `package.json#type` directly when oxc reports no `module_type`, keeping all js-like extensions consistent (and matching esbuild, which emits `__toESM(require(...), 1)` for both). closed #9690
7923b0d to
289c520
Compare
Closed
Merged
shulaoda
pushed a commit
that referenced
this pull request
Jun 11, 2026
## [1.1.1] - 2026-06-11 ### 🚀 Features - ecmascript_utils: introduce AstFactory for AST construction (#9682) by @hyf0 - drop no-op init calls for empty wrapped-ESM modules (#9678) by @IWANABETHATGUY ### 🐛 Bug Fixes - resolver: honor package.json#type for .jsx/.tsx extensions (#9690) (#9699) by @IWANABETHATGUY - hmr: report the full-reload reason for the invalidate-loop case (#9708) by @hyf0 - explicit `moduleSideEffects` from a hook must take priority over the `package.json#sideEffects` (#9688) by @sapphi-red - keep the `rolldown-runtime` name for the standalone runtime chunk (#9685) by @shulaoda - lazy-barrel: request all exports for entry barrels on first encounter (#9672) by @shulaoda - finalizer: skip init_*() for tree-shaken wrapped ESM owners (#9669) by @IWANABETHATGUY - order `chunk.imports` by execution order (#9654) by @chuganzy - vite-resolve: preserve slash separators for Sass partial exports (#9546) by @cjc0013 - cross-chunk CJS wrapper shadowed by author-local binding (#9648) by @IWANABETHATGUY ### 🚜 Refactor - precompute wrapped-ESM init metadata in generate stage (#9712) by @IWANABETHATGUY - ecmascript_utils: fold construction ext traits onto AstFactory and delete AstSnippet (#9702) by @hyf0 - finalizer: finish ScopeHoistingFinalizer migration to AstFactory (#9701) by @hyf0 - finalizer: migrate module_finalizers/mod.rs to AstFactory (#9700) by @hyf0 - hmr: migrate hmr finalizer to AstFactory (#9695) by @hyf0 - plugin: migrate vite_build_import_analysis to AstFactory (#9693) by @hyf0 - scanner: migrate tweak_ast_for_scanning to AstFactory (#9683) by @hyf0 - always split runtime module first (#9419) by @IWANABETHATGUY ### 📚 Documentation - tsconfig: correct auto-discovery resolution to match TypeScript (#9714) by @shulaoda - design: plan to unify all internal AST construction (#9673) by @hyf0 - tsconfig: align reference resolution docs with TypeScript behavior (#9641) by @shulaoda ### ⚡ Performance - avoid per-module join Strings in scope-hoisting concatenation (#9645) by @Boshen - avoid intermediate Strings in the ESM export clause (#9644) by @Boshen - reuse a scratch buffer for facade namespace names in the scanner (#9642) by @Boshen - reuse the import-matching tracker stack across named imports (#9643) by @Boshen - avoid cloning the per-chunk export-items map in render_chunk_exports (#9639) by @Boshen - avoid CompactStr allocation in sorted-exports membership check (#9640) by @Boshen ### 🧪 Testing - add more `moduleSideEffects` precedence tests (#9689) by @sapphi-red - 9651: drop shimMissingExports from regression fixture (#9674) by @IWANABETHATGUY ### ⚙️ Miscellaneous Tasks - update react repo links (#9711) by @iiio2 - remove outdated pnpm configurations (#9666) by @btea - deps: update github actions to v2.81.5 (#9665) by @renovate[bot] - testing: migrate test harness off deprecated inlineDynamicImports (#9710) by @IWANABETHATGUY - hmr: delete commented-out dead code left from old register-module codegen (#9707) by @hyf0 - deps: update napi-rs toolchain (#9706) by @shulaoda - deps: update rollup submodule for tests to v4.61.1 (#9676) by @rolldown-guard[bot] - deps: update test262 submodule for tests (#9677) by @rolldown-guard[bot] - deps: update oxc to 0.135.0 (#9670) by @Boshen - pass ALGOLIA_APP_ID and ALGOLIA_API_KEY to void deploy (#9667) by @Boshen - deps: update github actions (#9662) by @renovate[bot] - deps: update rust crates (#9663) by @renovate[bot] - deps: update rolldown-plugin-dts to v0.25.2 (#9661) by @renovate[bot] - deps: update typos to v1.47.2 (#9660) by @renovate[bot] - deps: update docs dependencies (#9657) by @bddjr - deps: update typos to v1.47.1 (#9655) by @renovate[bot] - deploy website in its own workflow, only on docs output change (#9650) by @Boshen - publish to pkg.pr.new add pm and `commentWithDev` option (#9638) by @btea ### ❤️ New Contributors * @chuganzy made their first contribution in [#9654](#9654) * @cjc0013 made their first contribution in [#9546](#9546)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.

oxc_resolveronly resolvesmodule_typefrompackage.json#typefor.js/.tsextensions, following the Node.js ESM spec which has no native.jsx/.tsx. rolldown relied on thatmodule_type, so a.tsxentry under"type": "module"fell back toUnknowninstead ofEsmPackageJson.This made identical
.tsand.tsxsources emit different CJS interop for a default import of an external CJS module: the.tsfile got node-mode__toESM(mod, 1)while the.tsxfile got__toESM(mod), breakingmod.defaultat runtime.As a bundler, rolldown already treats
.jsx/.tsxas js-like extensions, so fall back to reading the nearestpackage.json#typedirectly when oxc reports nomodule_type, keeping all js-like extensions consistent (and matching esbuild, which emits__toESM(require(...), 1)for both).closed #9690