GeezerVison

The World Wide Web, restored to 800 pixels of glory.

GeezerVison is showing github.com — esm: bypass CommonJS loader under `--default-type=module` by GeoffreyBooth · Pull Request #49986 · nodejs/node

Conversation

GeoffreyBooth commented Oct 1, 2023 •

edited

Loading

Uh oh!

There was an error while loading. Please reload this page.

Copy link

Copy Markdown

Member

This PR implements the part of #49432 where we avoid using the CommonJS loader for ESM entry points in the new --default-type=module mode. This effectively completes some TODOs in the code that note that particular code paths should be deprecated.

The only user-observable effects of this are that, under --experimental-default-type=module:

Extension searching is no longer provided for the main entry point. Users need to type node entry.js, not node entry.
Libraries that were monkey-patching the CommonJS loader via code loaded by --require will no longer achieve anything. They weren’t achieving much before, as monkey-patching the CommonJS loader only affects CommonJS and not ES modules, so it’s not like the previous behavior was really working very well (if at all) for applications with an ESM entry point.

This inches us closer to actually deprecating the CommonJS loader and having the ESM loader handle all modules, which is a long term goal that will aid in maintainability and future functionality for the modules subsystem.

Sorry, something went wrong.

Uh oh!

There was an error while loading. Please reload this page.

JakobJingleheimer reacted with hooray emoji

All reactions

🎉 1 reaction

GeoffreyBooth added module

Issues and PRs related to the module subsystem.

cli

Issues and PRs related to the Node.js command-line interface.

esm Issues and PRs related to the ECMAScript Modules implementation. labels

Oct 1, 2023

GeoffreyBooth requested a review from aduh95

October 1, 2023 04:13

nodejs-github-bot commented Oct 1, 2023

Copy link

Copy Markdown

Collaborator

Review requested:

@nodejs/loaders

All reactions

Sorry, something went wrong.

Uh oh!

There was an error while loading. Please reload this page.

nodejs-github-bot added needs-ci

PRs that need a full CI run.

process Issues and PRs related to the process subsystem. labels

Oct 1, 2023

GeoffreyBooth force-pushed the bypass-cjs-loader branch from a0fba47 to 4669340 Compare

October 1, 2023 04:27

aduh95 reviewed

Oct 1, 2023

View reviewed changes

Comment thread

lib/internal/modules/run_main.js Resolved

Uh oh!

There was an error while loading. Please reload this page.

GeoffreyBooth force-pushed the bypass-cjs-loader branch from 4669340 to 6c6aefe Compare

October 1, 2023 17:32

GeoffreyBooth commented Oct 1, 2023 •

edited

Loading

Uh oh!

There was an error while loading. Please reload this page.

Copy link

Copy Markdown

Member Author

So there are really two places where we resolve the input into a path: the resolveMainPath function, and this code within runMainESM:

    const main = path.isAbsolute(mainPath) ?
      pathToFileURL(mainPath).href : mainPath;

If we bypass or immediately return from resolveMainPath, then the mainPath input to runMainESM is the raw input from the CLI, like entry.js. And it usually won’t be absolute, so pathToFileURL isn’t getting called for it. So then we’re passing that input into esmLoader.import, which might work when the relative file path is parseable as a relative URL, which is most of the time—but not all of the time.

So I think either we undo the commit I just pushed, and always resolve the input into a path within resolveMainPath for both flows; or do something else. The more I think about this, the more it feels like runMainESM should always expect either an absolute file path or an absolute file URL, and code upstream of it is responsible for that conversion. We shouldn’t be resolving the input multiple times in multiple places. Thoughts @aduh95?

All reactions

Sorry, something went wrong.

Uh oh!

There was an error while loading. Please reload this page.

aduh95 reviewed

Oct 1, 2023

View reviewed changes

Comment thread

lib/internal/modules/run_main.js Resolved

Uh oh!

There was an error while loading. Please reload this page.

GeoffreyBooth commented Oct 2, 2023

Copy link

Copy Markdown

Member Author

@aduh95 I pushed some more commits, that undo the “streamline” commit, and add some tests and comments and improve an error message.

I’m open to being convinced otherwise, but it feels like a mistake to completely bypass resolveMainPath in ESM mode. Even if the esmLoader.import call within runMainESM somehow correctly parses a relative path as an URL (and please educate me on if that’s the case) we still should preserve the part of resolveMainPath that follows symlinks if --preserve-symlinks-main is not passed, right? We should presumably still support that in the new mode, right?

Assuming the above is all correct so far, I wanted a better error message for the “left off the file extension” case than just ENOENT: no such file or directory, lstat '/Users/geoffrey/Sites/node/test/fixtures/es-modules/package-without-type/index'. So I abstracted the “see if the resolution would’ve worked according to CommonJS rules” logic from esm/resolve.js and import it into the error path in run_main.js, so that node index prints an error message including Did you mean to import /Users/geoffrey/Sites/node/test/fixtures/es-modules/package-without-type/index.js? like we show for import specifiers that lack an extension. We were getting this message for free by passing through invalid input into esmLoader.import, but if we follow through with the toRealPath call in resolveMainPath then we need to handle it there.

All reactions

Sorry, something went wrong.

Uh oh!

There was an error while loading. Please reload this page.

GeoffreyBooth force-pushed the bypass-cjs-loader branch from 6f61ca0 to 4cf437c Compare

October 3, 2023 05:34

JakobJingleheimer approved these changes

Oct 4, 2023

View reviewed changes

JakobJingleheimer left a comment

Copy link

Copy Markdown

Member

There was a problem hiding this comment.

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Huzzah 🙂

A small nit: somewhere (I can't find the line now—on my mobile), it declares a typedef for default-type as a string: it's not any old string, it's an enum of specific strings. Ideally, those are numerated in the typedef.

Sorry, something went wrong.

Uh oh!

There was an error while loading. Please reload this page.

All reactions

ljharb reviewed

Oct 5, 2023

View reviewed changes

Comment thread

lib/internal/modules/esm/resolve.js Outdated

Comment on lines 1161 to 1164

ljharb Oct 5, 2023

Copy link

Copy Markdown

Member

There was a problem hiding this comment.

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change

	error.stack =
	ArrayPrototypeShift(lines) + '\n' +
	hint + '\n' +
	ArrayPrototypeJoin(lines, '\n');
	error.stack = `${ArrayPrototypeShift(lines)}\n${hint}\n${ArrayPrototypeJoin(lines, '\n')}`;

Sorry, something went wrong.

Uh oh!

There was an error while loading. Please reload this page.

All reactions

aduh95 reviewed

Oct 5, 2023

View reviewed changes

Comment thread

lib/internal/modules/run_main.js Resolved

Uh oh!

There was an error while loading. Please reload this page.

aduh95 reviewed

Oct 5, 2023

View reviewed changes

Comment thread

lib/internal/modules/run_main.js Resolved

Uh oh!

There was an error while loading. Please reload this page.

GeoffreyBooth commented Oct 5, 2023

Copy link

Copy Markdown

Member Author

I reverted the optimization for the legacy flow. I don’t want to make changes to the legacy flow in this PR, I’m too nervous about such changes introducing bugs since the changes don’t break any tests when I’d expect that they should. I want to just land the new feature intended by this PR, bypassing the CommonJS loader under --default-type=module, and we can refactor the legacy flow in a separate PR (and add more tests for it).

All reactions

Sorry, something went wrong.

Uh oh!

There was an error while loading. Please reload this page.

GeoffreyBooth and others added 11 commits

October 5, 2023 07:59


          esm: bypass CommonJS loader under --default-type

97b1966


          Streamline

30821db


          Revert "Streamline"

286c632

This reverts commit 6c6aefec5ede1070b2a17864075b2cc3d60718ee.


          Move/improve tests; decorate "not found" error to provide hint about …

…extension searching; fix lint


          Update docs

bdafa66


          Fix for Windows

83884e5


          Move out main re-resolution

162b6fa


          Optimize

22bddbd


          Revert "Optimize"

6c97b9a

This reverts commit 6d1a6e51118d1c4bd45cb21fad337ac2f87c8aa1.


          Update lib/internal/modules/run_main.js

32aaabc

Co-authored-by: Antoine du Hamel <[email protected]>


          Return early if possible

038dfb8

GeoffreyBooth mentioned this pull request

Oct 5, 2023

Proposal: Set --experimental-default-type mode by detecting ESM syntax in entry point #50043

Closed

GeoffreyBooth added commit-queue

PRs queued for automated landing through the Commit Queue.

commit-queue-squash PRs the Commit Queue should land as one squashed commit. labels

Oct 5, 2023

nodejs-github-bot removed the commit-queue PRs queued for automated landing through the Commit Queue. label

Oct 5, 2023

nodejs-github-bot merged commit 19b470f into nodejs:main

Oct 5, 2023

nodejs-github-bot commented Oct 5, 2023

Copy link

Copy Markdown

Collaborator

Landed in 19b470f

All reactions

Sorry, something went wrong.

Uh oh!

There was an error while loading. Please reload this page.

GeoffreyBooth deleted the bypass-cjs-loader branch

October 5, 2023 18:02

RafaelGSS mentioned this pull request

Oct 9, 2023

2023-10-17, Version 21.0.0 (Current) #49870

Merged

targos mentioned this pull request

Nov 11, 2023

[v18.x] Backport most ESM and customization hook changes #50669

Closed

targos pushed a commit that referenced this pull request

Nov 11, 2023


          esm: bypass CommonJS loader under --default-type

4de838f

PR-URL: #49986
Reviewed-By: Jacob Smith <[email protected]>
Reviewed-By: Antoine du Hamel <[email protected]>

targos pushed a commit to targos/node that referenced this pull request

Nov 11, 2023


          esm: bypass CommonJS loader under --default-type

151e812

PR-URL: nodejs#49986
Reviewed-By: Jacob Smith <[email protected]>
Reviewed-By: Antoine du Hamel <[email protected]>

targos mentioned this pull request

Nov 12, 2023

v20.10.0 release proposal #50682

Merged

targos added the backported-to-v18.x label

Nov 23, 2023

targos pushed a commit that referenced this pull request

Nov 23, 2023


          esm: bypass CommonJS loader under --default-type

c3c945e

PR-URL: #49986
Backport-PR-URL: #50669
Reviewed-By: Jacob Smith <[email protected]>
Reviewed-By: Antoine du Hamel <[email protected]>

targos mentioned this pull request

Nov 28, 2023

v18.19.0 release proposal #50953

Merged

juanarbol mentioned this pull request

Dec 6, 2023

2023-12-06, N|Solid v5.0.1 Iron Release nodesource/nsolid#41

Merged

ckerr added a commit to electron/electron that referenced this pull request

Dec 7, 2023


          chore: update feat_initialize_asar_support.patch

bcd8239

no code changes; patch just needed an update due to nearby upstream changes

Xref: nodejs/node#49986

ckerr added a commit to electron/electron that referenced this pull request

Dec 7, 2023


          chore: update chore_allow_the_node_entrypoint_to_be_a_builtin_module.…

88449ae

…patch

Xref: nodejs/node#49986

minor manual changes needed to sync with upstream change

codebytere pushed a commit to electron/electron that referenced this pull request

Dec 8, 2023


          chore: update feat_initialize_asar_support.patch

943ccb7

no code changes; patch just needed an update due to nearby upstream changes

Xref: nodejs/node#49986

codebytere pushed a commit to electron/electron that referenced this pull request

Dec 8, 2023


          chore: update chore_allow_the_node_entrypoint_to_be_a_builtin_module.…

3494aa6

…patch

Xref: nodejs/node#49986

minor manual changes needed to sync with upstream change

codebytere added a commit to electron/electron that referenced this pull request

Dec 8, 2023


          esm: bypass CommonJS loader under --default-type=module

19f12ca

nodejs/node#49986

codebytere added a commit to electron/electron that referenced this pull request

Dec 11, 2023


          chore: bump node to v20.10.0 (main) (#40675)

890a557

* chore: bump node in DEPS to v20.10.0

* chore: update feat_initialize_asar_support.patch

no code changes; patch just needed an update due to nearby upstream changes

Xref: nodejs/node#49986

* chore: update pass_all_globals_through_require.patch

no manual changes; patch applied with fuzz

Xref: nodejs/node#49657

* chore: update refactor_allow_embedder_overriding_of_internal_fs_calls

Xref: nodejs/node#49912

no code changes; patch just needed an update due to nearby upstream changes

* chore: update chore_allow_the_node_entrypoint_to_be_a_builtin_module.patch

Xref: nodejs/node#49986

minor manual changes needed to sync with upstream change

* update fix_expose_the_built-in_electron_module_via_the_esm_loader.patch

Xref: nodejs/node#50096
Xref: nodejs/node#50314
in lib/internal/modules/esm/load.js, update the code that checks for
`format === 'electron'`. I'd like 👀 on this

Xref: nodejs/node#49657
add braces in lib/internal/modules/esm/translators.js to sync with upstream

* fix: lazyload fs in esm loaders to apply asar patches

* nodejs/node#50127
* nodejs/node#50096

* esm: jsdoc for modules code

nodejs/node#49523

* test: set test-cli-node-options as flaky

nodejs/node#50296

* deps: update c-ares to 1.20.1

nodejs/node#50082

* esm: bypass CommonJS loader under --default-type=module

nodejs/node#49986

* deps: update uvwasi to 0.0.19

nodejs/node#49908

* lib,test: do not hardcode Buffer.kMaxLength

nodejs/node#49876

* crypto: account for disabled SharedArrayBuffer

nodejs/node#50034

* test: fix edge snapshot stack traces

nodejs/node#49659

* src: generate snapshot with --predictable

nodejs/node#48749

* chore: fixup patch indices

* fs: throw errors from sync branches instead of separate implementations

nodejs/node#49913

* crypto: ensure valid point on elliptic curve in SubtleCrypto.importKey

nodejs/node#50234

* esm: detect ESM syntax in ambiguous JavaScrip

nodejs/node#50096

* fixup! test: fix edge snapshot stack traces

* esm: unflag extensionless ES module JavaScript and Wasm in module scope

nodejs/node#49974

* [tagged-ptr] Arrowify objects

https://chromium-review.googlesource.com/c/v8/v8/+/4705331

---------

Co-authored-by: electron-roller[bot] <84116207+electron-roller[bot]@users.noreply.github.com>
Co-authored-by: Charles Kerr <[email protected]>
Co-authored-by: Shelley Vohr <[email protected]>

debadree25 pushed a commit to debadree25/node that referenced this pull request

Apr 15, 2024


          esm: bypass CommonJS loader under --default-type

672f793

PR-URL: nodejs#49986
Reviewed-By: Jacob Smith <[email protected]>
Reviewed-By: Antoine du Hamel <[email protected]>

sercher added a commit to sercher/graaljs that referenced this pull request

Apr 25, 2024


          esm: bypass CommonJS loader under --default-type

a792185

PR-URL: nodejs/node#49986
Backport-PR-URL: nodejs/node#50669
Reviewed-By: Jacob Smith <[email protected]>
Reviewed-By: Antoine du Hamel <[email protected]>

sercher added a commit to sercher/graaljs that referenced this pull request

Apr 25, 2024


          esm: bypass CommonJS loader under --default-type

fdf2ca1

PR-URL: nodejs/node#49986
Backport-PR-URL: nodejs/node#50669
Reviewed-By: Jacob Smith <[email protected]>
Reviewed-By: Antoine du Hamel <[email protected]>

bilbospocketses added a commit to bilbospocketses/ws-scrcpy-web that referenced this pull request

Apr 18, 2026


          chore(multiplexer): drop Event/CloseEvent/MessageEvent polyfills, kee…

a912426

…p ErrorEvent

Node 24 (repo engines floor) has Event, CloseEvent, and MessageEvent as
globals — the typeof checks in Event.ts / CloseEventClass.ts /
MessageEventClass.ts always picked native on any supported runtime.
Inlining native constructors at the 5 Multiplexer/Message call sites and
deleting the three dead polyfill files.

ErrorEvent is NOT yet a Node global — verified on Node 24.14.1 via
'typeof ErrorEvent === "undefined"' (tracked upstream at
nodejs/node#49986; internal-only ErrorEvent lives in undici). Kept
ErrorEventClass.ts and rewrote it to extend native Event (it previously
extended the now-deleted Event2). Runtime smoke test confirms the
polyfill dispatches and listeners see error/message/filename/lineno/colno
fields correctly.

Net: -137/+18 across 6 files, 0 tsc errors, 89/89 tests pass, webpack
clean, generated .d.ts byte-identical (920 bytes).

bilbospocketses added a commit to bilbospocketses/ws-scrcpy-web that referenced this pull request

Apr 27, 2026


          chore(multiplexer): drop Event/CloseEvent/MessageEvent polyfills, kee…

1cea742

…p ErrorEvent

Node 24 (repo engines floor) has Event, CloseEvent, and MessageEvent as
globals — the typeof checks in Event.ts / CloseEventClass.ts /
MessageEventClass.ts always picked native on any supported runtime.
Inlining native constructors at the 5 Multiplexer/Message call sites and
deleting the three dead polyfill files.

ErrorEvent is NOT yet a Node global — verified on Node 24.14.1 via
'typeof ErrorEvent === "undefined"' (tracked upstream at
nodejs/node#49986; internal-only ErrorEvent lives in undici). Kept
ErrorEventClass.ts and rewrote it to extend native Event (it previously
extended the now-deleted Event2). Runtime smoke test confirms the
polyfill dispatches and listeners see error/message/filename/lineno/colno
fields correctly.

Net: -137/+18 across 6 files, 0 tsc errors, 89/89 tests pass, webpack
clean, generated .d.ts byte-identical (920 bytes).

Reviewers

ljharb

ljharb left review comments

JakobJingleheimer

JakobJingleheimer approved these changes

aduh95

aduh95 approved these changes

Uh oh!

There was an error while loading. Please reload this page.

Assignees

No one assigned

Labels

author ready

PRs with CI started, the required approvals, and no outstanding review comments.

cli

Issues and PRs related to the Node.js command-line interface.

commit-queue-squash

PRs the Commit Queue should land as one squashed commit.

esm

Issues and PRs related to the ECMAScript Modules implementation.

module

Issues and PRs related to the module subsystem.

needs-ci

PRs that need a full CI run.

process

Issues and PRs related to the process subsystem.

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

Uh oh!

There was an error while loading. Please reload this page.

6 participants

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.

You are visitor 01402503 — GeezerVison does not store the pages you look up.
Best viewed with Netscape Navigator 4.7 or Internet Explorer 6 at 800×600. No cookies required.

GeezerVison

Conversation

GeoffreyBooth commented Oct 1, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

nodejs-github-bot commented Oct 1, 2023

Uh oh!

Uh oh!

GeoffreyBooth commented Oct 1, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

GeoffreyBooth commented Oct 2, 2023

Uh oh!

JakobJingleheimer left a comment

Choose a reason for hiding this comment

Uh oh!

ljharb Oct 5, 2023

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

GeoffreyBooth commented Oct 5, 2023

Uh oh!

nodejs-github-bot commented Oct 5, 2023

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

GeoffreyBooth commented Oct 1, 2023 •

edited

Loading

Uh oh!

There was an error while loading. Please reload this page.

GeoffreyBooth commented Oct 1, 2023 •

edited

Loading

Uh oh!

There was an error while loading. Please reload this page.