module: Allow runMain to be ESM #14369

bmeck · 2017-07-19T15:17:34Z

This follows the EPS an allows the node CLI to have ESM as an entry point.
node ./example.mjs. A newer V8 is needed for import() so that is not
included. import.meta is still in specification stage so that also is not
included.

Author: Bradley Farias bradley.meck@gmail.com
Author: Guy Bedford guybedford@gmail.com
Author: Jan Krems jan.krems@groupon.com

Checklist

make -j4 test (UNIX), or vcbuild test (Windows) passes
tests and/or benchmarks are included
documentation is changed or added
commit message follows commit guidelines

Affected core subsystem(s)

bmeck · 2017-07-19T15:19:32Z

Going to be on vacation til TC39 next week / tests need to be pulled in from sketchy. Opening this since it is going to want a lot of eyes.

jasnell · 2017-07-19T19:44:53Z

Great to see progress here. Won't have an opportunity to review likely until after NodeSummit next week.

TimothyGu

More to come later. Is the C++ resolution necessary or for performance? I kinda drew the line at parsing JSON in C++...

TimothyGu · 2017-07-20T02:03:04Z

lib/internal/loader/resolveRequestUrl.js

+ }
+
+ async createModule() {
+ const source = `${await asyncReadFile(this.url.pathname)}`;


There's a specialized function you could use instead of this.url.pathname.

TimothyGu · 2017-07-20T02:03:27Z

lib/internal/loader/resolveRequestUrl.js

+ `${this.url}`);
+ }
+}
+Object.setPrototypeOf(StandardModuleRequest.prototype, Object.create(null));


Is this necessary? If so, why not just set the prototype to null?

This is an encapsulation general thing so that it does not inherit from Object so cannot be affected by mutated globals. In line with #10135

Yeah I get that, but you don't need to set the prototype object to an object with prototype null, you can just set the prototype of the prototype object to null.

A chart for easier consumption of what I just wrote:

Current: | StandardModuleRequest.prototype | || [[Prototype]] vv | Object.create(null) | || [[Prototype]] vv | null | Why not: | StandardModuleRequest.prototype | || [[Prototype]] vv | null |

oh, yes. we should change that.

TimothyGu · 2017-07-20T02:07:47Z

lib/internal/loader/Loader.js

+const ModuleMap = require('internal/loader/ModuleMap');
+const ModuleJob = require('internal/loader/ModuleJob');
+const resolveRequestUrl = require('internal/loader/resolveRequestUrl');
+const {Error, Promise: {resolve: PromiseResolve}} = global;


Is this (and safe_globals and other such things) necessary?

TimothyGu · 2017-07-20T02:08:11Z

lib/internal/safe_globals.js

@@ -0,0 +1,24 @@
+const copyProps = (unsafe, safe) => {


If this is necessary, 'use strict'.

was fixed, forgot to reply

TimothyGu · 2017-07-20T02:24:19Z

lib/internal/loader/resolveRequestUrl.js

+const {URL} = require('url');
+const internalCJSModule = require('internal/module');
+const NativeModule = require('native_module');
+const Path = require('path');


Why not use destructuring to prevent monkey patching as has been done elsewhere? Either way s/P/p/.

TimothyGu · 2017-07-20T02:40:02Z

src/loader/module_wrap.cc

+ return;
+ }
+
+ const Local<Value> ret(result.ToObject(


Why not just ret =?

@TimothyGu I agree with your comments here, but it might be easiest to just do what @targos did and push changes, esp. for suggestions like these, yourself

@addaleax 👍

TimothyGu · 2017-07-20T02:42:14Z

src/loader/module_wrap.cc

+}
+
+using namespace node::url;
+ URL __init_cwd() {


Weird identation...

TimothyGu · 2017-07-20T02:44:32Z

src/loader/module_wrap.cc

+ else if (specifier[0] == '.') {
+ if (len == 1 || specifier[1] == '/') return true;
+ else if (specifier[1] == '.') {
+ if (len == 2 || specifier[2] == '/') return true;


Does this support Windows?

Never mind, this is to be consistent with https://html.spec.whatwg.org/multipage/webappapis.html#resolve-a-module-specifier.

TimothyGu · 2017-07-20T02:46:50Z

src/loader/module_wrap.cc

+ return URL("");
+ }
+ URL Resolve(std::string specifier, URL* base, bool read_pkg_json) {
+ auto pure_url = URL(specifier.c_str());


URL takes a std::string as input.

TimothyGu · 2017-07-20T02:48:01Z

src/loader/module_wrap.cc

+using v8::EscapableHandleScope;
+using v8::PropertyDescriptor;
+
+const std::string EXTENSIONS[] = {".mjs", ".js", ".json", ".node"};


Did we reach a conclusion on the non-Web-compatible automatic extension adding behavior?

mcollina

This is a very complex feature to review without any attached unit tests. Would you mind adding them?

targos · 2017-07-20T10:23:19Z

I took the liberty to fix ESLint issues and compilation.
I don't think this is ready to be reviewed:

make test

$ make test
make -C out BUILDTYPE=Release V=1
  touch 9349840c76d301fd95cb4ac8c5f455128549de34.intermediate
  LD_LIBRARY_PATH=/home/mzasso/git/nodejs/node/out/Release/lib.host:/home/mzasso/git/nodejs/node/out/Release/lib.target:$LD_LIBRARY_PATH; export LD_LIBRARY_PATH; cd ../deps/v8/src/inspector; mkdir -p /home/mzasso/git/nodejs/node/out/Release/obj/gen/src/inspector/protocol /home/mzasso/git/nodejs/node/out/Release/obj/gen/include/inspector; python ../../third_party/inspector_protocol/CodeGenerator.py --jinja_dir ../../third_party --output_base "/home/mzasso/git/nodejs/node/out/Release/obj/gen/src/inspector" --config inspector_protocol_config.json
rm 9349840c76d301fd95cb4ac8c5f455128549de34.intermediate
if [ ! -r node -o ! -L node ]; then ln -fs out/Release/node node; fi
make build-addons
make -C out BUILDTYPE=Release V=1
  touch 9349840c76d301fd95cb4ac8c5f455128549de34.intermediate
  LD_LIBRARY_PATH=/home/mzasso/git/nodejs/node/out/Release/lib.host:/home/mzasso/git/nodejs/node/out/Release/lib.target:$LD_LIBRARY_PATH; export LD_LIBRARY_PATH; cd ../deps/v8/src/inspector; mkdir -p /home/mzasso/git/nodejs/node/out/Release/obj/gen/src/inspector/protocol /home/mzasso/git/nodejs/node/out/Release/obj/gen/include/inspector; python ../../third_party/inspector_protocol/CodeGenerator.py --jinja_dir ../../third_party --output_base "/home/mzasso/git/nodejs/node/out/Release/obj/gen/src/inspector" --config inspector_protocol_config.json
rm 9349840c76d301fd95cb4ac8c5f455128549de34.intermediate
if [ ! -r node -o ! -L node ]; then ln -fs out/Release/node node; fi
rm -f -r test/addons/??_*/
./node tools/doc/addon-verify.js
/home/mzasso/git/nodejs/node/tools/doc/addon-verify.js:13
const tokens = marked.lexer(contents);
                     ^

TypeError: Cannot read property 'lexer' of undefined
    at Object.<anonymous> (/home/mzasso/git/nodejs/node/tools/doc/addon-verify.js:13:22)
    at Module._compile (module.js:598:30)
    at Object.Module._extensions..js (module.js:609:10)
    at Module.load (module.js:532:32)
    at tryModuleLoad (module.js:495:12)
    at Function.Module._load (module.js:489:3)
    at Function.Module.runMain (module.js:634:10)
    at startup (bootstrap_node.js:158:16)
    at bootstrap_node.js:575:3
Makefile:230: recipe for target 'test/addons/.docbuildstamp' failed
make[1]: *** [test/addons/.docbuildstamp] Error 1
Makefile:196: recipe for target 'test' failed
make: *** [test] Error 2

make test-parallel

$ make test-parallel
make -C out BUILDTYPE=Release V=1
  touch 9349840c76d301fd95cb4ac8c5f455128549de34.intermediate
  LD_LIBRARY_PATH=/home/mzasso/git/nodejs/node/out/Release/lib.host:/home/mzasso/git/nodejs/node/out/Release/lib.target:$LD_LIBRARY_PATH; export LD_LIBRARY_PATH; cd ../deps/v8/src/inspector; mkdir -p /home/mzasso/git/nodejs/node/out/Release/obj/gen/src/inspector/protocol /home/mzasso/git/nodejs/node/out/Release/obj/gen/include/inspector; python ../../third_party/inspector_protocol/CodeGenerator.py --jinja_dir ../../third_party --output_base "/home/mzasso/git/nodejs/node/out/Release/obj/gen/src/inspector" --config inspector_protocol_config.json
rm 9349840c76d301fd95cb4ac8c5f455128549de34.intermediate
if [ ! -r node -o ! -L node ]; then ln -fs out/Release/node node; fi
/usr/bin/python tools/test.py --mode=release parallel -J
=== release test-assert-checktag ===
Path: parallel/test-assert-checktag
/home/mzasso/git/nodejs/node/test/parallel/test-assert-checktag.js:17
  return common.expectsError({
               ^

TypeError: Cannot read property 'expectsError' of undefined
    at re (/home/mzasso/git/nodejs/node/test/parallel/test-assert-checktag.js:17:16)
    at Object.<anonymous> (/home/mzasso/git/nodejs/node/test/parallel/test-assert-checktag.js:39:19)
    at Module._compile (module.js:598:30)
    at Object.Module._extensions..js (module.js:609:10)
    at Module.load (module.js:532:32)
    at tryModuleLoad (module.js:495:12)
    at Function.Module._load (module.js:489:3)
    at Function.Module.runMain (module.js:634:10)
    at startup (bootstrap_node.js:158:16)
    at bootstrap_node.js:575:3
Command: out/Release/node /home/mzasso/git/nodejs/node/test/parallel/test-assert-checktag.js
=== release test-assert-deep ===
Path: parallel/test-assert-deep
/home/mzasso/git/nodejs/node/test/parallel/test-assert-deep.js:17
  return common.expectsError({
               ^

TypeError: Cannot read property 'expectsError' of undefined
    at re (/home/mzasso/git/nodejs/node/test/parallel/test-assert-deep.js:17:16)
    at Object.<anonymous> (/home/mzasso/git/nodejs/node/test/parallel/test-assert-deep.js:34:17)
    at Module._compile (module.js:598:30)
    at Object.Module._extensions..js (module.js:609:10)
    at Module.load (module.js:532:32)
    at tryModuleLoad (module.js:495:12)
    at Function.Module._load (module.js:489:3)
    at Function.Module.runMain (module.js:634:10)
    at startup (bootstrap_node.js:158:16)
    at bootstrap_node.js:575:3
Command: out/Release/node /home/mzasso/git/nodejs/node/test/parallel/test-assert-deep.js
=== release test-assert ===
Path: parallel/test-assert
/home/mzasso/git/nodejs/node/test/parallel/test-assert.js:42
  common.expectsError({
        ^

TypeError: Cannot read property 'expectsError' of undefined
    at Object.<anonymous> (/home/mzasso/git/nodejs/node/test/parallel/test-assert.js:42:9)
    at Module._compile (module.js:598:30)
    at Object.Module._extensions..js (module.js:609:10)
    at Module.load (module.js:532:32)
    at tryModuleLoad (module.js:495:12)
    at Function.Module._load (module.js:489:3)
    at Function.Module.runMain (module.js:634:10)
    at startup (bootstrap_node.js:158:16)
    at bootstrap_node.js:575:3
Command: out/Release/node /home/mzasso/git/nodejs/node/test/parallel/test-assert.js
=== release test-async-hooks-close-during-destroy ===
Path: parallel/test-async-hooks-close-during-destroy
/home/mzasso/git/nodejs/node/test/parallel/test-async-hooks-close-during-destroy.js:14
  init: common.mustCallAtLeast((id, provider, triggerAsyncId) => {
              ^

TypeError: Cannot read property 'mustCallAtLeast' of undefined
    at Object.<anonymous> (/home/mzasso/git/nodejs/node/test/parallel/test-async-hooks-close-during-destroy.js:14:15)
    at Module._compile (module.js:598:30)
    at Object.Module._extensions..js (module.js:609:10)
    at Module.load (module.js:532:32)
    at tryModuleLoad (module.js:495:12)
    at Function.Module._load (module.js:489:3)
    at Function.Module.runMain (module.js:634:10)
    at startup (bootstrap_node.js:158:16)
    at bootstrap_node.js:575:3
Command: out/Release/node /home/mzasso/git/nodejs/node/test/parallel/test-async-hooks-close-during-destroy.js

...

targos · 2017-07-20T14:05:04Z

src/loader/module_wrap.cc

+ TryCatch compileCatch(iso);
+ auto maybe_mod = ScriptCompiler::CompileModule(iso, &source);
+ if (compileCatch.HasCaught()) {
+ compileCatch.ReThrow();


Not sure but I think all the TryCatch can be replaced by maybe.isEmpty() checks. /cc @addaleax

Yes, checks like these should just be maybe.IsEmpty()

(Unless there is some weird interaction with uncaught exception handlers that this code relies on, but I don’t think that is the case, and if it is, that should be documented)

Fishrock123 · 2017-07-20T14:58:47Z

adding myself as an assignee in an attempt to not forget about it

TimothyGu · 2017-07-21T13:13:24Z

The entire test suite passes here after dbd1ec3550423b0476d57bf638594bdd41e5f607. @targos can you recheck?

targos · 2017-07-21T13:51:26Z

@TimothyGu I pushed a last fix. Now make test passes.

addaleax · 2017-07-21T13:54:16Z

@targos @TimothyGu You are awesome! 💙

CI: https://ci.nodejs.org/job/node-test-commit/11292/

Fishrock123

Some rough review. Will give it a try to day hopefully.

I suppose we will need quite the assortment of test cases? Or are there already standard tests somewhere?

Also all errors should probably switched to the new internal/errors system.

Fishrock123 · 2017-07-21T13:54:08Z

lib/internal/loader/index.js

+const loader = new Loader();
+
+exports = module.exports = {
+ loader,


Is this file only necessary to keep a global loader instance?

In other modules e.g. the repl, the REPL object has it as a static reference.

we could move it to be a static ref since these files are never exposed to user code.

Fishrock123 · 2017-07-21T13:55:18Z

src/node_url.cc

@@ -1283,7 +1283,9 @@ void URL::Parse(const char* input,
 }
 break;
 case kNoScheme:
- cannot_be_base = base->flags & URL_FLAGS_CANNOT_BE_BASE;
+ if (has_base) {


Should this be in a separate commit/pr?

dunno / after #14096 I think PRs are more intrisicly bundled if you need changes to land things, but might be mistaken

I'll split it out as it's a bug in a code path actually used by code in master right now.

please do split it out

+1 to splitting it out. It may be worthwhile having a test case for it also since it doesn't seem to be caught by the WPT tests at all.

Fishrock123 · 2017-07-21T14:02:07Z

node.gyp

 # headers to make for a more pleasant IDE experience
+ 'src/loader/module_wrap.h',


All of these should be alphabetical. lib/internal/bootstrap_node.js', is an exception for some historical reasons.

I’m also wondering, is there any particular reason for having this be in its own directory?

@addaleax was for my own organization but we can keep with all files in a single dir if you wish.

Fishrock123 · 2017-07-21T14:03:10Z

lib/module.js

@@ -32,6 +32,7 @@ const path = require('path');
 const internalModuleReadFile = process.binding('fs').internalModuleReadFile;
 const internalModuleStat = process.binding('fs').internalModuleStat;
 const preserveSymlinks = !!process.binding('config').preserveSymlinks;
+const ESM = require('internal/loader/index');


wouldn't internal/loader be enough with an index.js?

The internal modules work differently, there is no path resolution mechanism for them

Fishrock123 · 2017-07-21T14:08:51Z

lib/module.js

+ filename = Module._resolveFilename(request, parent, isMain);
+ } catch (e) {
+ // try to keep stack
+ e.stack;


What's this for?

Is this actually a getter?

Fishrock123 · 2017-07-21T14:10:52Z

lib/module.js

+ throw e;
+ }
+ );
+ return;


Should this somehow set these, as below?

process.mainModule = module; module.id = '.';

Fishrock123 · 2017-07-21T14:12:05Z

lib/internal/safe_globals.js

+ }
+ }
+};
+const makeSafe = (unsafe, safe) => {


These should be function () {}s?

(Please don't export arrow functions unless absolutely necessary.)

Do we have a policy around that? What’s the reason?

Why? Arrow functions aren't constructable, so you'd be exporting a tighter API

@Fishrock123 can you clarify why?

We haven't had much of a defined policy about this. There have been cases where we have avoided changing existing exports from regular functions to arrow functions but given that this is a new API, I'm not sure if there's an issue here.

Fishrock123 · 2017-07-21T14:17:22Z

lib/internal/loader/Loader.js

+ try {
+ return pathToFileURL(process.cwd());
+ } catch (e) {
+ // If the current working directory no longer exists.


I feel like the regular modular loader may explode in this case?

We should keep the behavior consistent whatever it is.

The regular loader doesn't explode. There are actually test cases for that. Preserving the original behavior is what I'm trying to do.

Fishrock123 · 2017-07-21T14:18:18Z

lib/internal/loader/Loader.js

+class Loader {
+ constructor(base = getBase()) {
+ this.moduleMap = new ModuleMap();
+ this.base = base ? new URL(`${base}`) : base;


should this throw or something if base is undefined or null?

See above, this is necessary to keep Loader construction from throwing, in a case when it's still perfectly fine to use native modules and absolute paths.

targos · 2017-07-21T14:41:02Z

If someone wants to port the existing tests, they are here: https://github.com/nodejs/sketchy-as-heck-loader-impl/tree/7e248d2b6fe8082532e15da1802b79d262a1d132/lib/internal/loader/test

TimothyGu · 2017-07-21T14:41:21Z

@Fishrock123 There are some test cases in @bmeck's sketchy-as-heck repo, I assume he has not yet had the time to port them over.

TimothyGu · 2017-07-21T14:43:08Z

TBH I'm still not sure about the whole safe_globals business. If someone can explain its necessity that would be best.

addaleax · 2017-07-21T14:54:13Z

New CI: https://ci.nodejs.org/job/node-test-commit/11293/ (edit: green)

TimothyGu · 2017-07-23T15:26:11Z

New CI: https://ci.nodejs.org/job/node-test-pull-request/9303/

A recap of my work done:

CJS now works fully with module loading
Some basic .mjs tests were imported from sketchy, but I'm not yet sure if I want to import the ones using the internal Loader class yet.
Linter now supports .mjs too
process.exit(1) instead of emitting the annoying rejected promise w/o handler warning and exiting with code 1; this can be done because the ESM loader is only used for the main script right now.

Pending the resolution from nodejs/node-eps#60, maybe we should delay the work on this a bit before the resolver algorithm is set in stone. Some good arguments from both sides.

robpalme · 2017-07-25T01:11:29Z

lib/internal/loader/Loader.js

+ return getNamespaceOf(module);
+ }
+}
+Object.setPrototypeOf(Loader.prototype, Object.create(null));


What is the reason for this line? I am curious about its effect on performance.

This is an encapsulation general thing so that it does not inherit from Object so cannot be affected by mutated globals. In line with #10135

bmeck · 2017-07-25T11:14:49Z

@TimothyGu ( #14369 (comment) )

Pending the resolution from nodejs/node-eps#60, maybe we should delay the work on this a bit before the resolver algorithm is set in stone. Some good arguments from both sides.

That PR is additive and does not remove or conflict with any features of this PR. There is no need to delay this PR unless you have concerns with conflict either now or in the future.

This follows the EPS an allows the node CLI to have ESM as an entry point. `node ./example.mjs`. A newer V8 is needed for `import()` so that is not included. `import.meta` is still in specification stage so that also is not included. PR-URL: #14369 Author: Bradley Farias <bradley.meck@gmail.com> Author: Guy Bedford <guybedford@gmail.com> Author: Jan Krems <jan.krems@groupon.com> Author: Timothy Gu <timothygu99@gmail.com> Author: Michaël Zasso <targos@protonmail.com> Author: Anna Henningsen <anna@addaleax.net> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Jeremiah Senkpiel <fishrock123@rocketmail.com>

PR-URL: #15290 Refs: #14369 Reviewed-By: Bradley Farias <bradley.meck@gmail.com> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Luigi Pinca <luigipinca@gmail.com> Reviewed-By: Colin Ihrig <cjihrig@gmail.com>

Notable Changes * build: * Snapshots are now re-enabled in V8 #14875 * console: * Implement minimal `console.group()`. #14910 * deps: * upgrade libuv to 1.14.1 #14866 * update nghttp2 to v1.25.0 #14955 * dns: * Add `verbatim` option to dns.lookup(). When true, results from the DNS resolver are passed on as-is, without the reshuffling that Node.js otherwise does that puts IPv4 addresses before IPv6 addresses. #14731 * fs: * add fs.copyFile and fs.copyFileSync which allows for more efficient copying of files. #15034 * inspector: * Enable async stack traces #13870 * module: * Add support for ESM. This is currently behind the `--experimental-modules` flag and requires the .mjs extension. `node --experimental-modules index.mjs` #14369 * napi: * implement promise #14365 * os: * Add support for CIDR notation to the output of the networkInterfaces() method. #14307 * perf_hooks: * An initial implementation of the Performance Timing API for Node.js. This is the same Performance Timing API implemented by modern browsers with a number of Node.js specific properties. The User Timing mark() and measure() APIs are implemented, as is a Node.js specific flavor of the Frame Timing for measuring event loop duration. #14680 * tls: * multiple PFX in createSecureContext [#14793](#14793) * Added new collaborators: * BridgeAR – Ruben Bridgewater PR-URL: #15308

styfle · 2017-09-13T02:00:44Z

This is super exciting, thanks for all your hard work! 🎉

I read the Interop with existing modules part of the docs and noticed CommonJS modules are exported as default which makes sense since there were no named exports in CommonJS.

This means I can use the following syntax just fine:

import fs from 'fs';
import _ from 'lodash';

But I cannot use the following syntax:

import * as fs from 'fs';
import * as _ from 'lodash';

Most importantly, I cannot use the following syntax:

import { copyFile } from 'fs';
import { map } from 'lodash';

For node core APIs that's probably okay. But for all the existing CommonJS modules on npm, this could be a problem because tools that provide tree-shaking, expect the latter syntax.

Is there a plan to support the latter syntax for importing smaller pieces of CommonJS modules?

rodoabad · 2017-09-13T12:19:55Z

You're the man @bmeck GG!

bmeck · 2017-09-13T12:59:03Z

@styfle

But I cannot use the following syntax:

import * as fs from 'fs';
import * as _ from 'lodash';

Most importantly, I cannot use the following syntax:

import { copyFile } from 'fs';
import { map } from 'lodash';

The syntax by which things are imported does not change the ModuleNamespace objects that are created in the ES spec (the result of * as fs).

We have a few problems supporting namespace creation for CJS with custom names for a few reasons and are trying to figure out an opt-in solution that module authors could do. In particular, we need to know the list of property names that module.exports contains at parse time. We cannot do this with static analysis and will require authors to list these properties. Another problem is mutation of these properties cannot be safely observed and tracked for "live binding" that ES modules do. There will need to be invariant checks that the property descriptors are not configurable/writable/containing getters when module evalutation finishes, similar to Proxy invariant checks.

As for tree shaking, it should still work the same as tools using that do not tree shake CJS modules (except https://github.com/indutny/webpack-common-shake, which should be unaffected). In real ESM there is no tree shaking so that is not an issue.

Named properties are not on the top of my TODO list since they require opt-in changes and can not act like what babel does, but if someone wants to help spec out a pragma or something for opt-in; I am willing to review/help.

This follows the EPS an allows the node CLI to have ESM as an entry point. `node ./example.mjs`. A newer V8 is needed for `import()` so that is not included. `import.meta` is still in specification stage so that also is not included. PR-URL: nodejs#14369 Author: Bradley Farias <bradley.meck@gmail.com> Author: Guy Bedford <guybedford@gmail.com> Author: Jan Krems <jan.krems@groupon.com> Author: Timothy Gu <timothygu99@gmail.com> Author: Michaël Zasso <targos@protonmail.com> Author: Anna Henningsen <anna@addaleax.net> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Jeremiah Senkpiel <fishrock123@rocketmail.com>

PR-URL: nodejs#15290 Refs: nodejs#14369 Reviewed-By: Bradley Farias <bradley.meck@gmail.com> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Luigi Pinca <luigipinca@gmail.com> Reviewed-By: Colin Ihrig <cjihrig@gmail.com>

Notable Changes * build: * Snapshots are now re-enabled in V8 nodejs#14875 * console: * Implement minimal `console.group()`. nodejs#14910 * deps: * upgrade libuv to 1.14.1 nodejs#14866 * update nghttp2 to v1.25.0 nodejs#14955 * dns: * Add `verbatim` option to dns.lookup(). When true, results from the DNS resolver are passed on as-is, without the reshuffling that Node.js otherwise does that puts IPv4 addresses before IPv6 addresses. nodejs#14731 * fs: * add fs.copyFile and fs.copyFileSync which allows for more efficient copying of files. nodejs#15034 * inspector: * Enable async stack traces nodejs#13870 * module: * Add support for ESM. This is currently behind the `--experimental-modules` flag and requires the .mjs extension. `node --experimental-modules index.mjs` nodejs#14369 * napi: * implement promise nodejs#14365 * os: * Add support for CIDR notation to the output of the networkInterfaces() method. nodejs#14307 * perf_hooks: * An initial implementation of the Performance Timing API for Node.js. This is the same Performance Timing API implemented by modern browsers with a number of Node.js specific properties. The User Timing mark() and measure() APIs are implemented, as is a Node.js specific flavor of the Frame Timing for measuring event loop duration. nodejs#14680 * tls: * multiple PFX in createSecureContext [nodejs#14793](nodejs#14793) * Added new collaborators: * BridgeAR – Ruben Bridgewater PR-URL: nodejs#15308

jdalton · 2017-09-13T17:12:37Z

@bmeck

Named properties are not on the top of my TODO list since they require opt-in changes and can not act like what babel does, but if someone wants to help spec out a pragma or something for opt-in; I am willing to review/help.

From the July 26 TC39 notes you said:

Node is not going to diverge from the browser on this topic. If "use module" is not supported by browsers, Node is not going to use "use module" either.

Are you now open to Node specific pragmas or does this mean you'd want the CJS import named exports pragma to make its way to the TC39? Is it too Node specific?

bmeck · 2017-09-14T14:51:21Z

@jdalton Node specific things within Node only CJS things (module.exports) seems fine, there is no compatibility concerns since other environments would not be affected. I would be very against it being applied to all things/files that might rely on the pragma but are intended to work without Node. Also, this idea is adding invariant checks rather than changing runtime/compiletime behavior.

PR-URL: #15290 Refs: #14369 Reviewed-By: Bradley Farias <bradley.meck@gmail.com> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Luigi Pinca <luigipinca@gmail.com> Reviewed-By: Colin Ihrig <cjihrig@gmail.com>

miraage · 2017-10-06T08:53:13Z

lib/internal/loader/Loader.js

+ }
+
+ async resolve(specifier) {
+ const request = resolveRequestUrl(this.base, specifier);


Missing await statement.

bmeck requested review from bnoordhuis, jasnell, jkrems and Fishrock123 July 19, 2017 15:17

nodejs-github-bot added the lib / src Issues and PRs related to general changes in the lib or src directory. label Jul 19, 2017

TimothyGu reviewed Jul 20, 2017

View reviewed changes

targos mentioned this pull request Jul 20, 2017

Call with V8 team, #3 nodejs/CTC#156

Closed

mcollina reviewed Jul 20, 2017

View reviewed changes

targos reviewed Jul 20, 2017

View reviewed changes

Fishrock123 self-assigned this Jul 20, 2017

TimothyGu self-assigned this Jul 21, 2017

Fishrock123 previously requested changes Jul 21, 2017

View reviewed changes

TimothyGu added module Issues and PRs related to the module subsystem. c++ Issues and PRs that require attention from people who are familiar with C++. and removed lib / src Issues and PRs related to general changes in the lib or src directory. labels Jul 23, 2017

TimothyGu force-pushed the initial-pr branch from 19f86b9 to 576bf43 Compare July 24, 2017 04:58

robpalme reviewed Jul 25, 2017

View reviewed changes

bnoordhuis mentioned this pull request Sep 10, 2017

Chrome 61 now supports ES modules natively, but why v8 6.1 does not support ES modules? #15311

Closed

TimothyGu mentioned this pull request Sep 11, 2017

test: fix single test runner regression #15329

Closed

3 tasks

demurgos mentioned this pull request Sep 13, 2017

Support .mjs output microsoft/TypeScript#18442

Closed

SMotaal mentioned this pull request Sep 28, 2017

Support .mjs output [also a TypeScript issue] TypeStrong/ts-node#436

Closed

Timer mentioned this pull request Oct 4, 2017

Support .mjs file extensions facebook/create-react-app#3237

Closed

miraage reviewed Oct 6, 2017

View reviewed changes

VitorLuizC mentioned this pull request Oct 8, 2017

Request: JS icon to .mjs extension be5invis/vscode-iconset#21

Closed

bmeck mentioned this pull request Feb 12, 2018

esm: Add support for pjson cache and main without extensions #18728

Closed

4 tasks

qnighy mentioned this pull request Nov 20, 2021

Adding an option to interpret __esModule as Babel does #40891

Closed

guybedford mentioned this pull request Jul 11, 2022

esm: prevent erroneous re-initialization of ESMLoader #43763

Merged

		# headers to make for a more pleasant IDE experience
		'src/loader/module_wrap.h',

module: Allow runMain to be ESM #14369

module: Allow runMain to be ESM #14369

Conversation

bmeck commented Jul 19, 2017 • edited Loading

Checklist

Affected core subsystem(s)

bmeck commented Jul 19, 2017

jasnell commented Jul 19, 2017

TimothyGu left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

TimothyGu Jul 26, 2017 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

TimothyGu Jul 20, 2017 • edited Loading

Choose a reason for hiding this comment

mcollina left a comment

Choose a reason for hiding this comment

targos commented Jul 20, 2017

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Fishrock123 commented Jul 20, 2017

TimothyGu commented Jul 21, 2017 • edited Loading

targos commented Jul 21, 2017

addaleax commented Jul 21, 2017

Fishrock123 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

targos commented Jul 21, 2017

TimothyGu commented Jul 21, 2017

TimothyGu commented Jul 21, 2017

addaleax commented Jul 21, 2017 • edited Loading

TimothyGu commented Jul 23, 2017

Choose a reason for hiding this comment

Choose a reason for hiding this comment

bmeck commented Jul 25, 2017

styfle commented Sep 13, 2017

rodoabad commented Sep 13, 2017

bmeck commented Sep 13, 2017

jdalton commented Sep 13, 2017 • edited Loading

bmeck commented Sep 14, 2017

bmeck commented Jul 19, 2017 •

edited

Loading

TimothyGu Jul 26, 2017 •

edited

Loading

TimothyGu Jul 20, 2017 •

edited

Loading

TimothyGu commented Jul 21, 2017 •

edited

Loading

addaleax commented Jul 21, 2017 •

edited

Loading

jdalton commented Sep 13, 2017 •

edited

Loading