vm: sync-ify SourceTextModule linkage

Split `module.link(linker)` into two synchronous step `sourceTextModule.linkRequestedModules()` and `sourceTextModule.instantiate()`. This allows creating vm modules and resolving the dependencies in a complete synchronous procedure. This also makes `syntheticModule.link()` redundant. The link step for a SyntheticModule is no-op and is already taken care in the constructor by initializing the binding slots with the given export names.

Author: legendecas

doc/api/errors.md

@@ -2266,6 +2266,13 @@ The V8 platform used by this instance of Node.js does not support creating
 Workers. This is caused by lack of embedder support for Workers. In particular,
 this error will not occur with standard builds of Node.js.
 
+<a id="ERR_MODULE_LINK_MISMATCH"></a>
+
+### `ERR_MODULE_LINK_MISMATCH`
+
+A module can not be linked due to a mismatch of the requested modules, and the
+list of given dependency modules.
+
 <a id="ERR_MODULE_NOT_FOUND"></a>
 
 ### `ERR_MODULE_NOT_FOUND`

doc/api/vm.md

@@ -419,9 +419,7 @@ class that closely mirrors [Module Record][]s as defined in the ECMAScript
 specification.
 
 Unlike `vm.Script` however, every `vm.Module` object is bound to a context from
-its creation. Operations on `vm.Module` objects are intrinsically asynchronous,
-in contrast with the synchronous nature of `vm.Script` objects. The use of
-'async' functions can help with manipulating `vm.Module` objects.
+its creation.
 
 Using a `vm.Module` object requires three distinct steps: creation/parsing,
 linking, and evaluation. These three steps are illustrated in the following
@@ -449,7 +447,7 @@ const contextifiedObject = vm.createContext({
 // Here, we attempt to obtain the default export from the module "foo", and
 // put it into local binding "secret".
 
-const bar = new vm.SourceTextModule(`
+const rootModule = new vm.SourceTextModule(`
  import s from 'foo';
  s;
  print(s);
@@ -459,47 +457,56 @@ const bar = new vm.SourceTextModule(`
 //
 // "Link" the imported dependencies of this Module to it.
 //
-// The provided linking callback (the "linker") accepts two arguments: the
-// parent module (`bar` in this case) and the string that is the specifier of
-// the imported module. The callback is expected to return a Module that
-// corresponds to the provided specifier, with certain requirements documented
-// in `module.link()`.
-//
-// If linking has not started for the returned Module, the same linker
-// callback will be called on the returned Module.
+// Obtain the requested dependencies of a SourceTextModule by
+// `sourceTextModule.moduleRequests` and resolve them.
 //
 // Even top-level Modules without dependencies must be explicitly linked. The
-// callback provided would never be called, however.
-//
-// The link() method returns a Promise that will be resolved when all the
-// Promises returned by the linker resolve.
+// array passed to `sourceTextModule.linkRequestedModules(modules)` can be
+// empty, however.
 //
-// Note: This is a contrived example in that the linker function creates a new
+// Note: This is a contrived example in that the linker creates a new
 // "foo" module every time it is called. In a full-fledged module system, a
 // cache would probably be used to avoid duplicated modules.
 
-async function linker(specifier, referencingModule) {
- if (specifier === 'foo') {
- return new vm.SourceTextModule(`
- // The "secret" variable refers to the global variable we added to
- // "contextifiedObject" when creating the context.
- export default secret;
- `, { context: referencingModule.context });
+const moduleMap = new Map([
+ ['root', rootModule],
+]);
 
- // Using `contextifiedObject` instead of `referencingModule.context`
- // here would work as well.
- }
- throw new Error(`Unable to resolve dependency: ${specifier}`);
+function linker(module) {
+ const requestedModules = module.moduleRequests.map((request) => {
+ // In a full-fledged module system, the linker would resolve the
+ // module with the module cache key `[specifier, attributes]`.
+ // In this example, we just use the specifier as the key.
+ const specifier = request.specifier;
+
+ let requestedModule = moduleMap.get(specifier);
+ if (requestedModule === undefined) {
+ requestedModule = new vm.SourceTextModule(`
+ // The "secret" variable refers to the global variable we added to
+ // "contextifiedObject" when creating the context.
+ export default secret;
+ `, { context: referencingModule.context });
+ moduleMap.set(specifier, linkedModule);
+ // Resolve the dependencies of the new module as well.
+ linker(requestedModule);
+ }
+
+ return requestedModule;
+ });
+
+ module.linkRequestedModules(requestedModules);
 }
-await bar.link(linker);
+
+linker(rootModule);
+rootModule.instantiate();
 
 // Step 3
 //
 // Evaluate the Module. The evaluate() method returns a promise which will
 // resolve after the module has finished evaluating.
 
 // Prints 42.
-await bar.evaluate();
+await rootModule.evaluate();
 ```
 
 ```cjs
@@ -521,7 +528,7 @@ const contextifiedObject = vm.createContext({
  // Here, we attempt to obtain the default export from the module "foo", and
  // put it into local binding "secret".
 
- const bar = new vm.SourceTextModule(`
+ const rootModule = new vm.SourceTextModule(`
  import s from 'foo';
  s;
  print(s);
@@ -531,47 +538,55 @@ const contextifiedObject = vm.createContext({
  //
  // "Link" the imported dependencies of this Module to it.
  //
- // The provided linking callback (the "linker") accepts two arguments: the
- // parent module (`bar` in this case) and the string that is the specifier of
- // the imported module. The callback is expected to return a Module that
- // corresponds to the provided specifier, with certain requirements documented
- // in `module.link()`.
- //
- // If linking has not started for the returned Module, the same linker
- // callback will be called on the returned Module.
+ // Obtain the requested dependencies of a SourceTextModule by
+ // `sourceTextModule.moduleRequests` and resolve them.
  //
  // Even top-level Modules without dependencies must be explicitly linked. The
- // callback provided would never be called, however.
- //
- // The link() method returns a Promise that will be resolved when all the
- // Promises returned by the linker resolve.
+ // array passed to `sourceTextModule.linkRequestedModules(modules)` can be
+ // empty, however.
  //
- // Note: This is a contrived example in that the linker function creates a new
+ // Note: This is a contrived example in that the linker creates a new
  // "foo" module every time it is called. In a full-fledged module system, a
  // cache would probably be used to avoid duplicated modules.
 
- async function linker(specifier, referencingModule) {
- if (specifier === 'foo') {
- return new vm.SourceTextModule(`
- // The "secret" variable refers to the global variable we added to
- // "contextifiedObject" when creating the context.
- export default secret;
- `, { context: referencingModule.context });
+ const moduleMap = new Map([
+ ['root', rootModule],
+ ]);
+
+ function linker(module) {
+ const requestedModules = module.moduleRequests.map((request) => {
+ // In a full-fledged module system, the linker would resolve the
+ // module with the module cache key `[specifier, attributes]`.
+ // In this example, we just use the specifier as the key.
+ const specifier = request.specifier;
+
+ let requestedModule = moduleMap.get(specifier);
+ if (requestedModule === undefined) {
+ requestedModule = new vm.SourceTextModule(`
+ // The "secret" variable refers to the global variable we added to
+ // "contextifiedObject" when creating the context.
+ export default secret;
+ `, { context: referencingModule.context });
+ moduleMap.set(specifier, linkedModule);
+ // Resolve the dependencies of the new module as well.
+ linker(requestedModule);
+ }
+
+ return requestedModule;
+ });
 
- // Using `contextifiedObject` instead of `referencingModule.context`
- // here would work as well.
- }
- throw new Error(`Unable to resolve dependency: ${specifier}`);
+ module.linkRequestedModules(requestedModules);
  }
- await bar.link(linker);
+
+ linker(rootModule);
 
  // Step 3
  //
  // Evaluate the Module. The evaluate() method returns a promise which will
  // resolve after the module has finished evaluating.
 
  // Prints 42.
- await bar.evaluate();
+ await rootModule.evaluate();
 })();
 ```
 
@@ -635,6 +650,9 @@ changes:
  former name is still provided for backward compatibility.
 -->
 
+> Stability: 0 - Deprecated: Use [`sourceTextModule.linkRequestedModules(modules)`][] and
+> [`sourceTextModule.instantiate()`][] instead.
+
 * `linker` {Function}
  * `specifier` {string} The specifier of the requested module:
  ```mjs
@@ -898,6 +916,53 @@ to disallow any changes to it.
 Corresponds to the `[[RequestedModules]]` field of [Cyclic Module Record][]s in
 the ECMAScript specification.
 
+### `sourceTextModule.instantiate()`
+
+<!-- YAML
+added: REPLACEME
+-->
+
+* Returns: {void}
+
+Instantiate the module with the linked requested modules.
+
+This resolves the imported bindings of the module, including re-exported
+binding names.
+
+If the requested modules include cyclic dependencies, the
+[`sourceTextModule.linkRequestedModules(modules)`][] method must be called on all
+modules in the cycle before calling this method.
+
+### `sourceTextModule.linkRequestedModules(modules)`
+
+<!-- YAML
+added: REPLACEME
+-->
+
+* `modules` {vm.Module\[]} Array of `vm.Module` objects that this module depends on.
+ The order of the modules in the array is the order of
+ [`sourceTextModule.moduleRequests`][].
+* Returns: {void}
+
+Link module dependencies. This method must be called before evaluation, and
+can only be called once per module.
+
+The order of the `modules` array should respect the order of
+[`sourceTextModule.moduleRequests`][].
+
+If the module has no dependencies, the `modules` array can be empty, or skip this
+method call.
+
+Composing `sourceTextModule.moduleRequests` and `sourceTextModule.link()`,
+this acts similar to [HostLoadImportedModule][] and [FinishLoadingImportedModule][]
+abstract operations in the ECMAScript specification, respectively.
+
+It's up to the creator of the `SourceTextModule` to determine if the resolution
+of the dependencies is synchronous or asynchronous.
+
+After each module in the `modules` array is linked, call
+[`sourceTextModule.instantiate()`][].
+
 ### `sourceTextModule.moduleRequests`
 
 <!-- YAML
@@ -1017,14 +1082,17 @@ the module to access information outside the specified `context`. Use
 added:
  - v13.0.0
  - v12.16.0
+changes:
+ - version: REPLACEME
+ pr-url: https://github.com/nodejs/node/pull/XXXXX
+ description: No longer need to call `syntheticModule.link()` before
+ calling this method.
 -->
 
 * `name` {string} Name of the export to set.
 * `value` {any} The value to set the export to.
 
-This method is used after the module is linked to set the values of exports. If
-it is called before the module is linked, an [`ERR_VM_MODULE_STATUS`][] error
-will be thrown.
+This method sets the module export binding slots with the given value.
 
 ```mjs
 import vm from 'node:vm';
@@ -1033,7 +1101,6 @@ const m = new vm.SyntheticModule(['x'], () => {
  m.setExport('x', 1);
 });
 
-await m.link(() => {});
 await m.evaluate();
 
 assert.strictEqual(m.namespace.x, 1);
@@ -1045,7 +1112,6 @@ const vm = require('node:vm');
  const m = new vm.SyntheticModule(['x'], () => {
  m.setExport('x', 1);
  });
- await m.link(() => {});
  await m.evaluate();
  assert.strictEqual(m.namespace.x, 1);
 })();
@@ -2037,7 +2103,9 @@ const { Script, SyntheticModule } = require('node:vm');
 [Cyclic Module Record]: https://tc39.es/ecma262/#sec-cyclic-module-records
 [ECMAScript Module Loader]: esm.md#modules-ecmascript-modules
 [Evaluate() concrete method]: https://tc39.es/ecma262/#sec-moduleevaluation
+[FinishLoadingImportedModule]: https://tc39.es/ecma262/#sec-FinishLoadingImportedModule
 [GetModuleNamespace]: https://tc39.es/ecma262/#sec-getmodulenamespace
+[HostLoadImportedModule]: https://tc39.es/ecma262/#sec-HostLoadImportedModule
 [HostResolveImportedModule]: https://tc39.es/ecma262/#sec-hostresolveimportedmodule
 [ImportDeclaration]: https://tc39.es/ecma262/#prod-ImportDeclaration
 [Link() concrete method]: https://tc39.es/ecma262/#sec-moduledeclarationlinking
@@ -2049,13 +2117,14 @@ const { Script, SyntheticModule } = require('node:vm');
 [WithClause]: https://tc39.es/ecma262/#prod-WithClause
 [`ERR_VM_DYNAMIC_IMPORT_CALLBACK_MISSING_FLAG`]: errors.md#err_vm_dynamic_import_callback_missing_flag
 [`ERR_VM_DYNAMIC_IMPORT_CALLBACK_MISSING`]: errors.md#err_vm_dynamic_import_callback_missing
-[`ERR_VM_MODULE_STATUS`]: errors.md#err_vm_module_status
 [`Error`]: errors.md#class-error
 [`URL`]: url.md#class-url
 [`eval()`]: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/eval
 [`optionsExpression`]: https://tc39.es/proposal-import-attributes/#sec-evaluate-import-call
 [`script.runInContext()`]: #scriptrunincontextcontextifiedobject-options
 [`script.runInThisContext()`]: #scriptruninthiscontextoptions
+[`sourceTextModule.instantiate()`]: #sourcetextmoduleinstantiate
+[`sourceTextModule.linkRequestedModules(modules)`]: #sourcetextmodulelinkrequestedmodulesmodules
 [`sourceTextModule.moduleRequests`]: #sourcetextmodulemodulerequests
 [`url.origin`]: url.md#urlorigin
 [`vm.compileFunction()`]: #vmcompilefunctioncode-params-options

lib/internal/errors.js

@@ -1597,6 +1597,9 @@ E('ERR_MISSING_ARGS',
  return `${msg} must be specified`;
  }, TypeError);
 E('ERR_MISSING_OPTION', '%s is required', TypeError);
+E('ERR_MODULE_LINK_MISMATCH', function(message) {
+ return message;
+}, TypeError);
 E('ERR_MODULE_NOT_FOUND', function(path, base, exactUrl) {
  if (exactUrl) {
  lazyInternalUtil().setOwnProperty(this, 'url', `${exactUrl}`);