esm: fix base URL for network imports

bmeck · danielleadams · commit 166eb782f9d3 · 2022-04-23T22:47:23.000-04:00
PR-URL: #42131 Reviewed-By: Geoffrey Booth <webadmin@geoffreybooth.com> Reviewed-By: Guy Bedford <guybedford@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
diff --git a/lib/internal/modules/cjs/helpers.js b/lib/internal/modules/cjs/helpers.js
@@ -193,6 +193,11 @@ function addBuiltinLibsToObject(object, dummyModuleName) {
   });
 }
 
+/**
+ *
+ * @param {string | URL} referrer
+ * @returns {string}
+ */
 function normalizeReferrerURL(referrer) {
   if (typeof referrer === 'string' && path.isAbsolute(referrer)) {
     return pathToFileURL(referrer).href;
diff --git a/lib/internal/modules/cjs/loader.js b/lib/internal/modules/cjs/loader.js
@@ -1023,7 +1023,8 @@ function wrapSafe(filename, content, cjsModuleInstance) {
       displayErrors: true,
       importModuleDynamically: async (specifier, _, importAssertions) => {
         const loader = asyncESM.esmLoader;
-        return loader.import(specifier, normalizeReferrerURL(filename),
+        return loader.import(specifier,
+                             loader.getBaseURL(normalizeReferrerURL(filename)),
                              importAssertions);
       },
     });
@@ -1039,7 +1040,8 @@ function wrapSafe(filename, content, cjsModuleInstance) {
       filename,
       importModuleDynamically(specifier, _, importAssertions) {
         const loader = asyncESM.esmLoader;
-        return loader.import(specifier, normalizeReferrerURL(filename),
+        return loader.import(specifier,
+                             loader.getBaseURL(normalizeReferrerURL(filename)),
                              importAssertions);
       },
     });
diff --git a/lib/internal/modules/esm/initialize_import_meta.js b/lib/internal/modules/esm/initialize_import_meta.js
@@ -3,12 +3,9 @@
 const { getOptionValue } = require('internal/options');
 const experimentalImportMetaResolve =
   getOptionValue('--experimental-import-meta-resolve');
-const { fetchModule } = require('internal/modules/esm/fetch_module');
-const { URL } = require('internal/url');
 const {
   PromisePrototypeThen,
   PromiseReject,
-  StringPrototypeStartsWith,
 } = primordials;
 const asyncESM = require('internal/process/esm_loader');
 
@@ -24,6 +21,10 @@ function createImportMetaResolve(defaultParentUrl) {
   };
 }
 
+/**
+ * @param {object} meta
+ * @param {{url: string}} context
+ */
 function initializeImportMeta(meta, context) {
   let url = context.url;
 
@@ -32,14 +33,7 @@ function initializeImportMeta(meta, context) {
     meta.resolve = createImportMetaResolve(url);
   }
 
-  if (
-    StringPrototypeStartsWith(url, 'http:') ||
-    StringPrototypeStartsWith(url, 'https:')
-  ) {
-    // The request & response have already settled, so they are in fetchModule's
-    // cache, in which case, fetchModule returns immediately and synchronously
-    url = fetchModule(new URL(url), context).resolvedHREF;
-  }
+  url = asyncESM.esmLoader.getBaseURL(url);
 
   meta.url = url;
 }
diff --git a/lib/internal/modules/esm/loader.js b/lib/internal/modules/esm/loader.js
@@ -17,11 +17,13 @@ const {
   RegExpPrototypeExec,
   SafeArrayIterator,
   SafeWeakMap,
+  StringPrototypeStartsWith,
   globalThis,
 } = primordials;
 const { MessageChannel } = require('internal/worker/io');
 
 const {
+  ERR_INTERNAL_ASSERTION,
   ERR_INVALID_ARG_TYPE,
   ERR_INVALID_ARG_VALUE,
   ERR_INVALID_RETURN_PROPERTY_VALUE,
@@ -47,6 +49,9 @@ const { defaultLoad } = require('internal/modules/esm/load');
 const { translators } = require(
   'internal/modules/esm/translators');
 const { getOptionValue } = require('internal/options');
+const {
+  fetchModule,
+} = require('internal/modules/esm/fetch_module');
 
 /**
  * An ESMLoader instance is used as the main entry point for loading ES modules.
@@ -209,7 +214,9 @@ class ESMLoader {
       const module = new ModuleWrap(url, undefined, source, 0, 0);
       callbackMap.set(module, {
         importModuleDynamically: (specifier, { url }, importAssertions) => {
-          return this.import(specifier, url, importAssertions);
+          return this.import(specifier,
+                             this.getBaseURL(url),
+                             importAssertions);
         }
       });
 
@@ -225,6 +232,43 @@ class ESMLoader {
     };
   }
 
+  /**
+   * Returns the url to use for the resolution of a given cache key url
+   * These are not guaranteed to be the same.
+   *
+   * In WHATWG HTTP spec for ESM the cache key is the non-I/O bound
+   * synchronous resolution using only string operations
+   *   ~= resolveImportMap(new URL(specifier, importerHREF))
+   *
+   * The url used for subsequent resolution is the response URL after
+   * all redirects have been resolved.
+   *
+   * https://example.com/foo redirecting to https://example.com/bar
+   * would have a cache key of https://example.com/foo and baseURL
+   * of https://example.com/bar
+   *
+   * MUST BE SYNCHRONOUS for import.meta initialization
+   * MUST BE CALLED AFTER receiving the url body due to I/O
+   * @param {string} url
+   * @returns {string}
+   */
+  getBaseURL(url) {
+    if (
+      StringPrototypeStartsWith(url, 'http:') ||
+      StringPrototypeStartsWith(url, 'https:')
+    ) {
+      // The request & response have already settled, so they are in
+      // fetchModule's cache, in which case, fetchModule returns
+      // immediately and synchronously
+      url = fetchModule(new URL(url), { parentURL: url }).resolvedHREF;
+      // This should only occur if the module hasn't been fetched yet
+      if (typeof url !== 'string') {
+        throw new ERR_INTERNAL_ASSERTION(`Base url for module ${url} not loaded.`);
+      }
+    }
+    return url;
+  }
+
   /**
    * Get a (possibly still pending) module job from the cache,
    * or create one and return its Promise.
diff --git a/lib/internal/modules/esm/module_job.js b/lib/internal/modules/esm/module_job.js
@@ -76,7 +76,8 @@ class ModuleJob {
       // these `link` callbacks depending on each other.
       const dependencyJobs = [];
       const promises = this.module.link(async (specifier, assertions) => {
-        const jobPromise = this.loader.getModuleJob(specifier, url, assertions);
+        const baseURL = this.loader.getBaseURL(url);
+        const jobPromise = this.loader.getModuleJob(specifier, baseURL, assertions);
         ArrayPrototypePush(dependencyJobs, jobPromise);
         const job = await jobPromise;
         return job.modulePromise;
diff --git a/lib/internal/modules/esm/translators.js b/lib/internal/modules/esm/translators.js
@@ -103,7 +103,9 @@ function errPath(url) {
 }
 
 async function importModuleDynamically(specifier, { url }, assertions) {
-  return asyncESM.esmLoader.import(specifier, url, assertions);
+  return asyncESM.esmLoader.import(specifier,
+                                   asyncESM.esmLoader.getBaseURL(url),
+                                   assertions);
 }
 
 // Strategy for loading a standard JavaScript module.
diff --git a/test/es-module/test-http-imports.mjs b/test/es-module/test-http-imports.mjs
@@ -111,6 +111,21 @@ for (const { protocol, createServer } of [
     assert.strict.notEqual(redirectedNS.default, ns.default);
     assert.strict.equal(redirectedNS.url, url.href);
 
+    // Redirects have the same import.meta.url but different cache
+    // entry on Web
+    const relativeAfterRedirect = new URL(url.href + 'foo/index.js');
+    const redirected = new URL(url.href + 'bar/index.js');
+    redirected.searchParams.set('body', 'export let relativeDepURL = (await import("./baz.js")).url');
+    relativeAfterRedirect.searchParams.set('redirect', JSON.stringify({
+      status: 302,
+      location: redirected.href
+    }));
+    const relativeAfterRedirectedNS = await import(relativeAfterRedirect.href);
+    assert.strict.equal(
+      relativeAfterRedirectedNS.relativeDepURL,
+      url.href + 'bar/baz.js'
+    );
+
     const crossProtocolRedirect = new URL(url.href);
     crossProtocolRedirect.searchParams.set('redirect', JSON.stringify({
       status: 302,

Original file line number	Diff line number	Diff line change
`@@ -103,7 +103,9 @@ function errPath(url) {`
`103`	`103`	`}`
`104`	`104`
`105`	`105`	`async function importModuleDynamically(specifier, { url }, assertions) {`
`106`		`- return asyncESM.esmLoader.import(specifier, url, assertions);`
	`106`	`+ return asyncESM.esmLoader.import(specifier,`
	`107`	`+ asyncESM.esmLoader.getBaseURL(url),`
	`108`	`+ assertions);`
`107`	`109`	`}`
`108`	`110`
`109`	`111`	`// Strategy for loading a standard JavaScript module.`