policy: implement scopes field

PR-URL: #34552
Reviewed-By: James M Snell <[email protected]>

Author: bmeck

doc/api/policy.md

@@ -196,4 +196,93 @@ module.exports = function fn(...args) {
 };
 ```
 
+### Scopes
+
+Use the `"scopes"` field of a manifest to set configuration for many resources
+at once. The `"scopes"` field works by matching resources by their segments.
+If a scope or resource includes `"cascade": true` unknown specifiers will
+be searched for in their containing scope. The containing scope for cascading
+is found by recursively reducing the resource URL by removing segments for
+[special schemes][], keeping trailing `"/"` suffixes and removing the query and
+hash fragment. This leads to the eventual reduction of the URL to its origin.
+If the URL is non-special the scope will be located by the URL's origin. If no
+scope is found for the origin or in the case of opaque origins, a protocol
+string can be used as a scope.
+
+Note, `blob:` URLs adopt their origin from the path they contain, and so a scope
+of `"blob:https://nodejs.org"` will have no effect since no URL can have an
+origin of `blob:https://nodejs.org`; URLs starting with
+`blob:https://nodejs.org/` will use `https://nodejs.org` for its origin and
+thus `https:` for its protocol scope. For opaque origin `blob:` URLs they will
+have `blob:` for their protocol scope since they do not adopt origins.
+
+#### Integrity Using Scopes
+
+Setting an integrity to `true` on a scope will set the integrity for any
+resource not found in the manifest to `true`.
+
+Setting an integrity to `null` on a scope will set the integrity for any
+resource not found in the manifest to fail matching.
+
+Not including an integrity is the same as setting the integrity to `null`.
+
+`"cascade"` for integrity checks will be ignored if `"integrity"` is explicitly
+set.
+
+The following example allows loading any file:
+
+```json
+{
+  "scopes": {
+    "file:": {
+      "integrity": true
+    }
+  }
+}
+```
+
+#### Dependency Redirection Using Scopes
+
+The following example, would allow access to `fs` for all resources within
+`./app/`:
+
+```json
+{
+  "resources": {
+    "./app/checked.js": {
+      "cascade": true,
+      "integrity": true
+    }
+  },
+  "scopes": {
+    "./app/": {
+      "dependencies": {
+        "fs": true
+      }
+    }
+  }
+}
+```
+
+The following example, would allow access to `fs` for all `data:` resources:
+
+```json
+{
+  "resources": {
+    "data:text/javascript,import('fs');": {
+      "cascade": true,
+      "integrity": true
+    }
+  },
+  "scopes": {
+    "data:": {
+      "dependencies": {
+        "fs": true
+      }
+    }
+  }
+}
+```
+
 [relative url string]: https://url.spec.whatwg.org/#relative-url-with-fragment-string
+[special schemes]: https://url.spec.whatwg.org/#special-scheme

lib/internal/modules/cjs/loader.js

@@ -46,8 +46,8 @@ const {
   SafeMap,
   String,
   StringPrototypeEndsWith,
-  StringPrototypeIndexOf,
   StringPrototypeLastIndexOf,
+  StringPrototypeIndexOf,
   StringPrototypeMatch,
   StringPrototypeSlice,
   StringPrototypeStartsWith,
@@ -81,8 +81,9 @@ const { getOptionValue } = require('internal/options');
 const enableSourceMaps = getOptionValue('--enable-source-maps');
 const preserveSymlinks = getOptionValue('--preserve-symlinks');
 const preserveSymlinksMain = getOptionValue('--preserve-symlinks-main');
-const manifest = getOptionValue('--experimental-policy') ?
-  require('internal/process/policy').manifest :
+// Do not eagerly grab .manifest, it may be in TDZ
+const policy = getOptionValue('--experimental-policy') ?
+  require('internal/process/policy') :
   null;
 const { compileFunction } = internalBinding('contextify');
 
@@ -1029,10 +1030,10 @@ function wrapSafe(filename, content, cjsModuleInstance) {
 Module.prototype._compile = function(content, filename) {
   let moduleURL;
   let redirects;
-  if (manifest) {
+  if (policy?.manifest) {
     moduleURL = pathToFileURL(filename);
-    redirects = manifest.getRedirector(moduleURL);
-    manifest.assertIntegrity(moduleURL, content);
+    redirects = policy.manifest.getDependencyMapper(moduleURL);
+    policy.manifest.assertIntegrity(moduleURL, content);
   }
 
   maybeCacheSourceMap(filename, content, this);
@@ -1101,9 +1102,9 @@ Module._extensions['.js'] = function(module, filename) {
 Module._extensions['.json'] = function(module, filename) {
   const content = fs.readFileSync(filename, 'utf8');
 
-  if (manifest) {
+  if (policy?.manifest) {
     const moduleURL = pathToFileURL(filename);
-    manifest.assertIntegrity(moduleURL, content);
+    policy.manifest.assertIntegrity(moduleURL, content);
   }
 
   try {
@@ -1117,10 +1118,10 @@ Module._extensions['.json'] = function(module, filename) {
 
 // Native extension for .node
 Module._extensions['.node'] = function(module, filename) {
-  if (manifest) {
+  if (policy?.manifest) {
     const content = fs.readFileSync(filename);
     const moduleURL = pathToFileURL(filename);
-    manifest.assertIntegrity(moduleURL, content);
+    policy.manifest.assertIntegrity(moduleURL, content);
   }
   // Be aware this doesn't use `content`
   return process.dlopen(module, path.toNamespacedPath(filename));

lib/internal/modules/esm/get_source.js

@@ -1,8 +1,9 @@
 'use strict';
 
 const { getOptionValue } = require('internal/options');
-const manifest = getOptionValue('--experimental-policy') ?
-  require('internal/process/policy').manifest :
+// Do not eagerly grab .manifest, it may be in TDZ
+const policy = getOptionValue('--experimental-policy') ?
+  require('internal/process/policy') :
   null;
 
 const { Buffer } = require('buffer');
@@ -33,8 +34,8 @@ async function defaultGetSource(url, { format } = {}, defaultGetSource) {
   } else {
     throw new ERR_INVALID_URL_SCHEME(['file', 'data']);
   }
-  if (manifest) {
-    manifest.assertIntegrity(parsed, source);
+  if (policy?.manifest) {
+    policy.manifest.assertIntegrity(parsed, source);
   }
   return { source };
 }

lib/internal/modules/esm/resolve.js

@@ -30,8 +30,9 @@ const {
   Stats,
 } = require('fs');
 const { getOptionValue } = require('internal/options');
-const manifest = getOptionValue('--experimental-policy') ?
-  require('internal/process/policy').manifest :
+// Do not eagerly grab .manifest, it may be in TDZ
+const policy = getOptionValue('--experimental-policy') ?
+  require('internal/process/policy') :
   null;
 const { sep, relative } = require('path');
 const preserveSymlinks = getOptionValue('--preserve-symlinks');
@@ -714,8 +715,8 @@ function resolveAsCommonJS(specifier, parentURL) {
 
 function defaultResolve(specifier, context = {}, defaultResolveUnused) {
   let { parentURL, conditions } = context;
-  if (manifest) {
-    const redirects = manifest.getRedirector(parentURL);
+  if (parentURL && policy?.manifest) {
+    const redirects = policy.manifest.getDependencyMapper(parentURL);
     if (redirects) {
       const { resolve, reaction } = redirects;
       const destination = resolve(specifier, new SafeSet(conditions));

lib/internal/modules/package_json_reader.js

@@ -7,6 +7,8 @@ const { toNamespacedPath } = require('path');
 
 const cache = new SafeMap();
 
+let manifest;
+
 /**
  *
  * @param {string} jsonPath
@@ -22,10 +24,12 @@ function read(jsonPath) {
   const result = { string, containsKeys };
   const { getOptionValue } = require('internal/options');
   if (string !== undefined) {
-    const manifest = getOptionValue('--experimental-policy') ?
-      require('internal/process/policy').manifest :
-      null;
-    if (manifest) {
+    if (manifest === undefined) {
+      manifest = getOptionValue('--experimental-policy') ?
+        require('internal/process/policy').manifest :
+        null;
+    }
+    if (manifest !== null) {
       const jsonURL = pathToFileURL(jsonPath);
       manifest.assertIntegrity(jsonURL, string);
     }