Add support for serializers

README.md

@@ -7,13 +7,15 @@ Object redaction with whitelist and blacklist. Blacklist items have higher prior
 3. `options` _(Object)_: An object with optional options.
 
     `options.replacement` _(Function)_: A function that allows customizing the replacement value (default implementation is `--REDACTED--`).
-
+
+    `options.serializers` _(List[Object])_: A list with serializers to apply. Each serializers must contain two properties: `path` (path for the value to be serialized, must be a `string`) and `serializer` (function to be called on the path's value).
+
     `options.trim` _(Boolean)_: A flag that enables trimming all redacted values, saving their keys to a `__redacted__` list (default value is `false`).
 
 ### Example
 
 ```js
-const anonymizer = require('@uphold/anonymizer');
+const { anonymizer } = require('@uphold/anonymizer');
 const whitelist = ['foo.key', 'foo.depth.*', 'bar.*', 'toAnonymize.baz', 'toAnonymizeSuperString'];
 const blacklist = ['foo.depth.innerBlacklist', 'toAnonymize.*'];
 const anonymize = anonymizer({ blacklist, whitelist });
@@ -39,6 +41,51 @@ anonymize(data);
 // }
 ```
 
+#### Example using serializers
+
+```js
+const { anonymizer } = require('@uphold/anonymizer');
+const whitelist = ['foo.key', 'foo.depth.*', 'bar.*', 'toAnonymize.baz'];
+const blacklist = ['foo.depth.innerBlacklist'];
+const serializers = [
+  { path: 'foo.key', serializer: () => 'biz' },
+  { path: 'toAnonymize', serializer: () => ({ baz: 'baz' }) }
+]
+const anonymize = anonymizer({ blacklist, whitelist });
+
+const data = {
+  foo: { key: 'public', another: 'bar', depth: { bar: 10, innerBlacklist: 11 } },
+  bar: { foo: 1, bar: 2 },
+  toAnonymize: {}
+};
+
+// {
+//   foo: {
+//     key: 'biz',
+//     another: '--REDACTED--',
+//     depth: { bar: 10, innerBlacklist: '--REDACTED--' }
+//   },
+//   bar: { foo: 1, bar: 2 },
+//   toAnonymize: { baz: 'baz' }
+// }
+```
+
+### Default serializers
+
+The introduction of serializers also added the possibility of using serializer functions exported by our module. The list of default serializers is presented below:
+- error
+
+#### Example
+
+```js
+const { anonymizer, defaultSerializers } = require('@uphold/anonymizer');
+const serializers = [
+  { path: 'foo', serializer: defaultSerializers.error }
+];
+
+const anonymize = anonymizer({ whitelist }, { serializers });
+```
+
 ## Releasing a new version
 
 - Diff the current code with the latest tag and make sure the output is expected.

package.json

@@ -22,6 +22,8 @@
   "dependencies": {
     "json-stringify-safe": "^5.0.1",
     "lodash.get": "^4.4.2",
+    "lodash.set": "^4.3.2",
+    "serialize-error": "^5.0.0",
     "traverse": "^0.6.6"
   },
   "devDependencies": {

src/index.js

@@ -4,7 +4,9 @@
  * Module dependencies.
  */
 
+const { serializeError } = require('serialize-error');
 const get = require('lodash.get');
+const set = require('lodash.set');
 const stringify = require('json-stringify-safe');
 const traverse = require('traverse');
 
@@ -14,27 +16,88 @@ const traverse = require('traverse');
 
 const DEFAULT_REPLACEMENT = '--REDACTED--';
 
+/**
+ * Validate serializers.
+ */
+
+function validateSerializers(serializers) {
+  serializers.map(({ path, serializer }) => {
+    if (typeof serializer !== 'function') {
+      throw new TypeError(`Invalid serializer for \`${path}\` path: must be a function`);
+    }
+  });
+}
+
+/**
+ * Compute Mutations
+ */
+
+function computeSerializedChanges(values, serializers) {
+  const changes = {};
+
+  for (const { path, serializer } of serializers) {
+    const value = get(values, path);
+
+    if (value === undefined) {
+      continue;
+    }
+
+    try {
+      changes[path] = serializer(value);
+    } catch (error) {
+      changes[path] = `Anonymize ERROR: Error while applying ${path} serializer`;
+    }
+  }
+
+  return changes;
+}
+
 /**
  * Module exports.
+ *
+ * Example:
+ *
+ * anonymizer({
+ *   whitelist: ['foo']
+ * }, {
+ * replacement,
+ * serializers: [
+ *   { path: 'foo.bar', serializer: () => {} }
+ *  ]
+ * })
  */
 
-module.exports = (
+module.exports.anonymizer = (
   { blacklist = [], whitelist = [] } = {},
-  { replacement = () => DEFAULT_REPLACEMENT, trim = false } = {}
+  { replacement = () => DEFAULT_REPLACEMENT, serializers = [], trim = false } = {}
 ) => {
   const whitelistTerms = whitelist.join('|');
   const whitelistPaths = new RegExp(`^(${whitelistTerms.replace(/\./g, '\\.').replace(/\*/g, '.*')})$`, 'i');
   const blacklistTerms = blacklist.join('|');
   const blacklistPaths = new RegExp(`^(${blacklistTerms.replace(/\./g, '\\.').replace(/\*/g, '.*')})$`, 'i');
 
+  validateSerializers(serializers);
+
   return values => {
     if (!(values instanceof Object)) {
       return values;
     }
 
     const blacklistedKeys = new Set();
+    // JSON.parse(stringify(values)) builds an object copy that isn't an
+    // exact replication of the initial input. It destroys some relevant
+    // data that can't be lost. However, it can't be swapped for another
+    // solution due to its performance and because it can also handle
+    // classes correctly. Moreover, the `computeSerializedChanges()`
+    // also requires a copy to avoid updates by reference and in order
+    // to avoid two copies, we build an object with the result of applying
+    // the serializers to their respective paths. After we perform the copy,
+    // the serializers output is merged into the copy.
+    const changes = computeSerializedChanges(values, serializers);
     const obj = JSON.parse(stringify(values));
 
+    Object.entries(changes).forEach(([path, change]) => set(obj, path, change));
+
     traverse(obj).forEach(function() {
       const path = this.path.join('.');
       const isBuffer = Buffer.isBuffer(get(values, path));
@@ -78,3 +141,11 @@ module.exports = (
     return obj;
   };
 };
+
+/**
+ * Module exports defaultSerializers.
+ */
+
+module.exports.defaultSerializers = {
+  error: serializeError
+};

test/src/benchmark/samples.js

@@ -0,0 +1,65 @@
+'use strict';
+
+/**
+ * Function to generate a sample object.
+ */
+
+module.exports.generateObjectSample = ({ depth = 6, branches = 2, leafValue = () => 'foobar', leafs = 32 }) => {
+  const sample = {};
+
+  if (depth === 0) {
+    for (let leaf = 0; leaf < leafs; leaf++) {
+      sample[`leaf-${leaf}`] = leafValue();
+    }
+
+    return sample;
+  }
+
+  for (let branch = 0; branch < branches; branch++) {
+    sample[`branch-${branch}`] = this.generateObjectSample({ branches, depth: depth - 1, leafs });
+  }
+
+  return sample;
+};
+
+/**
+ * Function to generate all the paths contained in a sample generated with `generateObjectSample`.
+ */
+
+module.exports.generateObjectSamplesPaths = ({ depth = 6, branches = 2, leafs = 32, path = '' }) => {
+  let paths = [];
+
+  if (depth === 0) {
+    for (let leaf = 0; leaf < leafs; leaf++) {
+      paths.push(`${path}.leaf-${leaf}`);
+    }
+
+    return paths;
+  }
+
+  for (let branch = 0; branch < branches; branch++) {
+    const childPathString = path === '' ? `branch-${branch}` : `${path}.branch-${branch}`;
+    const childPaths = this.generateObjectSamplesPaths({ branches, depth: depth - 1, leafs, path: childPathString });
+
+    paths = paths.concat(childPaths);
+  }
+
+  return paths;
+};
+
+module.exports.samples = {
+  // Sample with 0 props.
+  SAMPLE_0x: {},
+  // Sample with 2048 props.
+  SAMPLE_1x: this.generateObjectSample({ depth: 6 }),
+  // Sample with 4096 props.
+  SAMPLE_2x: this.generateObjectSample({ depth: 7 }),
+  // Sample with 8192 props.
+  SAMPLE_4x: this.generateObjectSample({ depth: 8 }),
+  // Sample with 16384 props.
+  SAMPLE_8x: this.generateObjectSample({ depth: 9 }),
+  // Sample with 32768 props.
+  SAMPLE_16x: this.generateObjectSample({ depth: 10 }),
+  // Sample with 65536 props.
+  SAMPLE_32x: this.generateObjectSample({ depth: 11 })
+};