Allow appending to wordlist file

README.md

@@ -79,6 +79,9 @@ Optionally you can specify *wordCacheFile* to control the location and name of t
 
 To a custom instance you can pass an optional boolean parameter *fetch*. When false, the cached wordlist will be preferred but falls back to fetching if it could not be loaded.
 
+You can also mix different RSS sources into the same wordlist file with the `appendWordlist` parameter, increasing entropy, or regularly rebuild your cache file from time to time to feed it with new words. Altought having more words is always desirable,
+you can limit the number of items in the wordlist with the `limitWordlist` configuration parameter. When using append, wordlist is shuffled before being limited to the desired quantity.
+
 ```php
 include 'PasswordGenerator.php';
 
@@ -90,6 +93,16 @@ $gen = PasswordGenerator::EN();
 echo 'Password 1: ', $gen->generate();
 echo 'Password 2: ', $gen->generate();
 echo 'Password 3: ', $gen->generate();
+
+// Append NYTimes feed to the specified wordlist, limiting to 7000 items max
+$gen = new PasswordGenerator([
+    'wordCacheFile'  => 'mywords.json',
+    'url'            => 'https://rss.nytimes.com/services/xml/rss/nyt/World.xml',
+    'minLength'      => 3,
+    'maxLength'      => 8,
+    'appendWordlist' => true,
+    'limitWordlist'  => 7000,
+]);
 ```
 
 ### Caching
@@ -155,6 +168,15 @@ When creating an instance of PasswordGenerator you can provide the following par
 
   Controls if and how many HTTP redirects should be followed during URL access. Defaults to *2*. To prevent following redirects set its value to *0*.
 
+* **appendWordlist**
+
+  If true, the fetched wordlist will be appended to the current one. If false, wordlist is generated from scratch.
+
+* **limitWordlist**
+
+  Int value declaring the maximum words the wordlist must contain. Useful when using appendWordList.
+
+
 ## License
 
 This project is licensed under the MIT License - see the [LICENSE.md](LICENSE.md) file for details.

example/PasswordGenerator.example.php

@@ -51,3 +51,15 @@
 
 // generate a password with custom pattern
 echo $gen->generate('wiwsw'), "\n";
+
+// Append NYTimes feed to the last wordlist, limiting to 700 items max
+$gen = new PasswordGenerator([
+    'wordCacheFile'  => 'mywords.json',
+    'url'            => 'https://rss.nytimes.com/services/xml/rss/nyt/World.xml',
+    'minLength'      => 3,
+    'maxLength'      => 8,
+    'appendWordlist' => true,
+    'limitWordlist'  => 700,
+]);
+
+echo $gen->generate('wiws'), "\n";

src/PasswordGenerator.php

@@ -34,10 +34,10 @@ class PasswordGenerator
     /** @var string $url Url of word ressource */
     private string $url;
 
-    /** @var int $minLength Minimum length of a password */
+    /** @var int $minLength Minimum length of a word to be considered for the list */
     private int $minLength;
 
-    /** @var int $maxLength Maximum length of a password */
+    /** @var int $maxLength Maximum length of a word to be considered for the list */
     private int $maxLength;
 
     /** @var array $wordList Cache of the words */
@@ -52,6 +52,12 @@ class PasswordGenerator
     /** @var bool $verbose */
     private bool $verbose;
 
+    /** @var bool $appendWordlist If true the fetched words will be appended to the cached wordlist, rather than replacing the current file */
+    private bool $appendWordlist = false;
+
+    /** @var int $limitWordlist Limit the max number of words in the wordlist files, defaults to no limit */
+    private int $limitWordlist = 0;
+
     /**
      * Creates an instance of the password generator. You can pass an optional
      * array with values that should override the default values for the keys:
@@ -65,6 +71,10 @@ class PasswordGenerator
      *   <dd>The maxinum length of characters a word must have.</dd>
      *   <dt>wordCacheFile</dt>
      *   <dd>Filename of the cache file.</dd>
+     *   <dt>appendWordlist</dt>
+     *   <dd>If true, the fetched wordlist will be appended to the current one. If false, wordlist is generated from scratch.</dd>
+     *   <dt>limitWordlist</dt>
+     *   <dd>Int value declaring the maximum words the wordlist must contain. Useful when using appendWordList.</dd>
      *   <dt>httpRedirects</dt>
      *   <dd>Number of maximum HTTP redirects the script will follow.</dd>
      * </dl>
@@ -143,9 +153,21 @@ private function populate_wordlist(): void
                     }
                 }
             }
-            $this->wordList = array_keys($wordlist);
+            if ((bool)$this->appendWordlist) {
+                $this->wordList = array_unique($this->wordList + array_keys($wordlist));
+            } else {
+                $this->wordList = array_keys($wordlist);
+            }
+            $wordlistLimit = (int)$this->limitWordlist;
+            if ($wordlistLimit > 0) {
+                // Shuffle the array to mix existing and new words, if in append mode
+                if ((bool)$this->appendWordlist) {
+                    shuffle($this->wordList);
+                }
+                $this->wordList = array_slice($this->wordList, 0, $wordlistLimit);
+            }
 
-            if (count($wordlist) > 0) {
+            if (count($this->wordList) > 0) {
                 $this->save_wordlist();
             }
         } catch (Exception $e) {