From 6c97852799d409f7125010a1c4e5ba9d0295565f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Guillaume=20Souli=C3=A9?= <guillaume.soulie@gmail.com>
Date: Fri, 23 Aug 2019 14:17:18 +0200
Subject: [PATCH] Introduce splitRegex option.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Issue:
Using /\b/ to split the text is limited to languages using only the 63
characters:

```
a b c d e f g h i j k l m n o p q r s t u v w x y z
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
0 1 2 3 4 5 6 7 8 9 _
```

cf https://stackoverflow.com/a/2449892

For example, in French, the string `je suis français` will be split into
`["je", " ", "suis", " ", "fran", "ç", "ais"]` which won't allow to
perform the bad-words cleaning.

Therefore I added an option `splitRegex` which allow to overwrite the
regex used to split.
---
 lib/badwords.js |  6 ++++--
 test/options.js | 24 ++++++++++++++++++++++++
 2 files changed, 28 insertions(+), 2 deletions(-)
 create mode 100644 test/options.js

diff --git a/lib/badwords.js b/lib/badwords.js
index 9091708..3990c41 100644
--- a/lib/badwords.js
+++ b/lib/badwords.js
@@ -12,11 +12,13 @@ class Filter {
    * @param {string} options.placeHolder - Character used to replace profane words.
    * @param {string} options.regex - Regular expression used to sanitize words before comparing them to blacklist.
    * @param {string} options.replaceRegex - Regular expression used to replace profane words with placeHolder.
+   * @param {string} options.splitRegex - Regular expression used to split a string into words.
    */
   constructor(options = {}) {
     Object.assign(this, {
       list: options.emptyList && [] || Array.prototype.concat.apply(localList, [baseList, options.list || []]),
       exclude: options.exclude || [],
+      splitRegex: options.splitRegex || /\b/,
       placeHolder: options.placeHolder || '*',
       regex: options.regex || /[^a-zA-Z0-9|\$|\@]|\^/g,
       replaceRegex: options.replaceRegex || /\w/g
@@ -51,9 +53,9 @@ class Filter {
    * @param {string} string - Sentence to filter.
    */
   clean(string) {
-    return string.split(/\b/).map((word) => {
+    return string.split(this.splitRegex).map((word) => {
       return this.isProfane(word) ? this.replaceWord(word) : word;
-    }).join('');
+    }).join(this.splitRegex.exec(string)[0]);
   }
 
   /**
diff --git a/test/options.js b/test/options.js
new file mode 100644
index 0000000..cada98f
--- /dev/null
+++ b/test/options.js
@@ -0,0 +1,24 @@
+require('assert');
+var Filter = require('../lib/badwords.js'),
+assert = require('better-assert');
+
+describe('options', function() {
+  describe('split regex', function() {
+
+    it('default value', function() {
+      filter = new Filter();
+      filter.addWords('français');
+      assert(filter.clean('fucking asshole') == '******* *******');
+      assert(filter.clean('mot en français') == 'mot en français');
+    });
+
+    it('override value', function() {
+      filter = new Filter({splitRegex: / /});
+      filter.addWords('français');
+      assert(filter.clean('fucking asshole') == '******* *******');
+      assert(filter.clean('mot en français') == 'mot en *******');
+    });
+
+
+  });
+});