From d38d4b1fa85a8c876415e08b57626bd2b9656973 Mon Sep 17 00:00:00 2001
From: Peter Johnson <insomnia@rcpt.at>
Date: Tue, 26 Jun 2018 13:48:50 +0200
Subject: [PATCH] feat(text_sanitizer): trim whitespace and quotation marks
 from a range of natural languages

---
 sanitizer/_text.js           |  6 ++-
 test/unit/sanitizer/_text.js | 74 +++++++++++++++++++++++++++++++++++-
 2 files changed, 77 insertions(+), 3 deletions(-)

diff --git a/sanitizer/_text.js b/sanitizer/_text.js
index 1df63411..cb4dc737 100644
--- a/sanitizer/_text.js
+++ b/sanitizer/_text.js
@@ -1,6 +1,9 @@
 const check = require('check-types');
 const _ = require('lodash');
 
+// ref: https://en.wikipedia.org/wiki/Quotation_mark
+const QUOTES = `"'«»‘’‚‛“”„‟‹›⹂「」『』〝〞〟﹁﹂﹃﹄＂＇｢｣`;
+
 // validate texts, convert types and apply defaults
 function _sanitize( raw, clean ){
 
@@ -14,8 +17,7 @@ function _sanitize( raw, clean ){
     messages.errors.push('invalid param \'text\': text length, must be >0');
 
   } else {
-    clean.text = raw.text;
-
+    clean.text = _.trim( _.trim( raw.text ), QUOTES );
   }
 
   return messages;
diff --git a/test/unit/sanitizer/_text.js b/test/unit/sanitizer/_text.js
index 1fde352b..7a35fbbe 100644
--- a/test/unit/sanitizer/_text.js
+++ b/test/unit/sanitizer/_text.js
@@ -37,7 +37,7 @@ module.exports.tests.text_parser = function(test, common) {
       const messages = sanitizer.sanitize(raw, clean);
 
       t.deepEquals(clean, expected_clean);
-      t.deepEquals(messages.errors, ['invalid param \'text\': text length, must be >0'], 'no errors');
+      t.deepEquals(messages.errors, ['invalid param \'text\': text length, must be >0']);
       t.deepEquals(messages.warnings, [], 'no warnings');
 
     });
@@ -46,6 +46,78 @@ module.exports.tests.text_parser = function(test, common) {
 
   });
 
+  test('should trim whitespace', t => {
+    var clean = {};
+    var raw = { text: ` test \n ` };
+    const messages = sanitizer.sanitize(raw, clean);
+
+    t.equals(clean.text, 'test');
+    t.deepEquals(messages.errors, [], 'no errors');
+    t.deepEquals(messages.warnings, [], 'no warnings');
+
+    t.end();
+  });
+
+  test('should trim double quotes', t => {
+    var clean = {};
+    var raw = { text: ` "test" \n ` };
+    const messages = sanitizer.sanitize(raw, clean);
+
+    t.equals(clean.text, 'test');
+    t.deepEquals(messages.errors, [], 'no errors');
+    t.deepEquals(messages.warnings, [], 'no warnings');
+
+    t.end();
+  });
+
+  test('should trim single quotes', t => {
+    var clean = {};
+    var raw = { text: ` 'test' \n ` };
+    const messages = sanitizer.sanitize(raw, clean);
+
+    t.equals(clean.text, 'test');
+    t.deepEquals(messages.errors, [], 'no errors');
+    t.deepEquals(messages.warnings, [], 'no warnings');
+
+    t.end();
+  });
+
+  test('should trim German quotes', t => {
+    var clean = {};
+    var raw = { text: ` „test“ \n ` };
+    const messages = sanitizer.sanitize(raw, clean);
+
+    t.equals(clean.text, 'test');
+    t.deepEquals(messages.errors, [], 'no errors');
+    t.deepEquals(messages.warnings, [], 'no warnings');
+
+    t.end();
+  });
+
+  test('should trim guillemets', t => {
+    var clean = {};
+    var raw = { text: ` »test« \n ` };
+    const messages = sanitizer.sanitize(raw, clean);
+
+    t.equals(clean.text, 'test');
+    t.deepEquals(messages.errors, [], 'no errors');
+    t.deepEquals(messages.warnings, [], 'no warnings');
+
+    t.end();
+  });
+
+  test('should trim Chinese quotes', t => {
+    var clean = {};
+    var raw = { text: ` ﹁「test」﹂ \n ` };
+    const messages = sanitizer.sanitize(raw, clean);
+
+    t.equals(clean.text, 'test');
+    t.deepEquals(messages.errors, [], 'no errors');
+    t.deepEquals(messages.warnings, [], 'no warnings');
+
+    t.end();
+  });
+
   test('return an array of expected parameters in object form for validation', (t) => {
     const expected = [{ name: 'text' }];
     const validParameters = sanitizer.expected();