From 6d5d19fa091b4b497f4d3374fd3a84503690e210 Mon Sep 17 00:00:00 2001
From: Peter Rotich <peter@osticket.com>
Date: Fri, 23 Jan 2015 22:21:56 +0000
Subject: [PATCH] charset: Normalize charset
This pull request adds a cleanup util for bogus and invalid charsets, mostly
added by a nameless company out of Redmond, WA.
---
include/class.charset.php | 41 +++++++++++++++++++++++++++++++++++++++
include/class.format.php | 9 +++------
2 files changed, 44 insertions(+), 6 deletions(-)
create mode 100644 include/class.charset.php
diff --git a/include/class.charset.php b/include/class.charset.php
new file mode 100644
index 000000000..5a4f28ad7
--- /dev/null
+++ b/include/class.charset.php
@@ -0,0 +1,41 @@
+<?php
+/*********************************************************************
+ class.charset.php
+
+ Charset util class
+
+ Copyright (c) 2015 osTicket
+ http://www.osticket.com
+
+ Released under the GNU General Public License WITHOUT ANY WARRANTY.
+ See LICENSE.TXT for details.
+
+ vim: expandtab sw=4 ts=4 sts=4:
+**********************************************************************/
+
+class Charset {
+
+ // Cleanup invalid charsets
+ // Thanks in part to https://github.com/mikel/mail/commit/88457e
+ function normalize($charset) {
+
+ $match = array();
+ switch (true) {
+ // Windows charsets - force correct format
+ case preg_match('`^Windows-?(\d+)$`', $charset, $match):
+ return 'Windows-'.$match[1];
+ // ks_c_5601-1987: Korean alias for cp949
+ case preg_match('`^ks_c_5601-1987`', $charset):
+ return 'cp949';
+ // Incorrect, bogus, ambiguous or empty charsets
+ // ISO-8859-1 is assumed
+ case preg_match('`^(default|x-user-defined|iso|us-ascii)`', $charset):
+ case preg_match('`^\s*$`', $charset):
+ return 'ISO-8859-1';
+ }
+
+ // Hmmmm
+ return $charset;
+ }
+}
+?>
diff --git a/include/class.format.php b/include/class.format.php
index a124a577f..aaff16b8b 100644
--- a/include/class.format.php
+++ b/include/class.format.php
@@ -14,6 +14,7 @@
vim: expandtab sw=4 ts=4 sts=4:
**********************************************************************/
+include_once INCLUDE_DIR.'class.charset.php';
class Format {
@@ -47,12 +48,8 @@ class Format {
if (!$charset && function_exists('mb_detect_encoding'))
$charset = mb_detect_encoding($text);
- // Cleanup - incorrect, bogus, or ambiguous charsets
- // ISO-8859-1 is assumed for empty charset.
- if (!$charset || in_array(strtolower(trim($charset)),
- array('default','x-user-defined','iso','us-ascii')))
- $charset = 'ISO-8859-1';
-
+ // Normalize bogus or ambiguous charsets
+ $charset = Charset::normalize(trim($charset));
$original = $text;
if (function_exists('iconv'))
$text = iconv($charset, $encoding.'//IGNORE', $text);
--
GitLab