summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authoralecpl <alec@alec.pl>2009-07-09 21:02:34 +0000
committeralecpl <alec@alec.pl>2009-07-09 21:02:34 +0000
commitce72e0125cf4188b8e7018672739641f30ba88fe (patch)
treecd26dda86a3ffaab954aa5b3c3af1482a6f5aa22
parent5edb5b6634b13098ee2d72078bc317f5a770a893 (diff)
- simplify 'utf8' class use, make rcube_charset_convert() 5x faster on systems without mbstring and iconv installed
-rw-r--r--program/include/main.inc60
-rw-r--r--program/lib/utf8.class.php82
2 files changed, 66 insertions, 76 deletions
diff --git a/program/include/main.inc b/program/include/main.inc
index ac2906d4a..296e13fad 100644
--- a/program/include/main.inc
+++ b/program/include/main.inc
@@ -183,9 +183,9 @@ function rcube_charset_convert($str, $from, $to=NULL)
static $mbstring_loaded = null;
static $mbstring_list = null;
static $convert_warning = false;
-
+ static $conv = null;
+
$error = false;
- $conv = null;
$to = empty($to) ? $to = strtoupper(RCMAIL_CHARSET) : rcube_parse_charset($to);
$from = rcube_parse_charset($from);
@@ -223,34 +223,29 @@ function rcube_charset_convert($str, $from, $to=NULL)
}
}
- # try to convert with custom classes
- if (class_exists('utf8'))
- $conv = new utf8();
-
- // convert string to UTF-8
+ // convert charset using bundled classes/functions
if ($to == 'UTF-8') {
if ($from == 'UTF7-IMAP') {
if ($_str = utf7_to_utf8($str))
- $str = $_str;
- else
- $error = true;
+ return $_str;
}
else if ($from == 'UTF-7') {
if ($_str = rcube_utf7_to_utf8($str))
- $str = $_str;
- else
- $error = true;
+ return $_str;
}
else if (($from == 'ISO-8859-1') && function_exists('utf8_encode')) {
- $str = utf8_encode($str);
+ return utf8_encode($str);
}
- else if ($from != 'UTF-8' && $conv) {
- $from = preg_replace(array('/^WINDOWS-*125([0-8])$/', '/^CP-/'), array('CP125\\1', 'CP'), $from);
- $conv->loadCharset($from);
- $str = $conv->strToUtf8($str);
+ else if (class_exists('utf8')) {
+ if (!$conv)
+ $conv = new utf8($from);
+ else
+ $conv->loadCharset($from);
+
+ if($_str = $conv->strToUtf8($str))
+ return $_str;
}
- else if ($from != 'UTF-8')
- $error = true;
+ $error = true;
}
// encode string for output
@@ -258,36 +253,37 @@ function rcube_charset_convert($str, $from, $to=NULL)
// @TODO: we need a function for UTF-7 (RFC2152) conversion
if ($to == 'UTF7-IMAP' || $to == 'UTF-7') {
if ($_str = utf8_to_utf7($str))
- $str = $_str;
- else
- $error = true;
+ return $_str;
}
else if ($to == 'ISO-8859-1' && function_exists('utf8_decode')) {
return utf8_decode($str);
}
- else if ($to != 'UTF-8' && $conv) {
- $to = preg_replace(array('/^WINDOWS-*125([0-8])$/', '/^CP-/'), array('CP125\\1', 'CP'), $to);
- $conv->loadCharset($to);
- return $conv->utf8ToStr($str);
- }
- else if ($to != 'UTF-8') {
- $error = true;
+ else if (class_exists('utf8')) {
+ if (!$conv)
+ $conv = new utf8($to);
+ else
+ $conv->loadCharset($from);
+
+ if ($_str = $conv->strToUtf8($str))
+ return $_str;
}
+ $error = true;
}
// report error
- if ($error && !$convert_warning){
+ if ($error && !$convert_warning) {
raise_error(array(
'code' => 500,
'type' => 'php',
'file' => __FILE__,
+ 'line' => __LINE__,
'message' => "Could not convert string from $from to $to. Make sure iconv/mbstring is installed or lib/utf8.class is available."
), true, false);
$convert_warning = true;
}
- // return UTF-8 string
+ // return UTF-8 or original string
return $str;
}
diff --git a/program/lib/utf8.class.php b/program/lib/utf8.class.php
index 2bbe63663..9f718d52d 100644
--- a/program/lib/utf8.class.php
+++ b/program/lib/utf8.class.php
@@ -37,59 +37,48 @@ Note:
// Charset maps
// Adapted to fit RoundCube
define("UTF8_MAP_DIR", "program/lib/encoding");
-$utf8_maps = array(
- "CP1250" => UTF8_MAP_DIR . "/CP1250.map",
- "CP1251" => UTF8_MAP_DIR . "/CP1251.map",
- "CP1252" => UTF8_MAP_DIR . "/CP1252.map",
- "CP1253" => UTF8_MAP_DIR . "/CP1253.map",
- "CP1254" => UTF8_MAP_DIR . "/CP1254.map",
- "CP1255" => UTF8_MAP_DIR . "/CP1255.map",
- "CP1256" => UTF8_MAP_DIR . "/CP1256.map",
- "CP1257" => UTF8_MAP_DIR . "/CP1257.map",
- "CP1258" => UTF8_MAP_DIR . "/CP1258.map",
- "ISO-8859-1" => UTF8_MAP_DIR . "/ISO-8859-1.map",
- "ISO-8859-2" => UTF8_MAP_DIR . "/ISO-8859-2.map",
- "ISO-8859-3" => UTF8_MAP_DIR . "/ISO-8859-3.map",
- "ISO-8859-4" => UTF8_MAP_DIR . "/ISO-8859-4.map",
- "ISO-8859-5" => UTF8_MAP_DIR . "/ISO-8859-5.map",
- "ISO-8859-6" => UTF8_MAP_DIR . "/ISO-8859-6.map",
- "ISO-8859-7" => UTF8_MAP_DIR . "/ISO-8859-7.map",
- "ISO-8859-8" => UTF8_MAP_DIR . "/ISO-8859-8.map",
- "ISO-8859-9" => UTF8_MAP_DIR . "/ISO-8859-9.map",
- "KOI8-R" => UTF8_MAP_DIR . "/KOI8R.map",
- "KOI8R" => UTF8_MAP_DIR . "/KOI8R.map"
- );
//Error constants
-define("ERR_OPEN_MAP_FILE","ERR_OPEN_MAP_FILE");
+define("ERR_OPEN_MAP_FILE", "ERR_OPEN_MAP_FILE");
//Class definition
-Class utf8{
+Class utf8 {
var $charset = "ISO-8859-1";
var $ascMap = array();
var $utfMap = array();
+ var $aliases = array(
+ 'KOI8-R' => 'KOI8R'
+ );
+ var $error = null;
- function __construct($charset="ISO-8859-1"){
+ function __construct($charset="ISO-8859-1") {
$this->loadCharset($charset);
}
//Load charset
- function loadCharset($charset){
- global $utf8_maps;
+ function loadCharset($charset) {
+
+ $charset = preg_replace(array('/^WINDOWS-*125([0-8])$/', '/^CP-/'), array('CP125\\1', 'CP'), $charset);
+ if (isset($aliases[$charset]))
+ $charset = $aliases[$charset];
+
+ $this->charset = $charset;
- if (!is_file($utf8_maps[$charset]))
+ if (empty($this->ascMap[$charset]))
{
- $this->onError(ERR_OPEN_MAP_FILE, "Failed to open map file for $charset");
- return;
+ $file = UTF8_MAP_DIR.'/'.$charset.'.map';
+
+ if (!is_file($file)) {
+ $this->onError(ERR_OPEN_MAP_FILE, "Failed to open map file for $charset");
+ return;
}
- if (empty($this->ascMap[$charset]))
- {
- $lines = file_get_contents($utf8_maps[$charset]);
+ $lines = file_get_contents($file);
$lines = preg_replace("/#.*$/m","",$lines);
$lines = preg_replace("/\n\n/","",$lines);
$lines = explode("\n",$lines);
+
foreach($lines as $line){
$parts = explode('0x',$line);
if(count($parts)==3){
@@ -98,37 +87,42 @@ Class utf8{
$this->ascMap[$charset][$asc]=$utf;
}
}
+
+ $this->utfMap = array_flip($this->ascMap[$charset]);
}
-
- $this->charset = $charset;
- $this->utfMap = array_flip($this->ascMap[$charset]);
}
//Error handler
function onError($err_code,$err_text){
- //print($err_code . " : " . $err_text . "<hr>\n");
- raise_error(array('code' => 500,
- 'type' => 'php',
- 'file' => __FILE__,
- 'message' => $err_text), TRUE, FALSE);
+ $this->error = $err_text;
+ return null;
}
//Translate string ($str) to UTF-8 from given charset
function strToUtf8($str){
+ if (empty($this->ascMap[$this->charset]))
+ return null;
+
$chars = unpack('C*', $str);
$cnt = count($chars);
- for($i=1;$i<=$cnt;$i++) $this->_charToUtf8($chars[$i]);
+ for($i=1; $i<=$cnt; $i++)
+ $this->_charToUtf8($chars[$i]);
+
return implode("",$chars);
}
//Translate UTF-8 string to single byte string in the given charset
function utf8ToStr($utf){
+ if (empty($this->ascMap[$this->charset]))
+ return null;
+
$chars = unpack('C*', $utf);
$cnt = count($chars);
$res = ""; //No simple way to do it in place... concatenate char by char
- for ($i=1;$i<=$cnt;$i++){
+
+ for ($i=1; $i<=$cnt; $i++)
$res .= $this->_utf8ToChar($chars, $i);
- }
+
return $res;
}