+/**
+ * Tests if string contains 8bit symbols.
+ *
+ * If charset is not set, function defaults to default_charset.
+ * $default_charset global must be set correctly if $charset is
+ * not used.
+ * @param string $string tested string
+ * @param string $charset charset used in a string
+ * @return bool true if 8bit symbols are detected
+ * @since 1.5.1 and 1.4.4
+ */
+function sq_is8bit($string,$charset='') {
+ global $default_charset;
+
+ if ($charset=='') $charset=$default_charset;
+
+ /**
+ * Don't use \240 in ranges. Sometimes RH 7.2 doesn't like it.
+ * Don't use \200-\237 for iso-8859-x charsets. This range
+ * stores control symbols in those charsets.
+ * Use preg_match instead of ereg in order to avoid problems
+ * with mbstring overloading
+ */
+ if (preg_match("/^iso-8859/i",$charset)) {
+ $needle='/\240|[\241-\377]/';
+ } else {
+ $needle='/[\200-\237]|\240|[\241-\377]/';
+ }
+ return preg_match("$needle",$string);
+}
+
+/**
+ * Replacement of mb_list_encodings function
+ *
+ * This function provides replacement for function that is available only
+ * in php 5.x. Function does not test all mbstring encodings. Only the ones
+ * that might be used in SM translations.
+ *
+ * Supported strings are stored in session in order to reduce number of
+ * mb_internal_encoding function calls.
+ *
+ * If you want to test all mbstring encodings - fill $list_of_encodings
+ * array.
+ * @return array list of encodings supported by php mbstring extension
+ * @since 1.5.1
+ */
+function sq_mb_list_encodings() {
+ if (! function_exists('mb_internal_encoding'))
+ return array();
+
+ // don't try to test encodings, if they are already stored in session
+ if (sqgetGlobalVar('mb_supported_encodings',$mb_supported_encodings,SQ_SESSION))
+ return $mb_supported_encodings;
+
+ // save original encoding
+ $orig_encoding=mb_internal_encoding();
+
+ $list_of_encoding=array(
+ 'pass',
+ 'auto',
+ 'ascii',
+ 'jis',
+ 'utf-8',
+ 'sjis',
+ 'euc-jp',
+ 'iso-8859-1',
+ 'iso-8859-2',
+ 'iso-8859-7',
+ 'iso-8859-9',
+ 'iso-8859-15',
+ 'koi8-r',
+ 'koi8-u',
+ 'big5',
+ 'gb2312',
+ 'gb18030',
+ 'windows-1251',
+ 'windows-1255',
+ 'windows-1256',
+ 'tis-620',
+ 'iso-2022-jp',
+ 'euc-kr',
+ 'utf7-imap');
+
+ $supported_encodings=array();
+
+ foreach ($list_of_encoding as $encoding) {
+ // try setting encodings. suppress warning messages
+ if (@mb_internal_encoding($encoding))
+ $supported_encodings[]=$encoding;
+ }
+
+ // restore original encoding
+ mb_internal_encoding($orig_encoding);
+
+ // register list in session
+ sqsession_register($supported_encodings,'mb_supported_encodings');
+
+ return $supported_encodings;
+}
+
+/**
+ * Function returns number of characters in string.
+ *
+ * Returned number might be different from number of bytes in string,
+ * if $charset is multibyte charset. Detection depends on mbstring
+ * functions. If mbstring does not support tested multibyte charset,
+ * vanilla string length function is used.
+ * @param string $str string
+ * @param string $charset charset
+ * @since 1.5.1
+ * @return integer number of characters in string
+ */
+function sq_strlen($str, $charset=''){
+ // default option
+ if ($charset=='') return strlen($str);
+
+ // use automatic charset detection, if function call asks for it
+ if ($charset=='auto') {
+ global $default_charset;
+ set_my_charset();
+ $charset=$default_charset;
+ }
+
+ // lowercase charset name
+ $charset=strtolower($charset);
+
+ // Use mbstring only with listed charsets
+ $aList_of_mb_charsets=array('utf-8','big5','gb2312','gb18030','euc-jp','euc-cn','euc-tw','euc-kr');
+
+ // calculate string length according to charset
+ if (in_array($charset,$aList_of_mb_charsets) && in_array($charset,sq_mb_list_encodings())) {
+ $real_length = mb_strlen($str,$charset);
+ } else {
+ // own strlen detection code is removed because missing strpos,
+ // strtoupper and substr implementations break string wrapping.
+ $real_length=strlen($str);
+ }
+ return $real_length;
+}
+
+/**
+ * string padding with multibyte support
+ *
+ * @link http://www.php.net/str_pad
+ * @param string $string original string
+ * @param integer $width padded string width
+ * @param string $pad padding symbols
+ * @param integer $padtype padding type
+ * (internal php defines, see str_pad() description)
+ * @param string $charset charset used in original string
+ * @return string padded string
+ */
+function sq_str_pad($string, $width, $pad, $padtype, $charset='') {
+
+ $charset = strtolower($charset);
+ $padded_string = '';
+
+ switch ($charset) {
+ case 'utf-8':
+ case 'big5':
+ case 'gb2312':
+ case 'euc-kr':
+ /*
+ * all multibyte charsets try to increase width value by
+ * adding difference between number of bytes and real length
+ */
+ $width = $width - sq_strlen($string,$charset) + strlen($string);
+ default:
+ $padded_string=str_pad($string,$width,$pad,$padtype);
+ }
+ return $padded_string;
+}
+
+/**
+ * Wrapper that is used to switch between vanilla and multibyte substr
+ * functions.
+ * @param string $string
+ * @param integer $start
+ * @param integer $length
+ * @param string $charset
+ * @return string
+ * @since 1.5.1
+ * @link http://www.php.net/substr
+ * @link http://www.php.net/mb_substr
+ */
+function sq_substr($string,$start,$length,$charset='auto') {
+ // use automatic charset detection, if function call asks for it
+ if ($charset=='auto') {
+ global $default_charset;
+ set_my_charset();
+ $charset=$default_charset;
+ }
+ $charset = strtolower($charset);
+ if (function_exists('mb_internal_encoding') &&
+ in_array($charset,sq_mb_list_encodings())) {
+ return mb_substr($string,$start,$length,$charset);
+ }
+ // TODO: add mbstring independent code
+
+ // use vanilla string functions as last option
+ return substr($string,$start,$length);
+}
+
+/**
+ * Wrapper that is used to switch between vanilla and multibyte strpos
+ * functions.
+ * @param string $haystack
+ * @param mixed $needle
+ * @param integer $offset
+ * @param string $charset
+ * @return string
+ * @since 1.5.1
+ * @link http://www.php.net/strpos
+ * @link http://www.php.net/mb_strpos
+ */
+function sq_strpos($haystack,$needle,$offset,$charset='auto') {
+ // use automatic charset detection, if function call asks for it
+ if ($charset=='auto') {
+ global $default_charset;
+ set_my_charset();
+ $charset=$default_charset;
+ }
+ $charset = strtolower($charset);
+ if (function_exists('mb_internal_encoding') &&
+ in_array($charset,sq_mb_list_encodings())) {
+ return mb_strpos($haystack,$needle,$offset,$charset);
+ }
+ // TODO: add mbstring independent code
+
+ // use vanilla string functions as last option
+ return strpos($haystack,$needle,$offset);
+}
+
+/**
+ * Wrapper that is used to switch between vanilla and multibyte strtoupper
+ * functions.
+ * @param string $string
+ * @param string $charset
+ * @return string
+ * @since 1.5.1
+ * @link http://www.php.net/strtoupper
+ * @link http://www.php.net/mb_strtoupper
+ */
+function sq_strtoupper($string,$charset='auto') {
+ // use automatic charset detection, if function call asks for it
+ if ($charset=='auto') {
+ global $default_charset;
+ set_my_charset();
+ $charset=$default_charset;
+ }
+ $charset = strtolower($charset);
+ if (function_exists('mb_strtoupper') &&
+ in_array($charset,sq_mb_list_encodings())) {
+ return mb_strtoupper($string,$charset);
+ }
+ // TODO: add mbstring independent code
+
+ // use vanilla string functions as last option
+ return strtoupper($string);
+}