X-Git-Url: https://vcs.fsf.org/?a=blobdiff_plain;f=functions%2Fstrings.php;h=49addede63768e15b53622b685a6e9bb7a716551;hb=480feeacdef8ef7790e1b940901135520f4ea39d;hp=27dfa8a1f96cecbe28cfc061369296c20ab1bf5a;hpb=c9d61baf815700ed32427bf46fed3a5cc093216a;p=squirrelmail.git
diff --git a/functions/strings.php b/functions/strings.php
index 27dfa8a1..49addede 100644
--- a/functions/strings.php
+++ b/functions/strings.php
@@ -3,11 +3,11 @@
/**
* strings.php
*
- * Copyright (c) 1999-2004 The SquirrelMail Project Team
+ * Copyright (c) 1999-2005 The SquirrelMail Project Team
* Licensed under the GNU GPL. For full terms see the file COPYING.
*
* This code provides various string manipulation functions that are
- * used by the rest of the Squirrelmail code.
+ * used by the rest of the SquirrelMail code.
*
* @version $Id$
* @package squirrelmail
@@ -74,6 +74,35 @@ function sqMakeNewLine (&$str, $citeLevel, &$column) {
}
}
+/**
+ * Checks for spaces in strings - only used if PHP doesn't have native ctype support
+ *
+ * @author Tomas Kuliavas
+ *
+ * You might be able to rewrite the function by adding short evaluation form.
+ *
+ * possible problems:
+ * - iso-2022-xx charsets - hex 20 might be part of other symbol. I might
+ * be wrong. 0x20 is not used in iso-2022-jp. I haven't checked iso-2022-kr
+ * and iso-2022-cn mappings.
+ *
+ * - no-break space ( ) - it is 8bit symbol, that depends on charset.
+ * there are at least three different charset groups that have nbsp in
+ * different places.
+ *
+ * I don't see any charset/nbsp options in php ctype either.
+ *
+ * @param string $string tested string
+ * @return bool true when only whitespace symbols are present in test string
+ */
+function sm_ctype_space($string) {
+ if ( preg_match('/^[\x09-\x0D]|^\x20/', $string) || $string=='') {
+ return true;
+ } else {
+ return false;
+ }
+}
+
/**
* Wraps text at $wrap characters. While sqWordWrap takes
* a single line of text and wraps it, this function works
@@ -87,6 +116,13 @@ function sqMakeNewLine (&$str, $citeLevel, &$column) {
* @return string the wrapped text
*/
function &sqBodyWrap (&$body, $wrap) {
+ //check for ctype support, and fake it if it doesn't exist
+ if (!function_exists('ctype_space')) {
+ function ctype_space ($string) {
+ return sm_ctype_space($string);
+ }
+ }
+
// the newly wrapped text
$outString = '';
// current column since the last newline in the outstring
@@ -109,7 +145,9 @@ function &sqBodyWrap (&$body, $wrap) {
// skip over any spaces interleaved among the cite markers
while (($pos < $length) && ($body{$pos} == ' ')) {
+
$pos++;
+
}
if ($pos >= $length) {
break;
@@ -119,7 +157,7 @@ function &sqBodyWrap (&$body, $wrap) {
// special case: if this is a blank line then maintain it
// (i.e. try to preserve original paragraph breaks)
// unless they occur at the very beginning of the text
- if (($body{$pos} == "\n") && (strlen($outString) != 0)) {
+ if (($body{$pos} == "\n" ) && (strlen($outString) != 0)) {
$outStringLast = $outString{strlen($outString) - 1};
if ($outStringLast != "\n") {
$outString .= "\n";
@@ -171,7 +209,10 @@ function &sqBodyWrap (&$body, $wrap) {
$pos = $nextNewline + 1;
continue;
}
-
+ /**
+ * Set this to false to stop appending short strings to previous lines
+ */
+ $smartwrap = true;
// inner loop, (obviously) handles wrapping up to
// the next newline
while ($pos < $nextNewline) {
@@ -179,20 +220,59 @@ function &sqBodyWrap (&$body, $wrap) {
while (($pos < $nextNewline) && (ctype_space ($body{$pos}))) {
$pos++;
}
-
// if this is a short line then just append it and continue outer loop
- if (($outStringCol + $nextNewline - $pos) <= ($wrap - $citeLevel - 1)) {
+ if (($outStringCol + $nextNewline - $pos) <= ($wrap - $citeLevel - 1) ) {
// if this is the final line in the input string then include
// any trailing newlines
+ // echo substr($body,$pos,$wrap). "
";
if (($nextNewline + 1 == $length) && ($body{$nextNewline} == "\n")) {
$nextNewline++;
}
// trim trailing spaces
$lastRealChar = $nextNewline;
- while (($lastRealChar > $pos) && (ctype_space ($body{$lastRealChar}))) {
+ while (($lastRealChar > $pos && $lastRealChar < $length) && (ctype_space ($body{$lastRealChar}))) {
$lastRealChar--;
}
+ // decide if appending the short string is what we want
+ if (($nextNewline < $length && $body{$nextNewline} == "\n") &&
+ isset($lastRealChar)) {
+ $mypos = $pos;
+ //check the first word:
+ while (($mypos < $length) && ($body{$mypos} == '>')) {
+ $mypos++;
+ // skip over any spaces interleaved among the cite markers
+ while (($mypos < $length) && ($body{$mypos} == ' ')) {
+ $mypos++;
+ }
+ }
+/*
+ $ldnspacecnt = 0;
+ if ($mypos == $nextNewline+1) {
+ while (($mypos < $length) && ($body{$mypos} == ' ')) {
+ $ldnspacecnt++;
+ }
+ }
+*/
+
+ $firstword = substr($body,$mypos,strpos($body,' ',$mypos) - $mypos);
+ //if ($dowrap || $ldnspacecnt > 1 || ($firstword && (
+ if (!$smartwrap || $firstword && (
+ $firstword{0} == '-' ||
+ $firstword{0} == '+' ||
+ $firstword{0} == '*' ||
+ $firstword{0} == strtoupper($firstword{0}) ||
+ strpos($firstword,':'))) {
+ $outString .= substr($body,$pos,($lastRealChar - $pos+1));
+ $outStringCol += ($lastRealChar - $pos);
+ sqMakeNewLine($outString,$citeLevel,$outStringCol);
+ $nextNewline++;
+ $pos = $nextNewline;
+ $outStringCol--;
+ continue;
+ }
+
+ }
$outString .= substr ($body, $pos, ($lastRealChar - $pos + 1));
$outStringCol += ($lastRealChar - $pos);
@@ -207,7 +287,7 @@ function &sqBodyWrap (&$body, $wrap) {
// our current line is already too long, break immediately
// and restart outer loop
if ($eol <= $pos) {
- sqMakeNewLine ($outString, $citeLeve, $outStringCol);
+ sqMakeNewLine ($outString, $citeLevel, $outStringCol);
continue;
}
@@ -270,20 +350,22 @@ function &sqBodyWrap (&$body, $wrap) {
* Has a problem with special HTML characters, so call this before
* you do character translation.
*
- * Specifically, ' comes up as 5 characters instead of 1.
+ * Specifically, ' comes up as 5 characters instead of 1.
* This should not add newlines to the end of lines.
*
* @param string line the line of text to wrap, by ref
* @param int wrap the maximum line lenth
+ * @param string charset name of charset used in $line string. Available since v.1.5.1.
* @return void
*/
-function sqWordWrap(&$line, $wrap) {
+function sqWordWrap(&$line, $wrap, $charset='') {
global $languages, $squirrelmail_language;
+ // Use custom wrapping function, if translation provides it
if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
- function_exists($languages[$squirrelmail_language]['XTRA_CODE'])) {
+ function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_wordwrap')) {
if (mb_detect_encoding($line) != 'ASCII') {
- $line = $languages[$squirrelmail_language]['XTRA_CODE']('wordwrap', $line, $wrap);
+ $line = call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_wordwrap', $line, $wrap);
return;
}
}
@@ -302,9 +384,9 @@ function sqWordWrap(&$line, $wrap) {
while ($i < count($words)) {
/* Force one word to be on a line (minimum) */
$line .= $words[$i];
- $line_len = strlen($beginning_spaces) + strlen($words[$i]) + 2;
+ $line_len = strlen($beginning_spaces) + sq_strlen($words[$i],$charset) + 2;
if (isset($words[$i + 1]))
- $line_len += strlen($words[$i + 1]);
+ $line_len += sq_strlen($words[$i + 1],$charset);
$i ++;
/* Add more words (as long as they fit) */
@@ -312,7 +394,7 @@ function sqWordWrap(&$line, $wrap) {
$line .= ' ' . $words[$i];
$i++;
if (isset($words[$i]))
- $line_len += strlen($words[$i]) + 1;
+ $line_len += sq_strlen($words[$i],$charset) + 1;
else
$line_len += 1;
}
@@ -561,7 +643,7 @@ function sq_mt_seed($Val) {
$Val *= -1;
}
- if ($Val = 0) {
+ if ($Val == 0) {
return;
}
@@ -659,7 +741,7 @@ function show_readable_size($bytes) {
}
/**
- * Generates a random string from the caracter set you pass in
+ * Generates a random string from the character set you pass in
*
* @param int size the size of the string to generate
* @param string chars a string containing the characters to use
@@ -756,7 +838,7 @@ function makeComposeLink($url, $text = null, $target='')
// build the compose in new window link...
- // if javascript is on, use onClick event to handle it
+ // if javascript is on, use onclick event to handle it
if($javascript_on) {
sqgetGlobalVar('base_uri', $base_uri, SQ_SESSION);
return ''. $text.'';
@@ -785,6 +867,19 @@ function sm_print_r() {
foreach(func_get_args() as $var) {
print_r($var);
echo "\n";
+ // php has get_class_methods function that can print class methods
+ if (is_object($var)) {
+ // get class methods if $var is object
+ $aMethods=get_class_methods(get_class($var));
+ // make sure that $aMethods is array and array is not empty
+ if (is_array($aMethods) && $aMethods!=array()) {
+ echo "Object methods:\n";
+ foreach($aMethods as $method) {
+ echo '* ' . $method . "\n";
+ }
+ }
+ echo "\n";
+ }
}
$buffer = ob_get_contents(); // Grab the print_r output
ob_end_clean(); // Silently discard the output & stop buffering
@@ -874,7 +969,10 @@ function sq_get_html_translation_table($table,$quote_style=ENT_COMPAT,$charset='
* sq_htmlentities
*
* Convert all applicable characters to HTML entities.
- * Minimal php requirement - v.4.0.5
+ * Minimal php requirement - v.4.0.5.
+ *
+ * Function is designed for people that want to use full power of htmlentities() in
+ * i18n environment.
*
* @param string $string string that has to be sanitized
* @param integer $quote_style quote encoding style. Possible values (without quotes):
@@ -893,5 +991,243 @@ function sq_htmlentities($string,$quote_style=ENT_COMPAT,$charset='us-ascii') {
return str_replace(array_keys($sq_html_ent_table),array_values($sq_html_ent_table),$string);
}
+/**
+ * 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',
+ '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. Currently only utf-8 charset is
+ * supported.
+ * @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);
+
+ // set initial returned length number
+ $real_length=0;
+
+ // calculate string length according to charset
+ // function can be modulized same way we modulize decode/encode/htmlentities
+ if ($charset=='utf-8') {
+ if (function_exists('mb_strlen')) {
+ $real_length = mb_strlen($str,'utf-8');
+ } else {
+ // function needs length of string in bytes.
+ // mbstring overloading might break it
+ $str_length=strlen($str);
+ $str_index=0;
+ while ($str_index < $str_length) {
+ // start of internal utf-8 multibyte character detection
+ if (preg_match("/[\xC0-\xDF]/",$str[$str_index]) &&
+ isset($str[$str_index+1]) &&
+ preg_match("/[\x80-\xBF]/",$str[$str_index+1])) {
+ // two byte utf-8
+ $str_index=$str_index+2;
+ $real_length++;
+ } elseif (preg_match("/[\xE0-\xEF]/",$str[$str_index]) &&
+ isset($str[$str_index+2]) &&
+ preg_match("/[\x80-\xBF][\x80-\xBF]/",$str[$str_index+1].$str[$str_index+2])) {
+ // three byte utf-8
+ $str_index=$str_index+3;
+ $real_length++;
+ } elseif (preg_match("/[\xF0-\xF7]/",$str[$str_index]) &&
+ isset($str[$str_index+3]) &&
+ preg_match("/[\x80-\xBF][\x80-\xBF][\x80-\xBF]/",$str[$str_index+1].$str[$str_index+2].$str[$str_index+3])) {
+ // four byte utf-8
+ $str_index=$str_index+4;
+ $real_length++;
+ } elseif (preg_match("/[\xF8-\xFB]/",$str[$str_index]) &&
+ isset($str[$str_index+4]) &&
+ preg_match("/[\x80-\xBF][\x80-\xBF][\x80-\xBF][\x80-\xBF]/",
+ $str[$str_index+1].$str[$str_index+2].$str[$str_index+3].$str[$str_index+4])) {
+ // five byte utf-8
+ $str_index=$str_index+5;
+ $real_length++;
+ } elseif (preg_match("/[\xFC-\xFD]/",$str[$str_index]) &&
+ isset($str[$str_index+5]) &&
+ preg_match("/[\x80-\xBF][\x80-\xBF][\x80-\xBF][\x80-\xBF]/",
+ $str[$str_index+1].$str[$str_index+2].$str[$str_index+3].$str[$str_index+4].$str[$str_index+5])) {
+ // six byte utf-8
+ $str_index=$str_index+6;
+ $real_length++;
+ } else {
+ $str_index++;
+ $real_length++;
+ }
+ // end of internal utf-8 multibyte character detection
+ }
+ }
+ // end of utf-8 length detection
+ } elseif ($charset=='big5') {
+ // TODO: add big5 string length detection
+ $real_length=strlen($str);
+ } elseif ($charset=='gb2312') {
+ // TODO: add gb2312 string length detection
+ $real_length=strlen($str);
+ } elseif ($charset=='gb18030') {
+ // TODO: add gb18030 string length detection
+ $real_length=strlen($str);
+ } elseif ($charset=='euc-jp') {
+ // TODO: add euc-jp string length detection
+ $real_length=strlen($str);
+ } elseif ($charset=='euc-cn') {
+ // TODO: add euc-cn string length detection
+ $real_length=strlen($str);
+ } elseif ($charset=='euc-tw') {
+ // TODO: add euc-tw string length detection
+ $real_length=strlen($str);
+ } elseif ($charset=='euc-kr') {
+ // TODO: add euc-kr string length detection
+ $real_length=strlen($str);
+ } else {
+ $real_length=strlen($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;
+}
$PHP_SELF = php_self();
?>
\ No newline at end of file