X-Git-Url: https://vcs.fsf.org/?a=blobdiff_plain;f=functions%2Fstrings.php;h=84a70af4025bb2b7bed8385b016d52f2d9de62ca;hb=98abf40863b317860ae2cabba04f97b69103f556;hp=79b35f3e32ba84c0cb65e40d0aeebb768266eea7;hpb=6eaf5320bd2b723926fa9bead5706434e5f4e7fb;p=squirrelmail.git diff --git a/functions/strings.php b/functions/strings.php index 79b35f3e..84a70af4 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 @@ -77,8 +77,6 @@ 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: @@ -96,11 +94,11 @@ function sqMakeNewLine (&$str, $citeLevel, &$column) { * @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; - } + if ( preg_match('/^[\x09-\x0D]|^\x20/', $string) || $string=='') { + return true; + } else { + return false; + } } /** @@ -116,7 +114,6 @@ function sm_ctype_space($string) { * @return string the wrapped text */ function &sqBodyWrap (&$body, $wrap) { - //sm_print_r($body); //check for ctype support, and fake it if it doesn't exist if (!function_exists('ctype_space')) { function ctype_space ($string) { @@ -128,7 +125,7 @@ function &sqBodyWrap (&$body, $wrap) { $outString = ''; // current column since the last newline in the outstring $outStringCol = 0; - $length = strlen($body); + $length = sq_strlen($body); // where we are in the original string $pos = 0; // the number of >>> citation markers we are currently at @@ -140,15 +137,15 @@ function &sqBodyWrap (&$body, $wrap) { // we're at the beginning of a line, get the new cite level $newCiteLevel = 0; - while (($pos < $length) && ($body{$pos} == '>')) { + while (($pos < $length) && (sq_substr($body,$pos,1) == '>')) { $newCiteLevel++; $pos++; // skip over any spaces interleaved among the cite markers - while (($pos < $length) && ($body{$pos} == ' ')) { - + while (($pos < $length) && (sq_substr($body,$pos,1) == ' ')) { + $pos++; - + } if ($pos >= $length) { break; @@ -158,8 +155,8 @@ function &sqBodyWrap (&$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)) { - $outStringLast = $outString{strlen($outString) - 1}; + if ((sq_substr($body,$pos,1) == "\n" ) && (sq_strlen($outString) != 0)) { + $outStringLast = $outString{sq_strlen($outString) - 1}; if ($outStringLast != "\n") { $outString .= "\n"; } @@ -193,7 +190,7 @@ function &sqBodyWrap (&$body, $wrap) { } // find the next newline -- we don't want to go further than that - $nextNewline = strpos ($body, "\n", $pos); + $nextNewline = sq_strpos ($body, "\n", $pos); if ($nextNewline === FALSE) { $nextNewline = $length; } @@ -202,7 +199,7 @@ function &sqBodyWrap (&$body, $wrap) { // will work fine for this. Maybe revisit this later though // (for completeness more than anything else, I think) if ($citeLevel == 0) { - $outString .= substr ($body, $pos, ($nextNewline - $pos)); + $outString .= sq_substr ($body, $pos, ($nextNewline - $pos)); $outStringCol = $nextNewline - $pos; if ($nextNewline != $length) { sqMakeNewLine ($outString, 0, $outStringCol); @@ -210,71 +207,77 @@ function &sqBodyWrap (&$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 < $nextNewline) { // skip over initial spaces - while (($pos < $nextNewline) && (ctype_space ($body{$pos}))) { + while (($pos < $nextNewline) && (ctype_space (sq_substr($body,$pos,1)))) { $pos++; } - // if this is a short line then just append it and continue outer loop if (($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{$nextNewline} == "\n")) { + if (($nextNewline + 1 == $length) && (sq_substr($body,$nextNewline,1) == "\n")) { $nextNewline++; } - if (($nextNewline < $length && $body{$nextNewline} == "\n") && + // trim trailing spaces + $lastRealChar = $nextNewline; + while (($lastRealChar > $pos && $lastRealChar < $length) && (ctype_space (sq_substr($body,$lastRealChar,1)))) { + $lastRealChar--; + } + // decide if appending the short string is what we want + if (($nextNewline < $length && sq_substr($body,$nextNewline,1) == "\n") && isset($lastRealChar)) { - - // trim trailing spaces - $lastRealChar = $nextNewline; - while (($lastRealChar > $pos && $lastRealChar < $length) && (ctype_space ($body{$lastRealChar}))) { - $lastRealChar--; - } - - //check the first word: - $mypos = $nextNewline+1; - while (($mypos < $length) && ($body{$mypos} == '>')) { - $mypos++; - - // skip over any spaces interleaved among the cite markers - while (($mypos < $length) && ($body{$mypos} == ' ')) { - + $mypos = $pos; + //check the first word: + while (($mypos < $length) && (sq_substr($body,$mypos,1) == '>')) { + $mypos++; + // skip over any spaces interleaved among the cite markers + while (($mypos < $length) && (sq_substr($body,$mypos,1) == ' ')) { $mypos++; - + } + } +/* + $ldnspacecnt = 0; + if ($mypos == $nextNewline+1) { + while (($mypos < $length) && ($body{$mypos} == ' ')) { + $ldnspacecnt++; } - } - $firstword = substr($body,$mypos,strpos($body,' ',$mypos) - $mypos); - if ($firstword && ($firstword{0} == '-' || + } +*/ + + $firstword = sq_substr($body,$mypos,sq_strpos($body,' ',$mypos) - $mypos); + //if ($dowrap || $ldnspacecnt > 1 || ($firstword && ( + if (!$smartwrap || $firstword && ( + $firstword{0} == '-' || $firstword{0} == '+' || $firstword{0} == '*' || + sq_substr($firstword,0,1) == sq_strtoupper(sq_substr($firstword,0,1)) || strpos($firstword,':'))) { - $outString .= substr($body,$pos,($lastRealChar - $pos+1)); - $outStringCol += ($lastRealChar - $pos); - sqMakeNewLine($outString,$citeLevel,$outStringCol); - $nextNewline++; - $pos = $nextNewline; - $outStringCol--; - continue; //break 2; - } - } - + $outString .= sq_substr($body,$pos,($lastRealChar - $pos+1)); + $outStringCol += ($lastRealChar - $pos); + sqMakeNewLine($outString,$citeLevel,$outStringCol); + $nextNewline++; + $pos = $nextNewline; + $outStringCol--; + continue; + } - // trim trailing spaces - $lastRealChar = $nextNewline; - while (($lastRealChar > $pos && $lastRealChar < $length) && (ctype_space ($body{$lastRealChar}))) { - $lastRealChar--; } - $outString .= substr ($body, $pos, ($lastRealChar - $pos + 1)); + + $outString .= sq_substr ($body, $pos, ($lastRealChar - $pos + 1)); $outStringCol += ($lastRealChar - $pos); $pos = $nextNewline + 1; continue; } + $eol = $pos + $wrap - $citeLevel - $outStringCol; // eol is the tentative end of line. // look backwards for there for a whitespace to break at. @@ -288,7 +291,7 @@ function &sqBodyWrap (&$body, $wrap) { // start looking backwards for whitespace to break at. $breakPoint = $eol; - while (($breakPoint > $pos) && (! ctype_space ($body{$breakPoint}))) { + while (($breakPoint > $pos) && (! ctype_space (sq_substr($body,$breakPoint,1)))) { $breakPoint--; } @@ -321,13 +324,13 @@ function &sqBodyWrap (&$body, $wrap) { } // skip newlines or whitespace at the beginning of the string - $substring = substr ($body, $pos, ($breakPoint - $pos)); + $substring = sq_substr ($body, $pos, ($breakPoint - $pos)); $substring = rtrim ($substring); // do rtrim and ctype_space have the same ideas about whitespace? $outString .= $substring; - $outStringCol += strlen ($substring); + $outStringCol += sq_strlen ($substring); // advance past the whitespace which caused the wrap $pos = $breakPoint; - while (($pos < $length) && (ctype_space ($body{$pos}))) { + while (($pos < $length) && (ctype_space (sq_substr($body,$pos,1)))) { $pos++; } if ($pos < $length) { @@ -345,20 +348,22 @@ function &sqBodyWrap (&$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, &#039; 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, $wrap); return; } } @@ -377,9 +382,9 @@ function sqWordWrap(&$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); $i ++; /* Add more words (as long as they fit) */ @@ -387,7 +392,7 @@ function sqWordWrap(&$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 $line_len += 1; } @@ -636,7 +641,7 @@ function sq_mt_seed($Val) { $Val *= -1; } - if ($Val = 0) { + if ($Val == 0) { return; } @@ -734,7 +739,7 @@ function show_readable_size($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 @@ -831,7 +836,7 @@ 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.''; @@ -860,6 +865,19 @@ 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"; + } } $buffer = ob_get_contents(); // Grab the print_r output ob_end_clean(); // Silently discard the output & stop buffering @@ -949,7 +967,10 @@ function sq_get_html_translation_table($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 $quote_style quote encoding style. Possible values (without quotes): @@ -968,5 +989,265 @@ function sq_htmlentities($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', + '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_internal_encoding') && + 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); +} $PHP_SELF = php_self(); ?> \ No newline at end of file