]*?>.*?@siU', // Strip out javascript
'@&(quot|#34);@i', // Replace HTML entities
'@&(amp|#38);@i', // Ampersand &
'@&(lt|#60);@i', // Less Than <
'@&(gt|#62);@i', // Greater Than >
'@&(nbsp|#160);@i', // Non Breaking Space
'@&(iexcl|#161);@i', // Inverted Exclamation point
'@&(cent|#162);@i', // Cent
'@&(pound|#163);@i', // Pound
'@&(copy|#169);@i', // Copyright
'@&(reg|#174);@i', // Registered
'@&(trade|#8482);@i', // trade
'@'@i', // singleQuote
'@(\xc2\xa0)@', // nbsp or tab (encoded windows-style)
'@(\xe2\x80\x8b)@', // ZERO WIDTH SPACE
);
$Replace = array ('',
'"',
'#amper#sand#',
'<',
'>',
' ',
chr(161),
chr(162),
chr(163),
'(C)',//chr(169),// copyrighgt
'(R)',//chr(174),// registered
'(TM)',// trade
"'",
' ',
'',
);
$_html = preg_replace($Rules, $Replace, $_html);
// removing carriage return linefeeds, preserve those enclosed in
tags
if ($stripcrl === true )
{
if (stripos($_html,'')!==false)
{
$contentArr = self::splithtmlByPRE($_html);
foreach ($contentArr as $k =>&$elem)
{
if (stripos($elem,'')===false)
{
//$elem = str_replace('@(\r\n)@i',' ',$elem);
$elem = str_replace(array("\r\n","\n"),($isHTML?'':' '),$elem);
}
}
$_html = implode('',$contentArr);
}
else
{
$_html = str_replace(array("\r\n","\n"),($isHTML?'':' '),$_html);
}
}
$tags = array (
0 => '~]*>\r*\n*~si',
1 => '~]*>\r*\n*~si',
2 => '~]*>\r*\n*~si',
3 => '~]*>\r*\n*~si',
4 => '~]*>\r*\n*~si',
5 => '~]*>\r*\n*~si',
6 => '~]*>~si',
7 => '~]*>\r*\n*~si',
8 => '~]*>\r*\n*~si',
9 => '~]*>\r*\n*~si',
10 => '//',
11 => '//',
12 => '~
~si',
13 => '~]*>~si',
14 => '/<=([1234567890])/',
15 => '/>=([1234567890])/',
16 => '/<([1234567890])/',
17 => '/>([1234567890])/',
);
$Replace = array (
0 => "\r\n",
1 => "\r\n",
2 => "\r\n",
3 => "\r\n",
4 => "\r\n",
5 => "\r\n",
6 => "\r\n",
7 => "\r\n",
8 => "\r\n",
9 => "\r\n__________________________________________________\r\n",
10 => '#blockquote#type#cite#',
11 => '#blockquote#type#cite#',
12 => '#blockquote#end#cite#',
13 => '#blockquote#type#cite#',
14 => '#lowerorequal#than#$1',
15 => '#greaterorequal#than#$1',
16 => '#lower#than#$1',
17 => '#greater#than#$1',
);
$_html = preg_replace($tags,$Replace,$_html);
$_html = preg_replace('~\s*]*>~si',' - ',$_html);
$_html = preg_replace('~]+>~s','',$_html);
// replace emailaddresses eclosed in <> (eg.: ) with the emailaddress only (e.g: me@you.de)
self::replaceEmailAdresses($_html);
//convert hrefs to description -> URL
//$_html = preg_replace('~]+href=\"([^"]+)\"[^>]*>(.*)~si','[$2 -> $1]',$_html);
$_html = preg_replace_callback('~]+href=\"([^"]+)\"[^>]*>(.*?)~si','self::transform_url2text',$_html);
// reducing double \r\n to single ones, dont mess with pre sections
if ($stripcrl === true && $isHTML)
{
if (stripos($_html,'')!==false)
{
$contentArr = self::splithtmlByPRE($_html);
foreach ($contentArr as $k =>&$elem)
{
if (stripos($elem,'')===false)
{
//this is supposed to strip out all remaining stuff in tags, this is sometimes taking out whole sections off content
if ( $stripalltags ) {
$_html = preg_replace('~<[^>^@]+>~s','',$_html);
}
// strip out whitespace inbetween CR/LF
$elem = preg_replace('~\r\n\s+\r\n~si', "\r\n\r\n", $elem);
// strip out / reduce exess CR/LF
$elem = preg_replace('~\r\n{3,}~si',"\r\n\r\n",$elem);
}
}
$_html = implode('',$contentArr);
}
else
{
//this is supposed to strip out all remaining stuff in tags, this is sometimes taking out whole sections off content
if ( $stripalltags ) {
$_html = preg_replace('~<[^>^@]+>~s','',$_html);
}
// strip out whitespace inbetween CR/LF
$_html = preg_replace('~\r\n\s+\r\n~si', "\r\n\r\n", $_html);
// strip out / reduce exess CR/LF
$_html = preg_replace('~(\r\n){3,}~si',"\r\n\r\n",$_html);
}
}
//this is supposed to strip out all remaining stuff in tags, this is sometimes taking out whole sections off content
if ( $stripalltags ) {
$_html = preg_replace('~<[^>^@]+>~s','',$_html);
//$_html = strip_tags($_html, '');
}
// reducing spaces (not for input that was plain text from the beginning)
if ($isHTML) $_html = preg_replace('~ +~s',' ',$_html);
// restoring ampersands
$_html = str_replace('#amper#sand#','&',$_html);
// restoring lower|greater[or equal] than
$_html = str_replace('#lowerorequal#than#','<=',$_html);
$_html = str_replace('#greaterorequal#than#','>=',$_html);
$_html = str_replace('#lower#than#','<',$_html);
$_html = str_replace('#greater#than#','>',$_html);
//error_log(__METHOD__.__LINE__.' Charset:'.$displayCharset.' -> '.$_html);
$_html = html_entity_decode($_html, ENT_COMPAT, $displayCharset);
//error_log(__METHOD__.__LINE__.' Charset:'.$displayCharset.' After html_entity_decode: -> '.$_html);
//self::replaceEmailAdresses($_html);
$pos = strpos($_html, 'blockquote');
//error_log("convert HTML2Text: $_html");
if($pos === false) {
return $_html;
} else {
$indent = 0;
$indentString = '';
$quoteParts = preg_split('/#blockquote#type#cite#/', $_html, -1, PREG_SPLIT_OFFSET_CAPTURE);
foreach($quoteParts as $quotePart) {
if($quotePart[1] > 0) {
$indent++;
$indentString .= '>';
}
$quoteParts2 = preg_split('/#blockquote#end#cite#/', $quotePart[0], -1, PREG_SPLIT_OFFSET_CAPTURE);
foreach($quoteParts2 as $quotePart2) {
if($quotePart2[1] > 0) {
$indent--;
$indentString = substr($indentString, 0, $indent);
}
$quoteParts3 = explode("\r\n", $quotePart2[0]);
foreach($quoteParts3 as $quotePart3) {
//error_log(__METHOD__.__LINE__.'Line:'.$quotePart3);
$allowedLength = 76-strlen("\r\n$indentString");
// only break lines, if not already indented
if (substr($quotePart3,0,strlen($indentString)) != $indentString)
{
if (strlen($quotePart3) > $allowedLength) {
$s=explode(" ", $quotePart3);
$quotePart3 = "";
$linecnt = 0;
foreach ($s as $k=>$v) {
$cnt = strlen($v);
// only break long words within the wordboundaries,
// but it may destroy links, so we check for href and dont do it if we find it
if($cnt > $allowedLength && stripos($v,'href=')===false) {
//error_log(__METHOD__.__LINE__.'LongWordFound:'.$v);
$v=wordwrap($v, $allowedLength, "\r\n$indentString", true);
}
// the rest should be broken at the start of the new word that exceeds the limit
if ($linecnt+$cnt > $allowedLength) {
$v="\r\n$indentString$v";
//error_log(__METHOD__.__LINE__.'breaking here:'.$v);
$linecnt = 0;
} else {
$linecnt += $cnt;
}
if (strlen($v)) $quotePart3 .= (strlen($quotePart3) ? " " : "").$v;
}
}
}
//error_log(__METHOD__.__LINE__.'partString to return:'.$indentString . $quotePart3);
$asciiTextBuff[] = $indentString . $quotePart3 ;
}
}
}
return implode("\r\n",$asciiTextBuff);
}
}
/**
* split html by PRE tag, return array with all content pre-sections isolated in array elements
* @author Leithoff, Klaus
* @param string html
* @return mixed array of parts or unaffected html
*/
static function splithtmlByPRE($html)
{
$searchFor = '',$pos);
$length = $endofpre-$pos+6;
$html2ret[] = substr($html,$pos,$length);
$searchFor = '