mirror of
https://github.com/EGroupware/egroupware.git
synced 2024-12-27 09:09:04 +01:00
move html mail specific part of translation class to Api\Mail\Html
This commit is contained in:
Ralf Becker
parent
1bf9c25cba
commit
5f68bc6664
@ -14,9 +14,6 @@ namespace EGroupware\Api\Html;
|
||||
|
||||
use EGroupware\Api;
|
||||
|
||||
// explicitly name old, not yet converted api classes
|
||||
use translation; // mail specific stuff not in Api\Translation
|
||||
|
||||
require_once(__DIR__.'/htmLawed/htmLawed.php');
|
||||
|
||||
/**
|
||||
@ -141,7 +138,7 @@ class HtmLawed
|
||||
{
|
||||
//error_log(__METHOD__.__LINE__.array2string($newStyle[0]));
|
||||
$style2buffer = implode('',$newStyle[0]);
|
||||
// only replace what we have found, we use it here, as we use the same routine in translation::replaceTagsCompletley
|
||||
// only replace what we have found, we use it here, as we use the same routine in Api\Mail\Html::replaceTagsCompletley
|
||||
// no need to do the extra routine
|
||||
$html = str_ireplace($newStyle[0],'',$html);
|
||||
}
|
||||
@ -170,15 +167,15 @@ class HtmLawed
|
||||
// CSS Security
|
||||
// http://code.google.com/p/browsersec/wiki/Part1#Cascading_stylesheets
|
||||
$css = preg_replace('/(javascript|expession|-moz-binding)/i','',$style);
|
||||
if (stripos($css,'script')!==false) translation::replaceTagsCompletley($css,'script'); // Strip out script that may be included
|
||||
if (stripos($css,'script')!==false) Api\Mail\Html::replaceTagsCompletley($css,'script'); // Strip out script that may be included
|
||||
// we need this, as styledefinitions are enclosed with curly brackets; and template stuff tries to replace everything between curly brackets that is having no horizontal whitespace
|
||||
// as the comments as <!-- styledefinition --> in stylesheet are outdated, and ck-editor does not understand it, we remove it
|
||||
$css_no_comment = str_replace(array(':','<!--','-->'),array(': ','',''),$css);
|
||||
//error_log(__METHOD__.__LINE__.$css);
|
||||
// we already removed what we have found, above, as we used pretty much the same routine as in translation::replaceTagsCompletley
|
||||
// we already removed what we have found, above, as we used pretty much the same routine as in Api\Mail\Html::replaceTagsCompletley
|
||||
// no need to do the extra routine
|
||||
// TODO: we may have to strip urls and maybe comments and ifs
|
||||
//if (stripos($html,'style')!==false) translation::replaceTagsCompletley($html,'style'); // clean out empty or pagewide style definitions / left over tags
|
||||
//if (stripos($html,'style')!==false) Api\Mail\Html::replaceTagsCompletley($html,'style'); // clean out empty or pagewide style definitions / left over tags
|
||||
return $css_no_comment;
|
||||
}
|
||||
|
||||
|
||||
@ -31,7 +31,6 @@ use Horde_Mime_Mdn;
|
||||
use tidy;
|
||||
|
||||
// old not yet converted api classes
|
||||
use translation; // mail stuff not in Api\Translation
|
||||
use addressbook_merge; // should go to Contacts\Merge
|
||||
|
||||
/**
|
||||
@ -1529,7 +1528,7 @@ class Mail
|
||||
if (empty($headerObject['BODYPREVIEW'])&&$part->getPrimaryType()== 'text')
|
||||
{
|
||||
$charset = $part->getContentTypeParameter('charset');
|
||||
$buffer = translation::convertHTMLToText($part->toString(array(
|
||||
$buffer = Mail\Html::convertHTMLToText($part->toString(array(
|
||||
'encode' => Horde_Mime_Part::ENCODE_BINARY, // otherwise we cant recode charset
|
||||
)), $charset, 'utf-8');
|
||||
$headerObject['BODYPREVIEW']=trim(str_replace(array("\r\n","\r","\n"),' ',mb_substr(Translation::convert_jsonsafe($buffer),0,((int)$_fetchPreviews<300?300:$_fetchPreviews))));
|
||||
@ -1562,7 +1561,7 @@ class Mail
|
||||
($partdisposition !== 'attachment')) {
|
||||
$_structure=$part;
|
||||
$this->fetchPartContents($uid, $_structure, false,true);
|
||||
$headerObject['BODYPREVIEW']=trim(str_replace(array("\r\n","\r","\n"),' ',mb_substr(translation::convertHTMLToText($_structure->getContents()),0,((int)$_fetchPreviews<300?300:$_fetchPreviews))));
|
||||
$headerObject['BODYPREVIEW']=trim(str_replace(array("\r\n","\r","\n"),' ',mb_substr(Mail\Html::convertHTMLToText($_structure->getContents()),0,((int)$_fetchPreviews<300?300:$_fetchPreviews))));
|
||||
$charSet=Translation::detect_encoding($headerObject['BODYPREVIEW']);
|
||||
// add line breaks to $bodyParts
|
||||
//error_log(__METHOD__.' ('.__LINE__.') '.' Charset:'.$bodyParts[$i]['charSet'].'->'.$bodyParts[$i]['body']);
|
||||
@ -2212,7 +2211,7 @@ class Mail
|
||||
}
|
||||
else
|
||||
{
|
||||
$_string = translation::decodeMailHeader($_string,self::$displayCharset);
|
||||
$_string = Mail\Html::decodeMailHeader($_string,self::$displayCharset);
|
||||
$test = @json_encode($_string);
|
||||
//error_log(__METHOD__.__LINE__.' ->'.strlen($singleBodyPart['body']).' Error:'.json_last_error().'<- BodyPart:#'.$test.'#');
|
||||
if (($test=="null" || $test === false || is_null($test)) && strlen($_string)>0)
|
||||
@ -4175,18 +4174,18 @@ class Mail
|
||||
$_html = str_replace(array('&amp;','<DIV><BR></DIV>',"<DIV> </DIV>",'<div> </div>','</td></font>','<br><td>','<tr></tr>','<o:p></o:p>','<o:p>','</o:p>'),
|
||||
array('&', '<BR>', '<BR>', '<BR>', '</font></td>','<td>', '', '', '', ''),$_html);
|
||||
//$_html = str_replace(array('&amp;'),array('&'),$_html);
|
||||
if (stripos($_html,'style')!==false) translation::replaceTagsCompletley($_html,'style'); // clean out empty or pagewide style definitions / left over tags
|
||||
if (stripos($_html,'head')!==false) translation::replaceTagsCompletley($_html,'head'); // Strip out stuff in head
|
||||
//if (stripos($_html,'![if')!==false && stripos($_html,'<![endif]>')!==false) translation::replaceTagsCompletley($_html,'!\[if','<!\[endif\]>',false); // Strip out stuff in ifs
|
||||
//if (stripos($_html,'!--[if')!==false && stripos($_html,'<![endif]-->')!==false) translation::replaceTagsCompletley($_html,'!--\[if','<!\[endif\]-->',false); // Strip out stuff in ifs
|
||||
if (stripos($_html,'style')!==false) Mail\Html::replaceTagsCompletley($_html,'style'); // clean out empty or pagewide style definitions / left over tags
|
||||
if (stripos($_html,'head')!==false) Mail\Html::replaceTagsCompletley($_html,'head'); // Strip out stuff in head
|
||||
//if (stripos($_html,'![if')!==false && stripos($_html,'<![endif]>')!==false) Mail\Html::replaceTagsCompletley($_html,'!\[if','<!\[endif\]>',false); // Strip out stuff in ifs
|
||||
//if (stripos($_html,'!--[if')!==false && stripos($_html,'<![endif]-->')!==false) Mail\Html::replaceTagsCompletley($_html,'!--\[if','<!\[endif\]-->',false); // Strip out stuff in ifs
|
||||
//error_log(__METHOD__.' ('.__LINE__.') '.$_html);
|
||||
|
||||
if (get_magic_quotes_gpc() === 1) $_html = stripslashes($_html);
|
||||
// Strip out doctype in head, as htmlLawed cannot handle it TODO: Consider extracting it and adding it afterwards
|
||||
if (stripos($_html,'!doctype')!==false) translation::replaceTagsCompletley($_html,'!doctype');
|
||||
if (stripos($_html,'?xml:namespace')!==false) translation::replaceTagsCompletley($_html,'\?xml:namespace','/>',false);
|
||||
if (stripos($_html,'?xml version')!==false) translation::replaceTagsCompletley($_html,'\?xml version','\?>',false);
|
||||
if (strpos($_html,'!CURSOR')!==false) translation::replaceTagsCompletley($_html,'!CURSOR');
|
||||
if (stripos($_html,'!doctype')!==false) Mail\Html::replaceTagsCompletley($_html,'!doctype');
|
||||
if (stripos($_html,'?xml:namespace')!==false) Mail\Html::replaceTagsCompletley($_html,'\?xml:namespace','/>',false);
|
||||
if (stripos($_html,'?xml version')!==false) Mail\Html::replaceTagsCompletley($_html,'\?xml version','\?>',false);
|
||||
if (strpos($_html,'!CURSOR')!==false) Mail\Html::replaceTagsCompletley($_html,'!CURSOR');
|
||||
// htmLawed filter only the 'body'
|
||||
//preg_match('`(<htm.+?<body[^>]*>)(.+?)(</body>.*?</html>)`ims', $_html, $matches);
|
||||
//if ($matches[2])
|
||||
@ -4857,7 +4856,7 @@ class Mail
|
||||
if (!$preserveHTML)
|
||||
{
|
||||
$alreadyHtmlLawed=false;
|
||||
// as translation::convert reduces \r\n to \n and purifier eats \n -> peplace it with a single space
|
||||
// as Translation::convert reduces \r\n to \n and purifier eats \n -> peplace it with a single space
|
||||
$newBody = str_replace("\n"," ",$newBody);
|
||||
// convert HTML to text, as we dont want HTML in infologs
|
||||
if ($useTidy && extension_loaded('tidy'))
|
||||
@ -4902,7 +4901,7 @@ class Mail
|
||||
$alreadyHtmlLawed=true;
|
||||
}
|
||||
//error_log(__METHOD__.' ('.__LINE__.') '.' after purify:'.$newBody);
|
||||
if ($preserveHTML==false) $newBody = translation::convertHTMLToText($newBody,self::$displayCharset,true,true);
|
||||
if ($preserveHTML==false) $newBody = Mail\Html::convertHTMLToText($newBody,self::$displayCharset,true,true);
|
||||
//error_log(__METHOD__.' ('.__LINE__.') '.' after convertHTMLToText:'.$newBody);
|
||||
if ($preserveHTML==false) $newBody = nl2br($newBody); // we need this, as htmLawed removes \r\n
|
||||
/*if (!$alreadyHtmlLawed) */ $mailClass->getCleanHTML($newBody); // remove stuff we regard as unwanted
|
||||
@ -4913,9 +4912,9 @@ class Mail
|
||||
continue;
|
||||
}
|
||||
//error_log(__METHOD__.' ('.__LINE__.') '.' Body(after specialchars):'.$newBody);
|
||||
//use translation::convertHTMLToText instead of strip_tags, (even message is plain text) as strip_tags eats away too much
|
||||
//use Mail\Html::convertHTMLToText instead of strip_tags, (even message is plain text) as strip_tags eats away too much
|
||||
//$newBody = strip_tags($newBody); //we need to fix broken tags (or just stuff like "<800 USD/p" )
|
||||
$newBody = translation::convertHTMLToText($newBody,self::$displayCharset,false,false);
|
||||
$newBody = Mail\Html::convertHTMLToText($newBody,self::$displayCharset,false,false);
|
||||
//error_log(__METHOD__.' ('.__LINE__.') '.' Body(after strip tags):'.$newBody);
|
||||
$newBody = htmlspecialchars_decode($newBody,ENT_QUOTES);
|
||||
//error_log(__METHOD__.' ('.__LINE__.') '.' Body (after hmlspc_decode):'.$newBody);
|
||||
@ -5300,7 +5299,7 @@ class Mail
|
||||
// CSS Security
|
||||
// http://code.google.com/p/browsersec/wiki/Part1#Cascading_stylesheets
|
||||
$css = preg_replace('/(javascript|expression|-moz-binding)/i','',$style);
|
||||
if (stripos($css,'script')!==false) translation::replaceTagsCompletley($css,'script'); // Strip out script that may be included
|
||||
if (stripos($css,'script')!==false) Mail\Html::replaceTagsCompletley($css,'script'); // Strip out script that may be included
|
||||
// we need this, as styledefinitions are enclosed with curly brackets; and template stuff tries to replace everything between curly brackets that is having no horizontal whitespace
|
||||
// as the comments as <!-- styledefinition --> in stylesheet are outdated, and ck-editor does not understand it, we remove it
|
||||
$css = str_replace(array(':','<!--','-->'),array(': ','',''),$css);
|
||||
@ -5985,7 +5984,7 @@ class Mail
|
||||
{
|
||||
if ($bodyParts[$i]['mimeType']=='text/html')
|
||||
{
|
||||
$bodyParts[$i]['body'] = translation::convertHTMLToText($bodyParts[$i]['body'],$bodyParts[$i]['charSet'],true,$stripalltags=true);
|
||||
$bodyParts[$i]['body'] = Mail\Html::convertHTMLToText($bodyParts[$i]['body'],$bodyParts[$i]['charSet'],true,$stripalltags=true);
|
||||
$bodyParts[$i]['mimeType']='text/plain';
|
||||
}
|
||||
}
|
||||
|
||||
500
api/src/Mail/Html.php
Normal file
500
api/src/Mail/Html.php
Normal file
@ -0,0 +1,500 @@
|
||||
<?php
|
||||
/**
|
||||
* EGroupware API - Html mail handling
|
||||
*
|
||||
* @link http://www.egroupware.org
|
||||
* @author Klaus Leithoff <kl@stylite.de>
|
||||
* @license http://opensource.org/licenses/lgpl-license.php LGPL - GNU Lesser General Public License
|
||||
* @package api
|
||||
* @version $Id$
|
||||
*/
|
||||
|
||||
namespace EGroupware\Api\Mail;
|
||||
|
||||
use EGroupware\Api;
|
||||
|
||||
/**
|
||||
* Html mail handling
|
||||
*
|
||||
* @todo clean up IDE warnings, mostly multiple assigments
|
||||
*/
|
||||
class Html
|
||||
{
|
||||
/**
|
||||
* Return the decoded string meeting some additional requirements for mailheaders
|
||||
*
|
||||
* @param string $_string -> part of an mailheader
|
||||
* @param string $displayCharset the charset parameter specifies the character set to represent the result by (if iconv_mime_decode is to be used)
|
||||
* @return string
|
||||
*/
|
||||
static function decodeMailHeader($_string, $displayCharset='utf-8')
|
||||
{
|
||||
//error_log(__FILE__.','.__METHOD__.':'."called with $_string and CHARSET $displayCharset");
|
||||
if(function_exists('imap_mime_header_decode'))
|
||||
{
|
||||
// some characterreplacements, as they fail to translate
|
||||
$sar = array(
|
||||
'@(\x84|\x93|\x94)@',
|
||||
'@(\x96|\x97|\x1a)@',
|
||||
'@(\x91|\x92)@',
|
||||
'@(\x85)@',
|
||||
'@(\x86)@',
|
||||
);
|
||||
$rar = array(
|
||||
'"',
|
||||
'-',
|
||||
'\'',
|
||||
'...',
|
||||
'+',
|
||||
);
|
||||
|
||||
$newString = '';
|
||||
|
||||
$string = preg_replace('/\?=\s+=\?/', '?= =?', $_string);
|
||||
|
||||
$elements=imap_mime_header_decode($string);
|
||||
|
||||
$convertAtEnd = false;
|
||||
foreach((array)$elements as $element)
|
||||
{
|
||||
if ($element->charset == 'default') $element->charset = Api\Translation::detect_encoding($element->text);
|
||||
if ($element->charset != 'x-unknown')
|
||||
{
|
||||
if( strtoupper($element->charset) != 'UTF-8') $element->text = preg_replace($sar,$rar,$element->text);
|
||||
// check if there is a possible nested encoding; make sure that the inputstring and the decoded result are different to avoid loops
|
||||
if(preg_match('/\?=.+=\?/', $element->text) && $element->text != $_string)
|
||||
{
|
||||
$element->text = self::decodeMailHeader($element->text, $element->charset);
|
||||
$element->charset = $displayCharset;
|
||||
}
|
||||
$newString .= Api\Translation::convert($element->text,$element->charset);
|
||||
}
|
||||
else
|
||||
{
|
||||
$newString .= $element->text;
|
||||
$convertAtEnd = true;
|
||||
}
|
||||
}
|
||||
if ($convertAtEnd) $newString = self::decodeMailHeader($newString,$displayCharset);
|
||||
return preg_replace('/([\000-\012\015\016\020-\037\075])/','',$newString);
|
||||
}
|
||||
elseif(function_exists(mb_decode_mimeheader))
|
||||
{
|
||||
$matches = null;
|
||||
if(preg_match_all('/=\?.*\?Q\?.*\?=/iU', $string=$_string, $matches))
|
||||
{
|
||||
foreach($matches[0] as $match)
|
||||
{
|
||||
$fixedMatch = str_replace('_', ' ', $match);
|
||||
$string = str_replace($match, $fixedMatch, $string);
|
||||
}
|
||||
$string = str_replace('=?ISO8859-','=?ISO-8859-',
|
||||
str_replace('=?windows-1258','=?ISO-8859-1',$string));
|
||||
}
|
||||
$string = mb_decode_mimeheader($string);
|
||||
return preg_replace('/([\000-\012\015\016\020-\037\075])/','',$string);
|
||||
}
|
||||
elseif(function_exists(iconv_mime_decode))
|
||||
{
|
||||
// continue decoding also if an error occurs
|
||||
$string = @iconv_mime_decode($_string, 2, $displayCharset);
|
||||
return preg_replace('/([\000-\012\015\016\020-\037\075])/','',$string);
|
||||
}
|
||||
|
||||
// no decoding function available
|
||||
return preg_replace('/([\000-\012\015\016\020-\037\075])/','',$_string);
|
||||
}
|
||||
|
||||
/**
|
||||
* replace emailaddresses enclosed in <> (eg.: <me@you.de>) with the emailaddress only (e.g: me@you.de)
|
||||
* as well as those emailadresses in links, and within broken links
|
||||
* @param string the text to process
|
||||
* @return 1
|
||||
*/
|
||||
static function replaceEmailAdresses(&$text)
|
||||
{
|
||||
//error_log($text);
|
||||
//replace CRLF with something other to be preserved via preg_replace as CRLF seems to vanish
|
||||
$text2 = str_replace("\r\n",'<#cr-lf#>',$text);
|
||||
// replace emailaddresses eclosed in <> (eg.: <me@you.de>) with the emailaddress only (e.g: me@you.de)
|
||||
$text3 = preg_replace("/(<|<a href=\")*(mailto:([\w\.,-.,_.,0-9.]+)(@)([\w\.,-.,_.,0-9.]+))(>|>)*/i","$2 ", $text2);
|
||||
//$text = preg_replace_callback("/(<|<a href=\")*(mailto:([\w\.,-.,_.,0-9.]+)(@)([\w\.,-.,_.,0-9.]+))(>|>)*/i",'self::transform_mailto2text',$text);
|
||||
//$text = preg_replace('~<a[^>]+href=\"(mailto:)+([^"]+)\"[^>]*>~si','$2 ',$text);
|
||||
$text4 = preg_replace_callback('~<a[^>]+href=\"(mailto:)+([^"]+)\"[^>]*>([ @\w\.,-.,_.,0-9.]+)<\/a>~si','self::transform_mailto2text',$text3);
|
||||
$text5 = preg_replace("/(([\w\.,-.,_.,0-9.]+)(@)([\w\.,-.,_.,0-9.]+))( |\s)*(<\/a>)*( |\s)*(>|>)*/i","$1 ", $text4);
|
||||
$text6 = preg_replace("/(<|<)*(([\w\.,-.,_.,0-9.]+)@([\w\.,-.,_.,0-9.]+))(>|>)*/i","$2 ", $text5);
|
||||
$text = str_replace('<#cr-lf#>',"\r\n",$text6);
|
||||
return 1;
|
||||
}
|
||||
|
||||
/**
|
||||
* strip tags out of the message completely with their content
|
||||
* @param string $_body is the text to be processed
|
||||
* @param string $tag is the tagname which is to be removed. Note, that only the name of the tag is to be passed to the function
|
||||
* without the enclosing brackets
|
||||
* @param string $endtag can be different from tag but should be used only, if begin and endtag are known to be different e.g.: <!-- -->
|
||||
* @param bool $addbracesforendtag if endtag is given, you may decide if the </ and > braces are to be added,
|
||||
* or if you want the string to be matched as is
|
||||
* @return void the modified text is passed via reference
|
||||
*/
|
||||
static function replaceTagsCompletley(&$_body,$tag,$endtag='',$addbracesforendtag=true)
|
||||
{
|
||||
if ($tag) $tag = strtolower($tag);
|
||||
$singleton = false;
|
||||
if ($endtag=='/>') $singleton =true;
|
||||
if ($endtag == '' || empty($endtag) || !isset($endtag))
|
||||
{
|
||||
$endtag = $tag;
|
||||
} else {
|
||||
$endtag = strtolower($endtag);
|
||||
//error_log(__METHOD__.' Using EndTag:'.$endtag);
|
||||
}
|
||||
// strip tags out of the message completely with their content
|
||||
if ($_body) {
|
||||
if ($singleton)
|
||||
{
|
||||
//$_body = preg_replace('~<'.$tag.'[^>].*? '.$endtag.'~simU','',$_body);
|
||||
$_body = preg_replace('~<?'.$tag.'[^>].* '.$endtag.'~simU','',$_body); // we are in Ungreedy mode, so we expect * to be ungreedy without specifying ?
|
||||
}
|
||||
else
|
||||
{
|
||||
$found=null;
|
||||
if ($addbracesforendtag === true )
|
||||
{
|
||||
if (stripos($_body,'<'.$tag)!==false) $ct = preg_match_all('#<'.$tag.'(?:\s.*)?>(.+)</'.$endtag.'>#isU', $_body, $found);
|
||||
if ($ct>0)
|
||||
{
|
||||
//error_log(__METHOD__.__LINE__.array2string($found[0]));
|
||||
// only replace what we have found
|
||||
$_body = str_ireplace($found[0],'',$_body);
|
||||
}
|
||||
// remove left over tags, unfinished ones, and so on
|
||||
$_body = preg_replace('~<'.$tag.'[^>]*?>~si','',$_body);
|
||||
}
|
||||
if ($addbracesforendtag === false )
|
||||
{
|
||||
if (stripos($_body,'<'.$tag)!==false) $ct = preg_match_all('#<'.$tag.'(?:\s.*)?>(.+)'.$endtag.'#isU', $_body, $found);
|
||||
if ($ct>0)
|
||||
{
|
||||
//error_log(__METHOD__.__LINE__.array2string($found[0]));
|
||||
// only replace what we have found
|
||||
$_body = str_ireplace($found[0],'',$_body);
|
||||
}
|
||||
/*
|
||||
$_body = preg_replace('~<'.$tag.'[^>]*?>(.*?)'.$endtag.'~simU','',$_body);
|
||||
*/
|
||||
// remove left over tags, unfinished ones, and so on
|
||||
$_body = preg_replace(array('~<'.$tag.'[^>]*?>~si', '~'.$endtag.'~'), '', $_body);
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
static function transform_mailto2text($matches)
|
||||
{
|
||||
//error_log(__METHOD__.__LINE__.array2string($matches));
|
||||
// this is the actual url
|
||||
$matches[2] = trim(strip_tags($matches[2]));
|
||||
$matches[3] = trim(strip_tags($matches[3]));
|
||||
$matches[2] = str_replace(array('%40','%20'),array('@',' '),$matches[2]);
|
||||
$matches[3] = str_replace(array('%40','%20'),array('@',' '),$matches[3]);
|
||||
return $matches[1].$matches[2].($matches[2]==$matches[3]?' ':' -> '.$matches[3].' ');
|
||||
}
|
||||
|
||||
static function transform_url2text($matches)
|
||||
{
|
||||
//error_log(__METHOD__.__LINE__.array2string($matches));
|
||||
$linkTextislink = false;
|
||||
// this is the actual url
|
||||
$matches[2] = trim(strip_tags($matches[2]));
|
||||
if ($matches[2]==$matches[1]) $linkTextislink = true;
|
||||
$matches[1] = str_replace(' ','%20',$matches[1]);
|
||||
return ($linkTextislink?' ':'[ ').$matches[1].($linkTextislink?'':' -> '.$matches[2]).($linkTextislink?' ':' ]');
|
||||
}
|
||||
|
||||
/**
|
||||
* convertHTMLToText
|
||||
* @param string $_html : Text to be stripped down
|
||||
* @param string $displayCharset : charset to use; should be a valid charset
|
||||
* @param bool $stripcrl : flag to indicate for the removal of all crlf \r\n
|
||||
* @param bool $stripalltags : flag to indicate wether or not to strip $_html from all remaining tags
|
||||
* @return text $_html : the modified text.
|
||||
*/
|
||||
static function convertHTMLToText($_html,$displayCharset=false,$stripcrl=false,$stripalltags=true)
|
||||
{
|
||||
// assume input isHTML, but test the input anyway, because,
|
||||
// if it is not, we may not want to strip whitespace
|
||||
$isHTML = true;
|
||||
if (strlen(strip_tags($_html)) == strlen($_html))
|
||||
{
|
||||
$isHTML = false;
|
||||
// return $_html; // maybe we should not proceed at all
|
||||
}
|
||||
if ($displayCharset === false) $displayCharset = self::$system_charset;
|
||||
//error_log(__METHOD__.$_html);
|
||||
#print '<hr>';
|
||||
#print "<pre>"; print htmlspecialchars($_html);
|
||||
#print "</pre>";
|
||||
#print "<hr>";
|
||||
if (stripos($_html,'style')!==false) self::replaceTagsCompletley($_html,'style'); // clean out empty or pagewide style definitions / left over tags
|
||||
if (stripos($_html,'head')!==false) self::replaceTagsCompletley($_html,'head'); // Strip out stuff in head
|
||||
if (stripos($_html,'![if')!==false && stripos($_html,'<![endif]>')!==false) self::replaceTagsCompletley($_html,'!\[if','<!\[endif\]>',false); // Strip out stuff in ifs
|
||||
if (stripos($_html,'!--[if')!==false && stripos($_html,'<![endif]-->')!==false) self::replaceTagsCompletley($_html,'!--\[if','<!\[endif\]-->',false); // Strip out stuff in ifs
|
||||
$Rules = array ('@<script[^>]*?>.*?</script>@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, $_html);
|
||||
|
||||
// removing carriage return linefeeds, preserve those enclosed in <pre> </pre> tags
|
||||
if ($stripcrl === true )
|
||||
{
|
||||
if (stripos($_html,'<pre ')!==false || stripos($_html,'<pre>')!==false)
|
||||
{
|
||||
$contentArr = self::splithtmlByPRE($_html);
|
||||
foreach ($contentArr as $k =>&$elem)
|
||||
{
|
||||
if (stripos($elem,'<pre ')===false && stripos($elem,'<pre>')===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 => '~<h[123][^>]*>\r*\n*~si',
|
||||
1 => '~<h[456][^>]*>\r*\n*~si',
|
||||
2 => '~<table[^>]*>\r*\n*~si',
|
||||
3 => '~<tr[^>]*>\r*\n*~si',
|
||||
4 => '~<li[^>]*>\r*\n*~si',
|
||||
5 => '~<br[^>]*>\r*\n*~si',
|
||||
6 => '~<br[^>]*>~si',
|
||||
7 => '~<p[^>]*>\r*\n*~si',
|
||||
8 => '~<div[^>]*>\r*\n*~si',
|
||||
9 => '~<hr[^>]*>\r*\n*~si',
|
||||
10 => '/<blockquote type="cite">/',
|
||||
11 => '/<blockquote>/',
|
||||
12 => '~</blockquote>~si',
|
||||
13 => '~<blockquote[^>]*>~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);
|
||||
$_html = preg_replace('~</t(d|h)>\s*<t(d|h)[^>]*>~si',' - ',$_html);
|
||||
$_html = preg_replace('~<img[^>]+>~s','',$_html);
|
||||
// replace emailaddresses eclosed in <> (eg.: <me@you.de>) with the emailaddress only (e.g: me@you.de)
|
||||
self::replaceEmailAdresses($_html);
|
||||
//convert hrefs to description -> URL
|
||||
//$_html = preg_replace('~<a[^>]+href=\"([^"]+)\"[^>]*>(.*)</a>~si','[$2 -> $1]',$_html);
|
||||
$_html = preg_replace_callback('~<a[^>]+href=\"([^"]+)\"[^>]*>(.*?)</a>~si','self::transform_url2text',$_html);
|
||||
|
||||
// reducing double \r\n to single ones, dont mess with pre sections
|
||||
if ($stripcrl === true && $isHTML)
|
||||
{
|
||||
if (stripos($_html,'<pre ')!==false || stripos($_html,'<pre>')!==false)
|
||||
{
|
||||
$contentArr = self::splithtmlByPRE($_html);
|
||||
foreach ($contentArr as $k =>&$elem)
|
||||
{
|
||||
if (stripos($elem,'<pre ')===false && stripos($elem,'<pre>')===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, '<a>');
|
||||
}
|
||||
// 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 = '<pre ';
|
||||
$pos = stripos($html,$searchFor);
|
||||
if ($pos===false)
|
||||
{
|
||||
$searchFor = '<pre>';
|
||||
$pos = stripos($html,$searchFor);
|
||||
}
|
||||
if ($pos === false)
|
||||
{
|
||||
return $html;
|
||||
}
|
||||
$html2ret[] = substr($html,0,$pos);
|
||||
while ($pos!==false)
|
||||
{
|
||||
$endofpre = stripos($html,'</pre>',$pos);
|
||||
$length = $endofpre-$pos+6;
|
||||
$html2ret[] = substr($html,$pos,$length);
|
||||
$searchFor = '<pre ';
|
||||
$pos = stripos($html,$searchFor, $endofpre+6);
|
||||
if ($pos===false)
|
||||
{
|
||||
$searchFor = '<pre>';
|
||||
$pos = stripos($html,$searchFor, $endofpre+6);
|
||||
}
|
||||
$html2ret[] = ($pos ? substr($html,$endofpre+6,$pos-($endofpre+6)): substr($html,$endofpre+6));
|
||||
//$pos=false;
|
||||
}
|
||||
//error_log(__METHOD__.__LINE__.array2string($html2ret));
|
||||
return $html2ret;
|
||||
}
|
||||
}
|
||||
@ -16,7 +16,7 @@ use EGroupware\Api;
|
||||
/**
|
||||
* EGroupware API - Translations
|
||||
*
|
||||
* @deprecated use Api\Translation for non-mail specific methods
|
||||
* @deprecated use Api\Translation for non-mail specific methods or Api\Mail\Html for others
|
||||
*/
|
||||
class translation extends Api\Translation
|
||||
{
|
||||
@ -26,83 +26,11 @@ class translation extends Api\Translation
|
||||
* @param string $_string -> part of an mailheader
|
||||
* @param string $displayCharset the charset parameter specifies the character set to represent the result by (if iconv_mime_decode is to be used)
|
||||
* @return string
|
||||
* @deprecated use Api\Mail\Html::decodeMailHeader
|
||||
*/
|
||||
static function decodeMailHeader($_string, $displayCharset='utf-8')
|
||||
{
|
||||
//error_log(__FILE__.','.__METHOD__.':'."called with $_string and CHARSET $displayCharset");
|
||||
if(function_exists('imap_mime_header_decode'))
|
||||
{
|
||||
// some characterreplacements, as they fail to translate
|
||||
$sar = array(
|
||||
'@(\x84|\x93|\x94)@',
|
||||
'@(\x96|\x97|\x1a)@',
|
||||
'@(\x91|\x92)@',
|
||||
'@(\x85)@',
|
||||
'@(\x86)@',
|
||||
);
|
||||
$rar = array(
|
||||
'"',
|
||||
'-',
|
||||
'\'',
|
||||
'...',
|
||||
'+',
|
||||
);
|
||||
|
||||
$newString = '';
|
||||
|
||||
$string = preg_replace('/\?=\s+=\?/', '?= =?', $_string);
|
||||
|
||||
$elements=imap_mime_header_decode($string);
|
||||
|
||||
$convertAtEnd = false;
|
||||
foreach((array)$elements as $element)
|
||||
{
|
||||
if ($element->charset == 'default') $element->charset = self::detect_encoding($element->text);
|
||||
if ($element->charset != 'x-unknown')
|
||||
{
|
||||
if( strtoupper($element->charset) != 'UTF-8') $element->text = preg_replace($sar,$rar,$element->text);
|
||||
// check if there is a possible nested encoding; make sure that the inputstring and the decoded result are different to avoid loops
|
||||
if(preg_match('/\?=.+=\?/', $element->text) && $element->text != $_string)
|
||||
{
|
||||
$element->text = self::decodeMailHeader($element->text, $element->charset);
|
||||
$element->charset = $displayCharset;
|
||||
}
|
||||
$newString .= self::convert($element->text,$element->charset);
|
||||
}
|
||||
else
|
||||
{
|
||||
$newString .= $element->text;
|
||||
$convertAtEnd = true;
|
||||
}
|
||||
}
|
||||
if ($convertAtEnd) $newString = self::decodeMailHeader($newString,$displayCharset);
|
||||
return preg_replace('/([\000-\012\015\016\020-\037\075])/','',$newString);
|
||||
}
|
||||
elseif(function_exists(mb_decode_mimeheader))
|
||||
{
|
||||
$matches = null;
|
||||
if(preg_match_all('/=\?.*\?Q\?.*\?=/iU', $string=$_string, $matches))
|
||||
{
|
||||
foreach($matches[0] as $match)
|
||||
{
|
||||
$fixedMatch = str_replace('_', ' ', $match);
|
||||
$string = str_replace($match, $fixedMatch, $string);
|
||||
}
|
||||
$string = str_replace('=?ISO8859-','=?ISO-8859-',
|
||||
str_replace('=?windows-1258','=?ISO-8859-1',$string));
|
||||
}
|
||||
$string = mb_decode_mimeheader($string);
|
||||
return preg_replace('/([\000-\012\015\016\020-\037\075])/','',$string);
|
||||
}
|
||||
elseif(function_exists(iconv_mime_decode))
|
||||
{
|
||||
// continue decoding also if an error occurs
|
||||
$string = @iconv_mime_decode($_string, 2, $displayCharset);
|
||||
return preg_replace('/([\000-\012\015\016\020-\037\075])/','',$string);
|
||||
}
|
||||
|
||||
// no decoding function available
|
||||
return preg_replace('/([\000-\012\015\016\020-\037\075])/','',$_string);
|
||||
return Api\Mail\Html::decodeMailHeader($_string, $displayCharset);
|
||||
}
|
||||
|
||||
/**
|
||||
@ -110,21 +38,11 @@ class translation extends Api\Translation
|
||||
* as well as those emailadresses in links, and within broken links
|
||||
* @param string the text to process
|
||||
* @return 1
|
||||
* @deprecated use Api\Mail\Html::replaceEmailAdresses
|
||||
*/
|
||||
static function replaceEmailAdresses(&$text)
|
||||
{
|
||||
//error_log($text);
|
||||
//replace CRLF with something other to be preserved via preg_replace as CRLF seems to vanish
|
||||
$text2 = str_replace("\r\n",'<#cr-lf#>',$text);
|
||||
// replace emailaddresses eclosed in <> (eg.: <me@you.de>) with the emailaddress only (e.g: me@you.de)
|
||||
$text3 = preg_replace("/(<|<a href=\")*(mailto:([\w\.,-.,_.,0-9.]+)(@)([\w\.,-.,_.,0-9.]+))(>|>)*/i","$2 ", $text2);
|
||||
//$text = preg_replace_callback("/(<|<a href=\")*(mailto:([\w\.,-.,_.,0-9.]+)(@)([\w\.,-.,_.,0-9.]+))(>|>)*/i",'self::transform_mailto2text',$text);
|
||||
//$text = preg_replace('~<a[^>]+href=\"(mailto:)+([^"]+)\"[^>]*>~si','$2 ',$text);
|
||||
$text4 = preg_replace_callback('~<a[^>]+href=\"(mailto:)+([^"]+)\"[^>]*>([ @\w\.,-.,_.,0-9.]+)<\/a>~si','self::transform_mailto2text',$text3);
|
||||
$text5 = preg_replace("/(([\w\.,-.,_.,0-9.]+)(@)([\w\.,-.,_.,0-9.]+))( |\s)*(<\/a>)*( |\s)*(>|>)*/i","$1 ", $text4);
|
||||
$text6 = preg_replace("/(<|<)*(([\w\.,-.,_.,0-9.]+)@([\w\.,-.,_.,0-9.]+))(>|>)*/i","$2 ", $text5);
|
||||
$text = str_replace('<#cr-lf#>',"\r\n",$text6);
|
||||
return 1;
|
||||
return Api\Mail\Html::replaceEmailAdresses($text);
|
||||
}
|
||||
|
||||
/**
|
||||
@ -135,81 +53,21 @@ class translation extends Api\Translation
|
||||
* @param string $endtag can be different from tag but should be used only, if begin and endtag are known to be different e.g.: <!-- -->
|
||||
* @param bool $addbracesforendtag if endtag is given, you may decide if the </ and > braces are to be added,
|
||||
* or if you want the string to be matched as is
|
||||
* @return void the modified text is passed via reference
|
||||
* @deprecated use Api\Mail\Html::replaceTagsCompletley
|
||||
*/
|
||||
static function replaceTagsCompletley(&$_body,$tag,$endtag='',$addbracesforendtag=true)
|
||||
{
|
||||
if ($tag) $tag = strtolower($tag);
|
||||
$singleton = false;
|
||||
if ($endtag=='/>') $singleton =true;
|
||||
if ($endtag == '' || empty($endtag) || !isset($endtag))
|
||||
{
|
||||
$endtag = $tag;
|
||||
} else {
|
||||
$endtag = strtolower($endtag);
|
||||
//error_log(__METHOD__.' Using EndTag:'.$endtag);
|
||||
}
|
||||
// strip tags out of the message completely with their content
|
||||
if ($_body) {
|
||||
if ($singleton)
|
||||
{
|
||||
//$_body = preg_replace('~<'.$tag.'[^>].*? '.$endtag.'~simU','',$_body);
|
||||
$_body = preg_replace('~<?'.$tag.'[^>].* '.$endtag.'~simU','',$_body); // we are in Ungreedy mode, so we expect * to be ungreedy without specifying ?
|
||||
}
|
||||
else
|
||||
{
|
||||
$found=null;
|
||||
if ($addbracesforendtag === true )
|
||||
{
|
||||
if (stripos($_body,'<'.$tag)!==false) $ct = preg_match_all('#<'.$tag.'(?:\s.*)?>(.+)</'.$endtag.'>#isU', $_body, $found);
|
||||
if ($ct>0)
|
||||
{
|
||||
//error_log(__METHOD__.__LINE__.array2string($found[0]));
|
||||
// only replace what we have found
|
||||
$_body = str_ireplace($found[0],'',$_body);
|
||||
}
|
||||
// remove left over tags, unfinished ones, and so on
|
||||
$_body = preg_replace('~<'.$tag.'[^>]*?>~si','',$_body);
|
||||
}
|
||||
if ($addbracesforendtag === false )
|
||||
{
|
||||
if (stripos($_body,'<'.$tag)!==false) $ct = preg_match_all('#<'.$tag.'(?:\s.*)?>(.+)'.$endtag.'#isU', $_body, $found);
|
||||
if ($ct>0)
|
||||
{
|
||||
//error_log(__METHOD__.__LINE__.array2string($found[0]));
|
||||
// only replace what we have found
|
||||
$_body = str_ireplace($found[0],'',$_body);
|
||||
}
|
||||
/*
|
||||
$_body = preg_replace('~<'.$tag.'[^>]*?>(.*?)'.$endtag.'~simU','',$_body);
|
||||
*/
|
||||
// remove left over tags, unfinished ones, and so on
|
||||
$_body = preg_replace(array('~<'.$tag.'[^>]*?>~si', '~'.$endtag.'~'), '', $_body);
|
||||
}
|
||||
}
|
||||
}
|
||||
Api\Mail\Html::replaceTagsCompletley($_body, $tag, $endtag, $addbracesforendtag);
|
||||
}
|
||||
|
||||
static function transform_mailto2text($matches)
|
||||
{
|
||||
//error_log(__METHOD__.__LINE__.array2string($matches));
|
||||
// this is the actual url
|
||||
$matches[2] = trim(strip_tags($matches[2]));
|
||||
$matches[3] = trim(strip_tags($matches[3]));
|
||||
$matches[2] = str_replace(array('%40','%20'),array('@',' '),$matches[2]);
|
||||
$matches[3] = str_replace(array('%40','%20'),array('@',' '),$matches[3]);
|
||||
return $matches[1].$matches[2].($matches[2]==$matches[3]?' ':' -> '.$matches[3].' ');
|
||||
return Api\Mail\Html::transform_mailto2text($matches);
|
||||
}
|
||||
|
||||
static function transform_url2text($matches)
|
||||
{
|
||||
//error_log(__METHOD__.__LINE__.array2string($matches));
|
||||
$linkTextislink = false;
|
||||
// this is the actual url
|
||||
$matches[2] = trim(strip_tags($matches[2]));
|
||||
if ($matches[2]==$matches[1]) $linkTextislink = true;
|
||||
$matches[1] = str_replace(' ','%20',$matches[1]);
|
||||
return ($linkTextislink?' ':'[ ').$matches[1].($linkTextislink?'':' -> '.$matches[2]).($linkTextislink?' ':' ]');
|
||||
return Api\Mail\Html::transform_url2text($matches);
|
||||
}
|
||||
|
||||
/**
|
||||
@ -219,244 +77,11 @@ class translation extends Api\Translation
|
||||
* @param bool $stripcrl : flag to indicate for the removal of all crlf \r\n
|
||||
* @param bool $stripalltags : flag to indicate wether or not to strip $_html from all remaining tags
|
||||
* @return text $_html : the modified text.
|
||||
* @deprecated use Api\Mail\Html::convertHTMLToText
|
||||
*/
|
||||
static function convertHTMLToText($_html,$displayCharset=false,$stripcrl=false,$stripalltags=true)
|
||||
{
|
||||
// assume input isHTML, but test the input anyway, because,
|
||||
// if it is not, we may not want to strip whitespace
|
||||
$isHTML = true;
|
||||
if (strlen(strip_tags($_html)) == strlen($_html))
|
||||
{
|
||||
$isHTML = false;
|
||||
// return $_html; // maybe we should not proceed at all
|
||||
}
|
||||
if ($displayCharset === false) $displayCharset = self::$system_charset;
|
||||
//error_log(__METHOD__.$_html);
|
||||
#print '<hr>';
|
||||
#print "<pre>"; print htmlspecialchars($_html);
|
||||
#print "</pre>";
|
||||
#print "<hr>";
|
||||
if (stripos($_html,'style')!==false) self::replaceTagsCompletley($_html,'style'); // clean out empty or pagewide style definitions / left over tags
|
||||
if (stripos($_html,'head')!==false) self::replaceTagsCompletley($_html,'head'); // Strip out stuff in head
|
||||
if (stripos($_html,'![if')!==false && stripos($_html,'<![endif]>')!==false) self::replaceTagsCompletley($_html,'!\[if','<!\[endif\]>',false); // Strip out stuff in ifs
|
||||
if (stripos($_html,'!--[if')!==false && stripos($_html,'<![endif]-->')!==false) self::replaceTagsCompletley($_html,'!--\[if','<!\[endif\]-->',false); // Strip out stuff in ifs
|
||||
$Rules = array ('@<script[^>]*?>.*?</script>@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, $_html);
|
||||
|
||||
// removing carriage return linefeeds, preserve those enclosed in <pre> </pre> tags
|
||||
if ($stripcrl === true )
|
||||
{
|
||||
if (stripos($_html,'<pre ')!==false || stripos($_html,'<pre>')!==false)
|
||||
{
|
||||
$contentArr = self::splithtmlByPRE($_html);
|
||||
foreach ($contentArr as $k =>&$elem)
|
||||
{
|
||||
if (stripos($elem,'<pre ')===false && stripos($elem,'<pre>')===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 => '~<h[123][^>]*>\r*\n*~si',
|
||||
1 => '~<h[456][^>]*>\r*\n*~si',
|
||||
2 => '~<table[^>]*>\r*\n*~si',
|
||||
3 => '~<tr[^>]*>\r*\n*~si',
|
||||
4 => '~<li[^>]*>\r*\n*~si',
|
||||
5 => '~<br[^>]*>\r*\n*~si',
|
||||
6 => '~<br[^>]*>~si',
|
||||
7 => '~<p[^>]*>\r*\n*~si',
|
||||
8 => '~<div[^>]*>\r*\n*~si',
|
||||
9 => '~<hr[^>]*>\r*\n*~si',
|
||||
10 => '/<blockquote type="cite">/',
|
||||
11 => '/<blockquote>/',
|
||||
12 => '~</blockquote>~si',
|
||||
13 => '~<blockquote[^>]*>~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);
|
||||
$_html = preg_replace('~</t(d|h)>\s*<t(d|h)[^>]*>~si',' - ',$_html);
|
||||
$_html = preg_replace('~<img[^>]+>~s','',$_html);
|
||||
// replace emailaddresses eclosed in <> (eg.: <me@you.de>) with the emailaddress only (e.g: me@you.de)
|
||||
self::replaceEmailAdresses($_html);
|
||||
//convert hrefs to description -> URL
|
||||
//$_html = preg_replace('~<a[^>]+href=\"([^"]+)\"[^>]*>(.*)</a>~si','[$2 -> $1]',$_html);
|
||||
$_html = preg_replace_callback('~<a[^>]+href=\"([^"]+)\"[^>]*>(.*?)</a>~si','self::transform_url2text',$_html);
|
||||
|
||||
// reducing double \r\n to single ones, dont mess with pre sections
|
||||
if ($stripcrl === true && $isHTML)
|
||||
{
|
||||
if (stripos($_html,'<pre ')!==false || stripos($_html,'<pre>')!==false)
|
||||
{
|
||||
$contentArr = self::splithtmlByPRE($_html);
|
||||
foreach ($contentArr as $k =>&$elem)
|
||||
{
|
||||
if (stripos($elem,'<pre ')===false && stripos($elem,'<pre>')===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, '<a>');
|
||||
}
|
||||
// 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);
|
||||
}
|
||||
return Api\Mail\Html::convertHTMLToText($_html, $displayCharset, $stripcrl, $stripalltags);
|
||||
}
|
||||
|
||||
/**
|
||||
@ -464,37 +89,10 @@ class translation extends Api\Translation
|
||||
* @author Leithoff, Klaus
|
||||
* @param string html
|
||||
* @return mixed array of parts or unaffected html
|
||||
* @deprecated use Api\Mail\Html::splithtmlByPRE
|
||||
*/
|
||||
static function splithtmlByPRE($html)
|
||||
{
|
||||
$searchFor = '<pre ';
|
||||
$pos = stripos($html,$searchFor);
|
||||
if ($pos===false)
|
||||
{
|
||||
$searchFor = '<pre>';
|
||||
$pos = stripos($html,$searchFor);
|
||||
}
|
||||
if ($pos === false)
|
||||
{
|
||||
return $html;
|
||||
}
|
||||
$html2ret[] = substr($html,0,$pos);
|
||||
while ($pos!==false)
|
||||
{
|
||||
$endofpre = stripos($html,'</pre>',$pos);
|
||||
$length = $endofpre-$pos+6;
|
||||
$html2ret[] = substr($html,$pos,$length);
|
||||
$searchFor = '<pre ';
|
||||
$pos = stripos($html,$searchFor, $endofpre+6);
|
||||
if ($pos===false)
|
||||
{
|
||||
$searchFor = '<pre>';
|
||||
$pos = stripos($html,$searchFor, $endofpre+6);
|
||||
}
|
||||
$html2ret[] = ($pos ? substr($html,$endofpre+6,$pos-($endofpre+6)): substr($html,$endofpre+6));
|
||||
//$pos=false;
|
||||
}
|
||||
//error_log(__METHOD__.__LINE__.array2string($html2ret));
|
||||
return $html2ret;
|
||||
return Api\Mail\Html::splithtmlByPRE($html);
|
||||
}
|
||||
}
|
||||
|
||||
Loading…
Reference in New Issue
Block a user