mirror of
https://github.com/EGroupware/egroupware.git
synced 2024-11-22 16:03:47 +01:00
reintroduce tidy in calls from egw-mail-app when only getCleanHTML is called, as it is correcting html structure issues for us
This commit is contained in:
parent
9ec5116d7e
commit
714c75ef6d
@ -2132,6 +2132,22 @@ class mail_compose
|
||||
static $nonDisplayAbleCharacters = array('[\016]','[\017]',
|
||||
'[\020]','[\021]','[\022]','[\023]','[\024]','[\025]','[\026]','[\027]',
|
||||
'[\030]','[\031]','[\032]','[\033]','[\034]','[\035]','[\036]','[\037]');
|
||||
|
||||
if (extension_loaded('tidy'))
|
||||
{
|
||||
$tidy = new tidy();
|
||||
$cleaned = $tidy->repairString($_body, mail_bo::$tidy_config,'utf8');
|
||||
// Found errors. Strip it all so there's some output
|
||||
if($tidy->getStatus() == 2)
|
||||
{
|
||||
error_log(__METHOD__.' ('.__LINE__.') '.' ->'.$tidy->errorBuffer);
|
||||
}
|
||||
else
|
||||
{
|
||||
$_body = $cleaned;
|
||||
}
|
||||
}
|
||||
|
||||
mail_bo::getCleanHTML($_body);
|
||||
return preg_replace($nonDisplayAbleCharacters, '', $_body);
|
||||
}
|
||||
|
@ -2801,10 +2801,50 @@ class mail_ui
|
||||
$newBody = $singleBodyPart['body'];
|
||||
//TODO:$newBody = $this->highlightQuotes($newBody);
|
||||
#error_log(print_r($newBody,true));
|
||||
|
||||
if (extension_loaded('tidy'))
|
||||
{
|
||||
$tidy = new tidy();
|
||||
$cleaned = $tidy->repairString($newBody, mail_bo::$tidy_config,'utf8');
|
||||
// Found errors. Strip it all so there's some output
|
||||
if($tidy->getStatus() == 2)
|
||||
{
|
||||
error_log(__METHOD__.' ('.__LINE__.') '.' ->'.$tidy->errorBuffer);
|
||||
}
|
||||
else
|
||||
{
|
||||
$newBody = $cleaned;
|
||||
}
|
||||
if (!$preserveHTML)
|
||||
{
|
||||
// filter only the 'body', as we only want that part, if we throw away the html
|
||||
preg_match('`(<htm.+?<body[^>]*>)(.+?)(</body>.*?</html>)`ims', $newBody, $matches=array());
|
||||
if ($matches[2])
|
||||
{
|
||||
$hasOther = true;
|
||||
$newBody = $matches[2];
|
||||
}
|
||||
}
|
||||
}
|
||||
else
|
||||
{
|
||||
// htmLawed filter only the 'body'
|
||||
preg_match('`(<htm.+?<body[^>]*>)(.+?)(</body>.*?</html>)`ims', $newBody, $matches=array());
|
||||
if ($matches[2])
|
||||
{
|
||||
$hasOther = true;
|
||||
$newBody = $matches[2];
|
||||
}
|
||||
$htmLawed = new egw_htmLawed();
|
||||
// the next line should not be needed, but produces better results on HTML 2 Text conversion,
|
||||
// as we switched off HTMLaweds tidy functionality
|
||||
$newBody = str_replace(array('&amp;','<DIV><BR></DIV>',"<DIV> </DIV>",'<div> </div>'),array('&','<BR>','<BR>','<BR>'),$newBody);
|
||||
$newBody = $htmLawed->egw_htmLawed($newBody);
|
||||
if ($hasOther && $preserveHTML) $newBody = $matches[1]. $newBody. $matches[3];
|
||||
}
|
||||
// do the cleanup, set for the use of purifier
|
||||
$newBodyBuff = $newBody;
|
||||
//$newBodyBuff = $newBody;
|
||||
mail_bo::getCleanHTML($newBody);
|
||||
/*
|
||||
// in a way, this tests if we are having real utf-8 (the displayCharset) by now; we should if charsets reported (or detected) are correct
|
||||
if (strtoupper(mail_bo::$displayCharset) == 'UTF-8')
|
||||
{
|
||||
@ -2819,7 +2859,7 @@ class mail_ui
|
||||
mail_bo::$htmLawed_config['tidy'] = $tv;
|
||||
}
|
||||
}
|
||||
|
||||
*/
|
||||
// removes stuff between http and ?http
|
||||
$Protocol = '(http:\/\/|(ftp:\/\/|https:\/\/))'; // only http:// gets removed, other protocolls are shown
|
||||
$newBody = preg_replace('~'.$Protocol.'[^>]*\?'.$Protocol.'~sim','$1',$newBody); // removes stuff between http:// and ?http://
|
||||
|
Loading…
Reference in New Issue
Block a user