note 40131 deleted from function.strip-tags by nlopess
| From: | nlopess@php.net | Date: | Thu, 11 Mar 2004 14:52:08 +0000 |
| Subject: | note 40131 deleted from function.strip-tags by nlopess | ||
| References: | 1 | Groups: | php.notes |
| Request: | Send a blank email to php-notes+get-66638@lists.php.net to get a copy of this message | ||
Note Submitter: fil (at) rezox (dot) com - www.RezoX.com
----
/***************************************
** MatchUpTags Function
** --------------------
** Written by Fil Laborde - email: fil (at) rezox (dot) com
**
** More thoroughly documented function at:
** http://www.rezox.com/tools/matchuptags.txt
****************************************
**
** FUNCTION:
** Find all the tags in the HTML:
** 1) Then close all the tags that are unclosed (in order of shortest tag to longest)
** 2) Remove closing tags that have no matching openner tag
**
** METHOD:
** We do a shortcut to prevent having to do too many loops (which is slow), by adding
** length to each tag, and then appending the tag count at the end (which allows the
** array_diff() function to work and yet distinguish between all tags)
**
** We order them from shortest to largest tag, and then alphabetically to guestimate the
** closing order of importance (so that things like </td></tr></table> would be
ordered
** correctly
** We create a duplicate lowercase version of the page, so that for strpos searches
** we can search for lowercase tags, and then do the modification in the mirrored normal
** HTML page
**
***************************************/
function MatchUpTags( $HTML ){
$HTML_Lowercase = strtolower($HTML);
$OpenArray = array();
$CloseArray = array();
$OpenCount = array();
$CloseCount = array();
preg_match_all( "'<([^ |^>]+)(
[^>]+)?>'s",$HTML_Lowercase,$TagArray,PREG_PATTERN_ORDER );
foreach( $TagArray[1] AS $Tag ){
if( $Tag[0]=='/' ){
$Tag = substr($Tag,1); #strip the first character '/'
$TagLen = strlen($Tag);
$CloseCount[$Tag]+=1;
$CloseArray[] = str_pad($TagLen,3,'0', STR_PAD_LEFT).'
'.$Tag.'_'.$CloseCount[$Tag];
} else { # open tags
$TagLen = strlen($Tag);
$OpenCount[$Tag]+=1;
$OpenArray[] = str_pad($TagLen,3,'0', STR_PAD_LEFT).'
'.$Tag.'_'.$OpenCount[$Tag];
};
};
sort($OpenArray,SORT_STRING);
$MissingClose = array_diff($OpenArray,$CloseArray);
$MissingOpen = array_diff($CloseArray,$OpenArray);
$TagAppend = '';
foreach ($MissingClose AS $Tag )
$TagAppend .= '</'.substr($Tag,4,strrpos($Tag,'_')-4).'>';
$HTML .= $TagAppend;
foreach( $MissingOpen AS $Tag ){
$TagCnt = intval(substr($Tag,strrpos($Tag,'_')+1));
$Tag = '</'.substr($Tag,4,strrpos($Tag,'_')-4).'>';
#reconstruct the closing tag
$TagLen = strlen($Tag);
$Pos = -1;
for( $i=0; $i<$TagCnt;$i++ )
# with Pos=-1, we start with Pos=0, and loop through each find
$Pos = strpos($HTML_Lowercase,$Tag,$Pos+1);
if( $Pos!==false )
$HTML = substr_replace( $HTML,str_repeat(' ',$TagLen),$Pos,$TagLen );
};
return $HTML;
};
** EXAMPLE **
$OrigHTML = "<b>sfsfdsdfsdf</b><font
size=3><table></ta</tableble>fdsdff<table
width='sdfsd'><tr><TD width='sdffd'>sdfsdffsd</td><div
class='text'>sdfsdfsdfsdf<span
group='dfssdsdf'>fsdsfsdf</div></div><strike
style='sdfsd'>ssdfsdfsf</font></b>";
$HTML = MatchUpTags($OrigHTML);
==============
EXAMPLE OUTPUT:
==============
Original HTML:
<b>sfsfdsdfsdf</b><font
size=3><table></ta</tableble>fdsdff<table
width='sdfsd'><tr><TD width='sdffd'>sdfsdffsd</td><div
class='text'>sdfsdfsdfsdf<span
group='dfssdsdf'>fsdsfsdf</div></div><strike
style='sdfsd'>ssdfsdfsf</font></b>
Cleaned up HTML, with appended missing closing tags & removed excess closing tags:
<b>sfsfdsdfsdf</b><font size=3><table> fdsdff<table
width='sdfsd'><tr><TD width='sdffd'>sdfsdffsd</td><div
class='text'>sdfsdfsdfsdf<span group='dfssdsdf'>fsdsfsdf</div>
<strike style='sdfsd'>ssdfsdfsf</font>
</tr></span></table></table></strike>