[geeklog-cvs] Geeklog-2/system Kses.php,NONE,1.1
tony at iowaoutdoors.org
tony at iowaoutdoors.org
Tue Jan 4 14:34:42 EST 2005
Update of /var/cvs/Geeklog-2/system
In directory www:/tmp/cvs-serv4537
Added Files:
Kses.php
Log Message:
Ported old class.kses.php over to PHP5. Most changes were either to use
PHP5's new OO features or to adhere to GL2 coding standards.
--- NEW FILE: Kses.php ---
<?php
/* Reminder: always indent with 4 spaces (no tabs). */
/**
* Geeklog 2
*
* License Details To Be Determined
*
*/
/*
* This is a port of the kses filter written in PHP4 over to PHP5. The PHP4 version was a
* fork of a slick piece of procedural code called 'kses' written by Ulf Harnhammar
* The entire set of functions was wrapped in a PHP object with some internal modifications
* by Richard Vasquez (http://www.chaos.org/) 7/25/2003
*
* The original (procedural) version of the code can be found at:
* http://sourceforge.net/projects/kses/
*
* [kses strips evil scripts!]
*
* ==========================================================================================
*
* Geeklog_Kses - PHP5 port of class.kses.php v0.0.2 written by Richard R. Vasquez, Jr.
*
* Copyright (C) 2005 Tony Bibbs <tony at geeklog.net>
* class.kses.php v0.0.2 - Copyright (C) 2003 Richard R. Vasquez, Jr.
*
* Derived from kses 0.2.1 - HTML/XHTML filter that only allows some elements and attributes
* Copyright (C) 2002, 2003 Ulf Harnhammar
*
* ==========================================================================================
*
* This program is free software and open source software; you can redistribute
* it and/or modify it under the terms of the GNU General Public License as
* published by the Free Software Foundation; either version 2 of the License,
* or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
* more details.
*
* You should have received a copy of the GNU General Public License along
* with this program; if not, write to the Free Software Foundation, Inc.,
* 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA or visit
* http://www.gnu.org/licenses/gpl.html
*
* ==========================================================================================
* CONTACT INFORMATION:
*
* Email: View current valid email address at http://www.chaos.org/contact/
*/
class Geeklog_Kses
{
private $allowedProtocols = array('http', 'https', 'ftp', 'news', 'nntp', 'telnet', 'gopher', 'mailto');
private $allowedHTML = array();
public function parse($string = '')
{
if (get_magic_quotes_gpc())
{
$string = stripslashes($string);
}
$string = $this->noNull($string);
$string = $this->jsEntities($string);
$string = $this->normalizeEntities($string);
$string = $this->_hook($string);
return $this->split($string);
}
public function protocols()
{
$c_args = func_num_args();
if ($c_args != 1)
{
return false;
}
$protocol_data = func_get_arg(0);
if (is_array($protocol_data))
{
foreach ($protocol_data as $protocol)
{
$this->addProtocol($protocol);
}
}
elseif (is_string($protocol_data))
{
$this->addProtocol($protocol_data);
return true;
}
else
{
throw new Exception('kses::Protocols() did not receive a string or an array.');
}
}
public function addProtocol($protocol = '')
{
if (!is_string($protocol))
{
throw new Exception('kses::addProtocol() requires a string.');
}
$protocol = strtolower(trim($protocol));
if ($protocol == '')
{
throw new Exception('kses::addProtocol() tried to add an empty/NULL protocol.');
}
// Remove any inadvertent ':' at the end of the protocol.
if (substr($protocol, strlen($protocol) - 1, 1) == ':')
{
$protocol = substr($protocol, 0, strlen($protocol) - 1);
}
if (!in_array($protocol, $this->allowedProtocols))
{
array_push($this->allowedProtocols, $protocol);
sort($this->allowedProtocols);
}
return true;
}
public function addHTML($tag = '', $attribs = array())
{
if (!is_string($tag))
{
throw new Exception('kses::addHTML() requires the tag to be a string');
}
$tag = strtolower(trim($tag));
if ($tag == '')
{
throw new Exception('kses::addHTML() tried to add an empty/NULL tag');
}
if (!is_array($attribs))
{
throw new Exception("kses::addHTML() requires an array (even an empty one) of attributes for '$tag'");
}
$new_attribs = array();
foreach ($attribs as $idx1 => $val1)
{
$new_idx1 = strtolower($idx1);
$new_val1 = $attribs[$idx1];
if (is_array($new_val1))
{
$tmp_val = array();
foreach ($new_val1 as $idx2 => $val2)
{
$new_idx2 = strtolower($idx2);
$tmp_val[$new_idx2] = $val2;
}
$new_val1 = $tmp_val;
}
$new_attribs[$new_idx1] = $new_val1;
}
$this->allowedHTML[$tag] = $new_attribs;
return true;
}
/**
* This function removes any NULL or chr(173) characters in $string.
*/
protected function noNull($string)
{
$string = preg_replace('/\0+/', '', $string);
$string = preg_replace('/(\\\\0)+/', '', $string);
$string = preg_replace('/\xad+/', '', $string); //deals with Opera "feature"
return $string;
}
/**
* This function removes the HTML JavaScript entities found in early versions of
* Netscape 4.
*/
protected function jsEntities($string)
{
return preg_replace('%&\s*\{[^}]*(\}\s*;?|$)%', '', $string);
}
/**
* This function normalizes HTML entities. It will convert "AT&T" to the correct
* "AT&T", ":" to ":", "&#XYZZY;" to "&#XYZZY;" and so on.
*/
protected function normalizeEntities($string)
{
// Disarm all entities by converting & to &
$string = str_replace('&', '&', $string);
// Change back the allowed entities in our entity white list
$string = preg_replace('/&([A-Za-z][A-Za-z0-9]{0,19});/', '&\\1;', $string);
$string = preg_replace('/�*([0-9]{1,5});/e', '\$this->normalizeEntities2("\\1")', $string);
$string = preg_replace('/&#([Xx])0*(([0-9A-Fa-f]{2}){1,2});/', '&#\\1\\2;', $string);
return $string;
}
/**
* This function helps normalizeEntities() to only accept 16 bit values
* and nothing more for &#number; entities.
*/
protected function normalizeEntities2($i)
{
return (($i > 65535) ? "&#$i;" : "&#$i;");
}
/**
* You add any kses hooks here.
*/
protected function _hook($string)
{
return $string;
}
/**
* This function goes through an array, and changes the keys to all lower case.
*/
protected function arrayLC($inarray)
{
$outarray = array();
foreach ($inarray as $inkey => $inval)
{
$outkey = strtolower($inkey);
$outarray[$outkey] = array();
foreach ($inval as $inkey2 => $inval2)
{
$outkey2 = strtolower($inkey2);
$outarray[$outkey][$outkey2] = $inval2;
}
}
return $outarray;
}
/**
* This function searches for HTML tags, no matter how malformed. It also
* matches stray ">" characters.
*/
protected function split($string)
{
return preg_replace(
'%(<'. // EITHER: <
'[^>]*'. // things that aren't >
'(>|$)'. // > or end of string
'|>)%e', // OR: just a >
"\$this->split2('\\1')",
$string);
}
/**
* This function does a lot of work. It rejects some very malformed things
* like <:::>. It returns an empty string, if the element isn't allowed (look
* ma, no strip_tags()!). Otherwise it splits the tag into an element and an
* attribute list.
*/
protected function split2($string)
{
$string = $this->stripslashes($string);
if (substr($string, 0, 1) != '<')
{
// It matched a ">" character
return '>';
}
if (!preg_match('%^<\s*(/\s*)?([a-zA-Z0-9]+)([^>]*)>?$%', $string, $matches))
{
// It's seriously malformed
return '';
}
$slash = trim($matches[1]);
$elem = $matches[2];
$attrlist = $matches[3];
if (!is_array($this->allowedHTML[strtolower($elem)]))
{
// They are using a not allowed HTML element
return '';
}
return $this->attr("$slash$elem", $attrlist);
}
/**
* This function removes all attributes, if none are allowed for this element.
* If some are allowed it calls shair() to split them further, and then it
* builds up new HTML code from the data that hair() returns. It also
* removes "<" and ">" characters, if there are any left. One more thing it
* does is to check if the tag has a closing XHTML slash, and if it does,
* it puts one in the returned code as well.
*/
protected function attr($element, $attr)
{
// Is there a closing XHTML slash at the end of the attributes?
$xhtml_slash = '';
if (preg_match('%\s/\s*$%', $attr))
{
$xhtml_slash = ' /';
}
// Are any attributes allowed at all for this element?
if (count($this->allowedHTML[strtolower($element)]) == 0)
{
return "<$element$xhtml_slash>";
}
// Split it
$attrarr = $this->hair($attr);
// Go through $attrarr, and save the allowed attributes for this element
// in $attr2
$attr2 = '';
foreach ($attrarr as $arreach)
{
$current = $this->allowedHTML[strtolower($element)][strtolower($arreach['name'])];
if ($current == '')
{
// the attribute is not allowed
continue;
}
if (!is_array($current))
{
// there are no checks
$attr2 .= ' '.$arreach['whole'];
}
else
{
// there are some checks
$ok = true;
foreach ($current as $currkey => $currval)
{
if (!$this->checkAttVal($arreach['value'],
$arreach['vless'], $currkey, $currval))
{
$ok = false;
break;
}
}
if ($ok)
{
// it passed them
$attr2 .= ' '.$arreach['whole'];
}
}
}
// Remove any "<" or ">" characters
$attr2 = preg_replace('/[<>]/', '', $attr2);
return "<$element$attr2$xhtml_slash>";
}
/**
* This function does a lot of work. It parses an attribute list into an array
* with attribute data, and tries to do the right thing even if it gets weird
* input. It will add quotes around attribute values that don't have any quotes
* or apostrophes around them, to make it easier to produce HTML code that will
* conform to W3C's HTML specification. It will also remove bad URL protocols
* from attribute values.
*/
protected function hair($attr)
{
$attrarr = array();
$mode = 0;
$attrname = '';
// Loop through the whole attribute list
while (strlen($attr) != 0)
{
// Was the last operation successful?
$working = 0;
switch ($mode)
{
case 0: // attribute name, href for instance
if (preg_match('/^([-a-zA-Z]+)/', $attr, $match))
{
$attrname = $match[1];
$working = $mode = 1;
$attr = preg_replace('/^[-a-zA-Z]+/', '', $attr);
}
break;
case 1: // equals sign or valueless ("selected")
if (preg_match('/^\s*=\s*/', $attr)) // equals sign
{
$working = 1;
$mode = 2;
$attr = preg_replace('/^\s*=\s*/', '', $attr);
break;
}
if (preg_match('/^\s+/', $attr)) // valueless
{
$working = 1;
$mode = 0;
$attrarr[] = array(
'name' => $attrname,
'value' => '',
'whole' => $attrname,
'vless' => 'y'
);
$attr = preg_replace('/^\s+/', '', $attr);
}
break;
case 2: // attribute value, a URL after href= for instance
if (preg_match('/^"([^"]*)"(\s+|$)/', $attr, $match)) //"value"
{
$thisval = $this->badProtocol($match[1]);
$attrarr[] = array(
'name' => $attrname,
'value' => $thisval,
'whole' => "$attrname=\"$thisval\"",
'vless' => 'n'
);
$working = 1;
$mode = 0;
$attr = preg_replace('/^"[^"]*"(\s+|$)/', '', $attr);
break;
}
if (preg_match("/^'([^']*)'(\s+|$)/", $attr, $match)) //'value'
{
$thisval = $this->badProtocol($match[1]);
$attrarr[] = array(
'name' => $attrname,
'value' => $thisval,
'whole' => "$attrname='$thisval'",
'vless' => 'n'
);
$working = 1;
$mode = 0;
$attr = preg_replace("/^'[^']*'(\s+|$)/", '', $attr);
break;
}
if (preg_match("%^([^\s\"']+)(\s+|$)%", $attr, $match)) // value
{
$thisval = $this->badProtocol($match[1]);
$attrarr[] = array(
'name' => $attrname,
'value' => $thisval,
'whole' => "$attrname=\"$thisval\"",
'vless' => 'n'
);
// We add quotes to conform to W3C's HTML spec.
$working = 1;
$mode = 0;
$attr = preg_replace("%^[^\s\"']+(\s+|$)%", '', $attr);
}
break;
}
if ($working == 0) // not well formed, remove and try again
{
$attr = $this->htmlError($attr);
$mode = 0;
}
}
//special case, for when the attribute list ends with a valueless
//attribute like "selected"
if ($mode == 1)
{
$attrarr[] = array(
'name' => $attrname,
'value' => '',
'whole' => $attrname,
'vless' => 'y'
);
}
return $attrarr;
}
/**
* This function removes all non-allowed protocols from the beginning of
* $string. It ignores whitespace and the case of the letters, and it does
* understand HTML entities. It does its work in a while loop, so it won't be
* fooled by a string like "javascript:javascript:alert(57)".
*/
protected function badProtocol($string)
{
$string = $this->noNull($string);
$string2 = $string.'a';
while ($string != $string2)
{
$string2 = $string;
$string = $this->badProtocolOnce($string);
}
return $string;
}
/**
* This function searches for URL protocols at the beginning of $string, while
* handling whitespace and HTML entities.
*/
protected function badProtocolOnce($string)
{
return preg_replace(
'/^((&[^;]*;|[\sA-Za-z0-9])*)'.
'(:|:|&#[Xx]3[Aa];)\s*/e',
'\$this->badProtocolOnce2("\\1")',
$string
);
return $string;
}
/**
* This function processes URL protocols, checks to see if they're in the white-
* list or not, and returns different data depending on the answer.
*/
function badProtocolOnce2($string)
{
$string2 = $this->decodeEntities($string2);
$string2 = preg_replace('/\s/', '', $string);
$string2 = $this->noNull($string2);
$string2 = strtolower($string2);
$allowed = false;
foreach ($this->allowedProtocols as $one_protocol)
{
if (strtolower($one_protocol) == $string2)
{
$allowed = true;
break;
}
}
if ($allowed)
{
return "$string2:";
}
else
{
return '';
}
}
/**
* This function performs different checks for attribute values. The currently
* implemented checks are "maxlen", "minlen", "maxval", "minval" and "valueless"
* with even more checks to come soon.
*/
protected function checkAttVal($value, $vless, $checkname, $checkvalue)
{
$ok = true;
switch (strtolower($checkname))
{
// The maxlen check makes sure that the attribute value has a length not
// greater than the given value. This can be used to avoid Buffer Overflows
// in WWW clients and various Internet servers.
case 'maxlen':
if (strlen($value) > $checkvalue)
{
$ok = false;
}
break;
// The minlen check makes sure that the attribute value has a length not
// smaller than the given value.
case 'minlen':
if (strlen($value) < $checkvalue)
{
$ok = false;
}
break;
// The maxval check does two things: it checks that the attribute value is
// an integer from 0 and up, without an excessive amount of zeroes or
// whitespace (to avoid Buffer Overflows). It also checks that the attribute
// value is not greater than the given value.
// This check can be used to avoid Denial of Service attacks.
case 'maxval':
if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
{
$ok = false;
}
if ($value > $checkvalue)
{
$ok = false;
}
break;
// The minval check checks that the attribute value is a positive integer,
// and that it is not smaller than the given value.
case 'minval':
if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
{
$ok = false;
}
if ($value < $checkvalue)
{
$ok = false;
}
break;
// The valueless check checks if the attribute has a value
// (like <a href="blah">) or not (<option selected>). If the given value
// is a "y" or a "Y", the attribute must not have a value.
// If the given value is an "n" or an "N", the attribute must have one.
case 'valueless':
if (strtolower($checkvalue) != $vless)
{
$ok = false;
}
break;
} //switch
return $ok;
} //function checkAttVal
/**
* This function changes the character sequence \" to just "
* It leaves all other slashes alone. It's really weird, but the quoting from
* preg_replace(//e) seems to require this.
*/
protected function stripslashes($string)
{
return preg_replace('%\\\\"%', '"', $string);
}
/**
* This function deals with parsing errors in hair(). The general plan is
* to remove everything to and including some whitespace, but it deals with
* quotes and apostrophes as well.
*/
protected function htmlError($string)
{
return preg_replace('/^("[^"]*("|$)|\'[^\']*(\'|$)|\S)*\s*/', '', $string);
}
/**
* This function decodes numeric HTML entities (A and A). It doesn't
* do anything with other entities like ä, but we don't need them in the
* URL protocol white listing system anyway.
*/
protected function decodeEntities($string)
{
$string = preg_replace('/&#([0-9]+);/e', 'chr("\\1")', $string);
$string = preg_replace('/&#[Xx]([0-9A-Fa-f]+);/e', 'chr(hexdec("\\1"))', $string);
return $string;
}
// This function returns kses' version number.
public function version()
{
return '0.1 (PHP5 port of class.kses.php 0.0.2)';
}
}
?>
More information about the geeklog-cvs
mailing list