MDL-28627 remove buggy and obsolete KSES cleaning

This commit is contained in:
Petr Skoda 2011-10-22 09:25:06 +02:00
parent 3a81b37617
commit 73b309e6a3
10 changed files with 9 additions and 902 deletions

View file

@ -3716,14 +3716,6 @@ WHERE gradeitemid IS NOT NULL AND grademax IS NOT NULL");
upgrade_main_savepoint(true, 2010051801);
}
if ($oldversion < 2010052100) {
// Switch to html purifier as default cleaning engine - KSES is really very bad
if (empty($CFG->enablehtmlpurifier)) {
unset_config('enablehtmlpurifier');
}
upgrade_main_savepoint(true, 2010052100);
}
if ($oldversion < 2010052200) {
// Define field legacyfiles to be added to course - just in case we are upgrading from PR1
$table = new xmldb_table('course');
@ -6810,6 +6802,12 @@ FROM
upgrade_main_savepoint(true, 2011101200.01);
}
if ($oldversion < 2011101900.02) {
// remove unused setting
unset_config('enablehtmlpurifier');
upgrade_main_savepoint(true, 2011101900.02);
}
return true;
}

View file

@ -1,712 +0,0 @@
<?php
/**
* kses 0.2.2 - HTML/XHTML filter that only allows some elements and attributes
* Copyright (C) 2002, 2003, 2005 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 3 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 ***
*
* E-mail: metaur at users dot sourceforge dot net
* Web page: http://sourceforge.net/projects/kses
* Paper mail: Ulf Harnhammar
* Ymergatan 17 C
* 753 25 Uppsala
* SWEDEN
*
* [kses strips evil scripts!]
*
* @package moodlecore
* @copyright Ulf Harnhammar {@link http://sourceforge.net/projects/kses}
* @license http://www.gnu.org/copyleft/gpl.html GNU GPL v3 or later
*/
/**
* This function makes sure that only the allowed HTML element names, attribute
* names and attribute values plus only sane HTML entities will occur in
* $string. You have to remove any slashes from PHP's magic quotes before you
* call this function.
*
* @param string $string
* @param string $allowed_html
* @param array $allowed_protocols
* @return string
*/
function kses($string, $allowed_html, $allowed_protocols =
array('http', 'https', 'ftp', 'news', 'nntp', 'telnet',
'gopher', 'mailto'))
###############################################################################
# This function makes sure that only the allowed HTML element names, attribute
# names and attribute values plus only sane HTML entities will occur in
# $string. You have to remove any slashes from PHP's magic quotes before you
# call this function.
###############################################################################
{
$string = kses_no_null($string);
$string = kses_js_entities($string);
$string = kses_normalize_entities($string);
$string = kses_hook($string);
$allowed_html_fixed = kses_array_lc($allowed_html);
return kses_split($string, $allowed_html_fixed, $allowed_protocols);
} # function kses
/**
* You add any kses hooks here
*
* @param string $string
* @return string
*/
function kses_hook($string)
###############################################################################
# You add any kses hooks here.
###############################################################################
{
return $string;
} # function kses_hook
/**
* This function returns kses' version number.
*
* @return string
*/
function kses_version()
###############################################################################
# This function returns kses' version number.
###############################################################################
{
return '0.2.2';
} # function kses_version
/**
* This function searches for HTML tags, no matter how malformed. It also
* matches stray ">" characters.
*
* @param string $string
* @param string $allowed_html
* @param array $allowed_protocols
* @return string
*/
function kses_split($string, $allowed_html, $allowed_protocols)
###############################################################################
# This function searches for HTML tags, no matter how malformed. It also
# matches stray ">" characters.
###############################################################################
{
return preg_replace('%(<'. # EITHER: <
'[^>]*'. # things that aren't >
'(>|$)'. # > or end of string
'|>)%e', # OR: just a >
"kses_split2('\\1', \$allowed_html, ".
'$allowed_protocols)',
$string);
} # function kses_split
/**
* 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.
*
* @param string $string
* @param string $allowed_html
* @param array $allowed_protocols
* @return string
*/
function kses_split2($string, $allowed_html, $allowed_protocols)
###############################################################################
# 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.
###############################################################################
{
$string = kses_stripslashes($string);
if (substr($string, 0, 1) != '<')
return '&gt;';
# It matched a ">" character
if (!preg_match('%^<\s*(/\s*)?([a-zA-Z0-9]+)([^>]*)>?$%', $string, $matches))
return '';
# It's seriously malformed
$slash = trim($matches[1]);
$elem = $matches[2];
$attrlist = $matches[3];
if (!@isset($allowed_html[strtolower($elem)]))
return '';
# They are using a not allowed HTML element
if ($slash != '')
return "<$slash$elem>";
# No attributes are allowed for closing elements
return kses_attr("$slash$elem", $attrlist, $allowed_html,
$allowed_protocols);
} # function kses_split2
/**
* This function removes all attributes, if none are allowed for this element.
* If some are allowed it calls kses_hair() to split them further, and then it
* builds up new HTML code from the data that kses_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.
*
* @param string $element
* @param string $attr
* @param string $allowed_html
* @param array $allowed_protocols
* @return string
*/
function kses_attr($element, $attr, $allowed_html, $allowed_protocols)
###############################################################################
# This function removes all attributes, if none are allowed for this element.
# If some are allowed it calls kses_hair() to split them further, and then it
# builds up new HTML code from the data that kses_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.
###############################################################################
{
# Is there a closing XHTML slash at the end of the attributes?
$xhtml_slash = '';
if (preg_match('%\s/\s*$%', $attr))
$xhtml_slash = ' /';
# Are any attributes allowed at all for this element?
if (@count($allowed_html[strtolower($element)]) == 0)
return "<$element$xhtml_slash>";
# Split it
$attrarr = kses_hair($attr, $allowed_protocols);
# Go through $attrarr, and save the allowed attributes for this element
# in $attr2
$attr2 = '';
foreach ($attrarr as $arreach)
{
if (!@isset($allowed_html[strtolower($element)]
[strtolower($arreach['name'])]))
continue; # the attribute is not allowed
$current = $allowed_html[strtolower($element)]
[strtolower($arreach['name'])];
if (!is_array($current))
$attr2 .= ' '.$arreach['whole'];
# there are no checks
else
{
# there are some checks
$ok = true;
foreach ($current as $currkey => $currval)
if (!kses_check_attr_val($arreach['value'], $arreach['vless'],
$currkey, $currval))
{ $ok = false; break; }
if ($ok)
$attr2 .= ' '.$arreach['whole']; # it passed them
} # if !is_array($current)
} # foreach
# Remove any "<" or ">" characters
$attr2 = preg_replace('/[<>]/', '', $attr2);
return "<$element$attr2$xhtml_slash>";
} # function kses_attr
/**
* 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.
*
* @param string $attr
* @param array $allowed_protocols
* @return array
*/
function kses_hair($attr, $allowed_protocols)
###############################################################################
# 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.
###############################################################################
{
$attrarr = array();
$mode = 0;
$attrname = '';
# Loop through the whole attribute list
while (strlen($attr) != 0)
{
$working = 0; # Was the last operation successful?
switch ($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, $match))
# "value"
{
// MDL-2684 - kses stripping CSS styles that it thinks look like protocols
if ($attrname == 'style') {
$thisval = $match[1];
} else {
$thisval = kses_bad_protocol($match[1], $allowed_protocols);
}
$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'
{
$thisval = kses_bad_protocol($match[1], $allowed_protocols);
$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
{
$thisval = kses_bad_protocol($match[1], $allowed_protocols);
$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;
$attr = preg_replace("%^[^\s\"']+(\s+|$)%", '', $attr);
}
break;
} # switch
if ($working == 0) # not well formed, remove and try again
{
$attr = kses_html_error($attr);
$mode = 0;
}
} # while
if ($mode == 1)
# special case, for when the attribute list ends with a valueless
# attribute like "selected"
$attrarr[] = array
('name' => $attrname,
'value' => '',
'whole' => $attrname,
'vless' => 'y');
return $attrarr;
} # function kses_hair
/**
* 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.
*
* @param string $value
* @param string $vless
* @param string $checkname
* @param string $checkvalue
* @return bool
*/
function kses_check_attr_val($value, $vless, $checkname, $checkvalue)
###############################################################################
# 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.
###############################################################################
{
$ok = true;
switch (strtolower($checkname))
{
case 'maxlen':
# 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.
if (strlen($value) > $checkvalue)
$ok = false;
break;
case 'minlen':
# The minlen check makes sure that the attribute value has a length not
# smaller than the given value.
if (strlen($value) < $checkvalue)
$ok = false;
break;
case 'maxval':
# The maxval check does two things: it checks that the attribute value is
# an integer from 0 and up, without an excessive amount of zeroes or
# whitespace (to avoid Buffer Overflows). It also checks that the attribute
# value is not greater than the given value.
# This check can be used to avoid Denial of Service attacks.
if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
$ok = false;
if ($value > $checkvalue)
$ok = false;
break;
case 'minval':
# The minval check checks that the attribute value is a positive integer,
# and that it is not smaller than the given value.
if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
$ok = false;
if ($value < $checkvalue)
$ok = false;
break;
case 'valueless':
# The valueless check checks if the attribute has a value
# (like <a href="blah">) or not (<option selected>). If the given value
# is a "y" or a "Y", the attribute must not have a value.
# If the given value is an "n" or an "N", the attribute must have one.
if (strtolower($checkvalue) != $vless)
$ok = false;
break;
} # switch
return $ok;
} # function kses_check_attr_val
/**
* 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)".
*
* @param string $string
* @param array $$allowed_protocols
* @return string
*/
function kses_bad_protocol($string, $allowed_protocols)
###############################################################################
# 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)".
###############################################################################
{
$string = kses_no_null($string);
$string = preg_replace('/([^\xc3-\xcf])\xad+/', '\\1', $string); # deals with Opera "feature" -- moodle utf8 fix
$string2 = $string.'a';
while ($string != $string2)
{
$string2 = $string;
$string = kses_bad_protocol_once($string, $allowed_protocols);
} # while
return $string;
} # function kses_bad_protocol
/**
* This function removes any NULL characters in $string.
*
* @param string $string
* @return string
*/
function kses_no_null($string)
###############################################################################
# This function removes any NULL characters in $string.
###############################################################################
{
$string = preg_replace('/\0+/', '', $string);
$string = preg_replace('/(\\\\0)+/', '', $string);
return $string;
} # function kses_no_null
/**
* 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.
*
* @param string $string
* @return string
*/
function kses_stripslashes($string)
###############################################################################
# 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.
###############################################################################
{
return preg_replace('%\\\\"%', '"', $string);
} # function kses_stripslashes
/**
* This function goes through an array, and changes the keys to all lower case.
*
* @param array $inarray
* @return array
*/
function kses_array_lc($inarray)
###############################################################################
# This function goes through an array, and changes the keys to all lower case.
###############################################################################
{
$outarray = array();
foreach ($inarray as $inkey => $inval)
{
$outkey = strtolower($inkey);
$outarray[$outkey] = array();
foreach ($inval as $inkey2 => $inval2)
{
$outkey2 = strtolower($inkey2);
$outarray[$outkey][$outkey2] = $inval2;
} # foreach $inval
} # foreach $inarray
return $outarray;
} # function kses_array_lc
/**
* This function removes the HTML JavaScript entities found in early versions of
* Netscape 4.
*
* @param string $string
*/
function kses_js_entities($string)
###############################################################################
# This function removes the HTML JavaScript entities found in early versions of
# Netscape 4.
###############################################################################
{
return preg_replace('%&\s*\{[^}]*(\}\s*;?|$)%', '', $string);
} # function kses_js_entities
/**
* This function deals with parsing errors in kses_hair(). The general plan is
* to remove everything to and including some whitespace, but it deals with
* quotes and apostrophes as well.
*
* @param string $string
* @return string
*/
function kses_html_error($string)
###############################################################################
# This function deals with parsing errors in kses_hair(). The general plan is
# to remove everything to and including some whitespace, but it deals with
# quotes and apostrophes as well.
###############################################################################
{
return preg_replace('/^("[^"]*("|$)|\'[^\']*(\'|$)|\S)*\s*/', '', $string);
} # function kses_html_error
/**
* This function searches for URL protocols at the beginning of $string, while
* handling whitespace and HTML entities.
*
* @param string $string
* @param string $allowed_protocols
* @return string
*/
function kses_bad_protocol_once($string, $allowed_protocols)
###############################################################################
# This function searches for URL protocols at the beginning of $string, while
# handling whitespace and HTML entities.
###############################################################################
{
$string2 = preg_split('/:|&#58;|&#x3a;/i', $string, 2);
if(isset($string2[1]) && !preg_match('%/\?%',$string2[0]))
{
$string = kses_bad_protocol_once2($string2[0],$allowed_protocols).trim($string2[1]);
}
return $string;
} # function kses_bad_protocol_once
/**
* 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.
*
* @param string $string
* @param string $allowed_protocols
* @return string
*/
function kses_bad_protocol_once2($string, $allowed_protocols)
###############################################################################
# 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.
###############################################################################
{
$string2 = kses_decode_entities($string);
$string2 = preg_replace('/\s/', '', $string2);
$string2 = kses_no_null($string2);
$string2 = preg_replace('/\xad+/', '', $string2);
# deals with Opera "feature"
$string2 = strtolower($string2);
$allowed = false;
foreach ($allowed_protocols as $one_protocol)
if (strtolower($one_protocol) == $string2)
{
$allowed = true;
break;
}
if ($allowed)
return "$string2:";
else
return '';
} # function kses_bad_protocol_once2
/**
* This function normalizes HTML entities. It will convert "AT&T" to the correct
* "AT&amp;T", "&#00058;" to "&#58;", "&#XYZZY;" to "&amp;#XYZZY;" and so on.
*
* @param string $string
* @return string
*/
function kses_normalize_entities($string)
###############################################################################
# This function normalizes HTML entities. It will convert "AT&T" to the correct
# "AT&amp;T", "&#00058;" to "&#58;", "&#XYZZY;" to "&amp;#XYZZY;" and so on.
###############################################################################
{
# Disarm all entities by converting & to &amp;
$string = str_replace('&', '&amp;', $string);
# Change back the allowed entities in our entity whitelist
$string = preg_replace('/&amp;([A-Za-z][A-Za-z0-9]{0,19});/',
'&\\1;', $string);
$string = preg_replace('/&amp;#0*([0-9]{1,5});/e',
'kses_normalize_entities2("\\1")', $string);
$string = preg_replace('/&amp;#([Xx])0*(([0-9A-Fa-f]{2}){1,2});/',
'&#\\1\\2;', $string);
return $string;
} # function kses_normalize_entities
/**
* This function helps kses_normalize_entities() to only accept 16 bit values
* and nothing more for &#number; entities.
*
* @param int $i
* @return string
*/
function kses_normalize_entities2($i)
###############################################################################
# This function helps kses_normalize_entities() to only accept 16 bit values
# and nothing more for &#number; entities.
###############################################################################
{
return (($i > 65535) ? "&amp;#$i;" : "&#$i;");
} # function kses_normalize_entities2
/**
* This function decodes numeric HTML entities (&#65; and &#x41;). It doesn't
* do anything with other entities like &auml;, but we don't need them in the
* URL protocol whitelisting system anyway.
*
* @param string $string
* @return string
*/
function kses_decode_entities($string)
###############################################################################
# This function decodes numeric HTML entities (&#65; and &#x41;). It doesn't
# do anything with other entities like &auml;, but we don't need them in the
# URL protocol whitelisting system anyway.
###############################################################################
{
$string = preg_replace('/&#([0-9]+);/e', 'chr("\\1")', $string);
$string = preg_replace('/&#[Xx]([0-9A-Fa-f]+);/e', 'chr(hexdec("\\1"))',
$string);
return $string;
} # function kses_decode_entities
?>

View file

@ -811,9 +811,6 @@ if ($USER && function_exists('apache_note')
apache_note('MOODLEUSER', $logname);
}
// Adjust ALLOWED_TAGS
adjust_allowed_tags();
// Use a custom script replacement if one exists
if (!empty($CFG->customscripts)) {
if (($customscript = custom_script_path()) !== false) {

View file

@ -30,21 +30,16 @@ class purifier_test extends UnitTestCase {
public static $includecoverage = array('lib/htmlpurifier/HTMLPurifier.php');
private $enablehtmlpurifier = null;
private $cachetext = null;
function setUp() {
global $CFG;
$this->enablehtmlpurifier = $CFG->enablehtmlpurifier;
$CFG->enablehtmlpurifier = 1;
$this->cachetext = $CFG->cachetext;
$CFG->cachetext = 0;
}
function tearDown() {
global $CFG;
$CFG->enablehtmlpurifier = $this->enablehtmlpurifier;
$CFG->cachetext = $this->cachetext;
}

View file

@ -83,11 +83,6 @@ class web_test extends UnitTestCase {
$this->assertEqual(wikify_links('this is a <a href="http://someaddress.com/query">link</a>'), 'this is a link [ http://someaddress.com/query ]');
}
function test_fix_non_standard_entities() {
$this->assertEqual(fix_non_standard_entities('&#x00A3&#0228'), '&#xA3;&#228;');
$this->assertEqual(fix_non_standard_entities('&#x00A3;&#0228;'), '&#xA3;&#228;');
}
function test_compare_url() {
$url1 = new moodle_url('index.php', array('var1' => 1, 'var2' => 2));
$url2 = new moodle_url('index2.php', array('var1' => 1, 'var2' => 2, 'var3' => 3));

View file

@ -273,13 +273,6 @@
<version>1.0.0</version>
<licenseversion>3.0+</licenseversion>
</library>
<library>
<location>kses.php</location>
<name>KSES</name>
<license>GPL</license>
<version>0.2.2</version>
<licenseversion>3.0+</licenseversion>
</library>
<library>
<location>markdown.php</location>
<name>Markdown original+extra</name>

View file

@ -77,26 +77,6 @@ define('URL_MATCH_PARAMS', 1);
*/
define('URL_MATCH_EXACT', 2);
/**
* Allowed tags - string of html tags that can be tested against for safe html tags
* @global string $ALLOWED_TAGS
* @name $ALLOWED_TAGS
*/
global $ALLOWED_TAGS;
$ALLOWED_TAGS =
'<p><br><b><i><u><font><table><tbody><thead><tfoot><span><div><tr><td><th><ol><ul><dl><li><dt><dd><h1><h2><h3><h4><h5><h6><hr><img><a><strong><emphasis><em><sup><sub><address><cite><blockquote><pre><strike><param><acronym><nolink><lang><tex><algebra><math><mi><mn><mo><mtext><mspace><ms><mrow><mfrac><msqrt><mroot><mstyle><merror><mpadded><mphantom><mfenced><msub><msup><msubsup><munder><mover><munderover><mmultiscripts><mtable><mtr><mtd><maligngroup><malignmark><maction><cn><ci><apply><reln><fn><interval><inverse><sep><condition><declare><lambda><compose><ident><quotient><exp><factorial><divide><max><min><minus><plus><power><rem><times><root><gcd><and><or><xor><not><implies><forall><exists><abs><conjugate><eq><neq><gt><lt><geq><leq><ln><log><int><diff><partialdiff><lowlimit><uplimit><bvar><degree><set><list><union><intersect><in><notin><subset><prsubset><notsubset><notprsubset><setdiff><sum><product><limit><tendsto><mean><sdev><variance><median><mode><moment><vector><matrix><matrixrow><determinant><transpose><selector><annotation><semantics><annotation-xml><tt><code>';
/**
* Allowed protocols - array of protocols that are safe to use in links and so on
* @global string $ALLOWED_PROTOCOLS
*/
$ALLOWED_PROTOCOLS = array('http', 'https', 'ftp', 'news', 'mailto', 'rtsp', 'teamspeak', 'gopher', 'mms',
'color', 'callto', 'cursor', 'text-align', 'font-size', 'font-weight', 'font-style', 'font-family',
'border', 'border-bottom', 'border-left', 'border-top', 'border-right', 'margin', 'margin-bottom', 'margin-left', 'margin-top', 'margin-right',
'padding', 'padding-bottom', 'padding-left', 'padding-top', 'padding-right', 'vertical-align',
'background', 'background-color', 'text-decoration'); // CSS as well to get through kses
/// Functions
/**
@ -1309,19 +1289,6 @@ function wikify_links($string) {
return preg_replace('~(<a [^<]*href=["|\']?([^ "\']*)["|\']?[^>]*>([^<]*)</a>)~i','$3 [ $2 ]', $string);
}
/**
* Replaces non-standard HTML entities
*
* @param string $string
* @return string
*/
function fix_non_standard_entities($string) {
$text = preg_replace('/&#0*([0-9]+);?/', '&#$1;', $string);
$text = preg_replace('/&#x0*([0-9a-fA-F]+);?/', '&#x$1;', $text);
$text = preg_replace('[\x00-\x08\x0b-\x0c\x0e-\x1f]', '', $text);
return $text;
}
/**
* Given text in a variety of format codings, this function returns
* the text as plain text suitable for plain email.
@ -1436,8 +1403,6 @@ function trusttext_trusted($context) {
/**
* Is trusttext feature active?
*
* @global object
* @param object $context
* @return bool
*/
function trusttext_active() {
@ -1459,8 +1424,6 @@ function trusttext_active() {
* @return string The cleaned up text
*/
function clean_text($text, $format = FORMAT_HTML, $options = array()) {
global $ALLOWED_TAGS, $CFG;
if (empty($text) or is_numeric($text)) {
return (string)$text;
}
@ -1474,22 +1437,7 @@ function clean_text($text, $format = FORMAT_HTML, $options = array()) {
return $text;
}
if (!empty($CFG->enablehtmlpurifier)) {
$text = purify_html($text, $options);
} else {
/// Fix non standard entity notations
$text = fix_non_standard_entities($text);
/// Remove tags that are not allowed
$text = strip_tags($text, $ALLOWED_TAGS);
/// Clean up embedded scripts and , using kses
$text = cleanAttributes($text);
/// Again remove tags that are not allowed
$text = strip_tags($text, $ALLOWED_TAGS);
}
$text = purify_html($text, $options);
// Remove potential script events - some extra protection for undiscovered bugs in our code
$text = preg_replace("~([^a-z])language([[:space:]]*)=~i", "$1Xlanguage=", $text);
@ -1501,10 +1449,10 @@ function clean_text($text, $format = FORMAT_HTML, $options = array()) {
/**
* KSES replacement cleaning function - uses HTML Purifier.
*
* @global object
* @param string $text The (X)HTML string to purify
* @param array $options Array of options; currently only option supported is 'allowid' (if set,
* does not remove id attributes when cleaning)
* @return string
*/
function purify_html($text, $options = array()) {
global $CFG;
@ -1568,89 +1516,6 @@ function purify_html($text, $options = array()) {
return $text;
}
/**
* This function takes a string and examines it for HTML tags.
*
* If tags are detected it passes the string to a helper function {@link cleanAttributes2()}
* which checks for attributes and filters them for malicious content
*
* @param string $str The string to be examined for html tags
* @return string
*/
function cleanAttributes($str){
$result = preg_replace_callback(
'%(<[^>]*(>|$)|>)%m', #search for html tags
"cleanAttributes2",
$str
);
return $result;
}
/**
* This function takes a string with an html tag and strips out any unallowed
* protocols e.g. javascript:
*
* It calls ancillary functions in kses which are prefixed by kses
*
* @global object
* @global string
* @param array $htmlArray An array from {@link cleanAttributes()}, containing in its 1st
* element the html to be cleared
* @return string
*/
function cleanAttributes2($htmlArray){
global $CFG, $ALLOWED_PROTOCOLS;
require_once($CFG->libdir .'/kses.php');
$htmlTag = $htmlArray[1];
if (substr($htmlTag, 0, 1) != '<') {
return '&gt;'; //a single character ">" detected
}
if (!preg_match('%^<\s*(/\s*)?([a-zA-Z0-9]+)([^>]*)>?$%', $htmlTag, $matches)) {
return ''; // It's seriously malformed
}
$slash = trim($matches[1]); //trailing xhtml slash
$elem = $matches[2]; //the element name
$attrlist = $matches[3]; // the list of attributes as a string
$attrArray = kses_hair($attrlist, $ALLOWED_PROTOCOLS);
$attStr = '';
foreach ($attrArray as $arreach) {
$arreach['name'] = strtolower($arreach['name']);
if ($arreach['name'] == 'style') {
$value = $arreach['value'];
while (true) {
$prevvalue = $value;
$value = kses_no_null($value);
$value = preg_replace("/\/\*.*\*\//Us", '', $value);
$value = kses_decode_entities($value);
$value = preg_replace('/(&#[0-9]+)(;?)/', "\\1;", $value);
$value = preg_replace('/(&#x[0-9a-fA-F]+)(;?)/', "\\1;", $value);
if ($value === $prevvalue) {
$arreach['value'] = $value;
break;
}
}
$arreach['value'] = preg_replace("/j\s*a\s*v\s*a\s*s\s*c\s*r\s*i\s*p\s*t/i", "Xjavascript", $arreach['value']);
$arreach['value'] = preg_replace("/v\s*b\s*s\s*c\s*r\s*i\s*p\s*t/i", "Xvbscript", $arreach['value']);
$arreach['value'] = preg_replace("/e\s*x\s*p\s*r\s*e\s*s\s*s\s*i\s*o\s*n/i", "Xexpression", $arreach['value']);
$arreach['value'] = preg_replace("/b\s*i\s*n\s*d\s*i\s*n\s*g/i", "Xbinding", $arreach['value']);
} else if ($arreach['name'] == 'href') {
//Adobe Acrobat Reader XSS protection
$arreach['value'] = preg_replace('/(\.(pdf|fdf|xfdf|xdp|xfd)[^#]*)#.*$/i', '$1', $arreach['value']);
}
$attStr .= ' '.$arreach['name'].'="'.$arreach['value'].'"';
}
$xhtml_slash = '';
if (preg_match('%/\s*$%', $attrlist)) {
$xhtml_slash = ' /';
}
return '<'. $slash . $elem . $attStr . $xhtml_slash .'>';
}
/**
* Given plain text, makes it into HTML as nicely as possible.
* May contain HTML tags already
@ -1659,7 +1524,6 @@ function cleanAttributes2($htmlArray){
* by {@see format_text()} to convert FORMAT_MOODLE to HTML. You are supposed
* to call format_text() in most of cases.
*
* @global object
* @param string $text The string to convert.
* @param boolean $smiley_ignored Was used to determine if smiley characters should convert to smiley images, ignored now
* @param boolean $para If true then the returned string will be wrapped in div tags
@ -1668,8 +1532,6 @@ function cleanAttributes2($htmlArray){
*/
function text_to_html($text, $smiley_ignored=null, $para=true, $newlines=true) {
global $CFG;
/// Remove any whitespace that may be between HTML tags
$text = preg_replace("~>([[:space:]]+)<~i", "><", $text);
@ -2612,22 +2474,6 @@ function print_maintenance_message() {
die;
}
/**
* Adjust the list of allowed tags based on $CFG->allowobjectembed and user roles (admin)
*
* @global object
* @global string
* @return void
*/
function adjust_allowed_tags() {
global $CFG, $ALLOWED_TAGS;
if (!empty($CFG->allowobjectembed)) {
$ALLOWED_TAGS .= '<embed><object>';
}
}
/**
* A class for tabs, Some code to print tabs
*