Commit 6c3317dd authored by Richard Mansfield's avatar Richard Mansfield

Files orphaned by HTML purifier changes

parent 992c2a60
<?php
/**
* @package mahara
* @subpackage thirdparty/kses
*/
# 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 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 ***
#
# 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!]
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, $allowed_protocols);
} # function kses
function kses_hook($string)
###############################################################################
# You add any kses hooks here.
###############################################################################
{
return $string;
} # function kses_hook
function kses_version()
###############################################################################
# This function returns kses' version number.
###############################################################################
{
return '0.2.2';
} # function kses_version
function kses_split($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)',
$string);
} # function kses_split
function kses_split2($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,
$allowed_protocols);
} # function kses_split2
function kses_attr($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
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"
{
$thisval = kses_bad_protocol($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 $attrarr;
} # function kses_hair
function kses_check_attr_val($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 $ok;
} # function kses_check_attr_val
function kses_bad_protocol($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('/\xad+/', '', $string); # deals with Opera "feature"
$string2 = $string.'a';
while ($string != $string2)
{
$string2 = $string;
$string = kses_bad_protocol_once($string, $allowed_protocols);
} # while
return $string;
} # function kses_bad_protocol
function kses_no_null($string)
###############################################################################
# This function removes any NULL characters in $string.
###############################################################################
{
$string = preg_replace('/\0+/', '', $string);
$string = preg_replace('/(\\\\0)+/', '', $string);
return $string;
} # function kses_no_null
function kses_stripslashes($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('%\\\\"%', '"', $string);
} # function kses_stripslashes
function kses_array_lc($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 $outarray;
} # function kses_array_lc
function kses_js_entities($string)
###############################################################################
# This function removes the HTML JavaScript entities found in early versions of
# Netscape 4.
###############################################################################
{
return preg_replace('%&\s*\{[^}]*(\}\s*;?|$)%', '', $string);
} # function kses_js_entities
function kses_html_error($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*/', '', $string);
} # function kses_html_error
function kses_bad_protocol_once($string, $allowed_protocols)
###############################################################################
# This function searches for URL protocols at the beginning of $string, while
# handling whitespace and HTML entities.
###############################################################################
{
return preg_replace('/^((&[^;]*;|[\sA-Za-z0-9])*)'.
'(:|&#0*58;|&#[Xx]3[Aa];)\s*/e',
'kses_bad_protocol_once2("\\1", $allowed_protocols)',
$string);
} # function kses_bad_protocol_once
function kses_bad_protocol_once2($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 "$string2:";
else
return '';
} # function kses_bad_protocol_once2
function kses_normalize_entities($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 $string;
} # function kses_normalize_entities
function kses_normalize_entities2($i)
###############################################################################
# This function helps kses_normalize_entities() to only accept 16 bit values
# and nothing more for &#number; entities.
###############################################################################
{
return (($i > 65535) ? "&amp;#$i;" : "&#$i;");
} # function kses_normalize_entities2
function kses_decode_entities($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
?>
<?php
/**
* @package mahara
* @subpackage thirdparty/validateurlsyntax
*/
/*
About validateUrlSyntax():
This function will verify if a http URL is formatted properly, returning
either with true or false.
I used rfc #2396 URI: Generic Syntax as my guide when creating the
regular expression. For all the details see the comments below.
Usage:
validateUrlSyntax( url_to_check[, options])
url_to_check - string - The url to check
options - string - A optional string of options to set which parts of
the url are required, optional, or not allowed. Each option
must be followed by a "+" for required, "?" for optional, or
"-" for not allowed.
s - Scheme. Allows "+?-", defaults to "s?"
H - http:// Allows "+?-", defaults to "H?"
S - https:// (SSL). Allows "+?-", defaults to "S?"
E - mailto: (email). Allows "+?-", defaults to "E-"
F - ftp:// Allows "+?-", defaults to "F-"
Dependant on scheme being enabled
u - User section. Allows "+?-", defaults to "u?"
P - Password in user section. Allows "+?-", defaults to "P?"
Dependant on user section being enabled
a - Address (ip or domain). Allows "+?-", defaults to "a+"
I - Ip address. Allows "+?-", defaults to "I?"
If I+, then domains are disabled
If I-, then domains are required
Dependant on address being enabled
p - Port number. Allows "+?-", defaults to "p?"
f - File path. Allows "+?-", defaults to "f?"
q - Query section. Allows "+?-", defaults to "q?"
r - Fragment (anchor). Allows "+?-", defaults to "r?"
Paste the funtion code, or include_once() this template at the top of the page
you wish to use this function.
Examples:
validateUrlSyntax('http://george@www.cnn.com/#top')
validateUrlSyntax('https://games.yahoo.com:8080/board/chess.htm?move=true')
validateUrlSyntax('http://www.hotmail.com/', 's+u-I-p-q-r-')
validateUrlSyntax('/directory/file.php#top', 's-u-a-p-f+')
if (validateUrlSyntax('http://www.canowhoopass.com/', 'u-'))
{
echo 'URL SYNTAX IS VERIFIED';
} else {
echo 'URL SYNTAX IS ILLEGAL';
}
Last Edited:
December 15th 2004
Changelog:
December 15th 2004
-Added new TLD's - .jobs, .mobi, .post and .travel. They are official, but not yet active.
August 31th 2004
-Fixed bug allowing empty username even when it was required
-Changed and added a few options to add extra schemes
-Added mailto: ftp:// and http:// options
-https option was 'l' now it is 'S' (capital)
-Added password option. Now passwords can be disabled while usernames are ok (for email)
-IP Address option was 'i' now it is 'I' (capital)
-Options are now case sensitive
-Added validateEmailSyntax() and validateFtpSyntax() functions below<br>
August 27th, 2004
-IP group range is more specific. Used to allow 0-299. Now it is 0-255
-Port range more specific. Used to allow 0-69999. Now it is 0-65535<br>
-Fixed bug disallowing 'i-' option.<br>
-Changed license to GPL
July 8th, 2004
-Fixed bug disallowing 'l-' option. Thanks Dr. Cheap
June 15, 2004
-Added options parameter to make it easier for people to plug the function in
without needed to rework the code.
-Split the example application away from the function
June 1, 2004
-Complete rewrite
-Now more modular
-Easier to disable sections
-Easier to port to other languages
-Easier to port to verify email addresses
-Uses only simple regular expressions so it is more portable
-Follows RFC closer for domain names. Some "play" domains may break
-Renamed from 'verifyUrl()' to 'validateUrlSyntax()'
-Removed extra code which added 'http://' and trailing '/' if it was missing
-That code was better suited for a massaging function, not verifying
-Bug fixes:
-Now splits up and forces '/path?query#fragment' order
-No longer requires a path when using a query or fragment
August 29, 2003
-Allowed port numbers above 9999. Now allows up to 69999
Sometime, 2002
-Added new top level domains
-aero, coop, museum, name, info, biz, pro
October 5, 2000
-First Version
Intentional Limitations:
-Does not verify url actually exists. Only validates the syntax
-Strictly follows the RFC standards. Some urls exist in the wild which will
not validate. Including ones with square brackets in the query section '[]'
Known Problems:
-None at this time
Author(s):
Rod Apeldoorn - rod(at)canowhoopass(dot)com
Homepage:
http://www.canowhoopass.com/
Thanks!:
-WEAV -Several members of Weav helped to test - http://weav.bc.ca/
-There were also a number of emails from other developers expressing
thanks and suggestions. It is nice to be appreciated. Thanks!
License: