"\'\[\]/\#?@]*)' . // 2: Userinfo (Username)
'@)?' .
'(' .
// 3: Host
'\[[0-9a-f:.]+\]' . '|' . // IPv6([colon-hex and dot]): RFC2732
'(?:[0-9]{1,3}\.){3}[0-9]{1,3}' . '|' . // IPv4(dot-decimal): 001.22.3.44
'[a-z0-9_-][a-z0-9_.-]+[a-z0-9_-]' . // hostname(FQDN) : foo.example.org
')' .
'(?::([0-9]*))?' . // 4: Port
'((?:/+[^\s<>"\'\[\]/\#]+)*/+)?' . // 5: Directory path or path-info
'([^\s<>"\'\[\]\#?]+)?' . // 6: File?
'(?:\?([^\s<>"\'\[\]\#]+))?' . // 7: Query string
'(?:\#([a-z0-9._~%!$&\'()*+,;=:@-]*))?' . // 8: Fragment
'#i',
$string, $array, PREG_SET_ORDER | PREG_OFFSET_CAPTURE
);
// Format the $array
static $parts = array(
1 => 'scheme', 2 => 'userinfo', 3 => 'host', 4 => 'port',
5 => 'path', 6 => 'file', 7 => 'query', 8 => 'fragment'
);
$default = array('');
foreach(array_keys($array) as $uri) {
$_uri = & $array[$uri];
array_rename_keys($_uri, $parts, TRUE, $default);
$offset = $_uri['scheme'][1]; // Scheme's offset = URI's offset
foreach(array_keys($_uri) as $part) {
$_uri[$part] = & $_uri[$part][0]; // Remove offsets
}
}
foreach(array_keys($array) as $uri) {
$_uri = & $array[$uri];
if ($_uri['scheme'] === '') {
unset($array[$uri]); // Considererd harmless
continue;
}
unset($_uri[0]); // Matched string itself
$_uri['area']['offset'] = $offset; // Area offset for area_measure()
}
return $array;
}
// Pickupped URI array => An URI (See uri_pickup())
// USAGE:
// $pickups = uri_pickup('a string include some URIs');
// $uris = array();
// foreach (array_keys($pickups) as $key) {
// $uris[$key] = uri_pickup_implode($pickups[$key]);
// }
function uri_pickup_implode($uri = array())
{
if (empty($uri) || ! is_array($uri)) return NULL;
$tmp = array();
if (isset($uri['scheme']) && $uri['scheme'] !== '') {
$tmp[] = & $uri['scheme'];
$tmp[] = '://';
}
if (isset($uri['userinfo']) && $uri['userinfo'] !== '') {
$tmp[] = & $uri['userinfo'];
$tmp[] = '@';
}
if (isset($uri['host']) && $uri['host'] !== '') {
$tmp[] = & $uri['host'];
}
if (isset($uri['port']) && $uri['port'] !== '') {
$tmp[] = ':';
$tmp[] = & $uri['port'];
}
if (isset($uri['path']) && $uri['path'] !== '') {
$tmp[] = & $uri['path'];
}
if (isset($uri['file']) && $uri['file'] !== '') {
$tmp[] = & $uri['file'];
}
if (isset($uri['query']) && $uri['query'] !== '') {
$tmp[] = '?';
$tmp[] = & $uri['query'];
}
if (isset($uri['fragment']) && $uri['fragment'] !== '') {
$tmp[] = '#';
$tmp[] = & $uri['fragment'];
}
return implode('', $tmp);
}
// ---------------------
// URI normalization
// Normalize an array of URI arrays
// NOTE: Give me the uri_pickup() results
function uri_pickup_normalize(& $pickups, $destructive = TRUE)
{
if (! is_array($pickups)) return $pickups;
if ($destructive) {
foreach (array_keys($pickups) as $key) {
$_key = & $pickups[$key];
$_key['scheme'] = isset($_key['scheme']) ? scheme_normalize($_key['scheme']) : '';
$_key['host'] = isset($_key['host']) ? host_normalize($_key['host']) : '';
$_key['port'] = isset($_key['port']) ? port_normalize($_key['port'], $_key['scheme'], FALSE) : '';
$_key['path'] = isset($_key['path']) ? strtolower(path_normalize($_key['path'])) : '';
$_key['file'] = isset($_key['file']) ? file_normalize($_key['file']) : '';
$_key['query'] = isset($_key['query']) ? query_normalize($_key['query']) : '';
$_key['fragment'] = isset($_key['fragment']) ? strtolower($_key['fragment']) : '';
}
} else {
foreach (array_keys($pickups) as $key) {
$_key = & $pickups[$key];
$_key['scheme'] = isset($_key['scheme']) ? scheme_normalize($_key['scheme']) : '';
$_key['host'] = isset($_key['host']) ? strtolower($_key['host']) : '';
$_key['port'] = isset($_key['port']) ? port_normalize($_key['port'], $_key['scheme'], FALSE) : '';
$_key['path'] = isset($_key['path']) ? path_normalize($_key['path']) : '';
}
}
return $pickups;
}
// Scheme normalization: Renaming the schemes
// snntp://example.org => nntps://example.org
// NOTE: Keep the static lists simple. See also port_normalize().
function scheme_normalize($scheme = '', $abbrevs_harmfull = TRUE)
{
// Abbreviations they have no intention of link
static $abbrevs = array(
'ttp' => 'http',
'ttps' => 'https',
);
// Aliases => normalized ones
static $aliases = array(
'pop' => 'pop3',
'news' => 'nntp',
'imap4' => 'imap',
'snntp' => 'nntps',
'snews' => 'nntps',
'spop3' => 'pop3s',
'pops' => 'pop3s',
);
if (! is_string($scheme)) return '';
$scheme = strtolower($scheme);
if (isset($abbrevs[$scheme])) {
$scheme = $abbrevs_harmfull ? $abbrevs[$scheme] : '';
}
if (isset($aliases[$scheme])) {
$scheme = $aliases[$scheme];
}
return $scheme;
}
// Hostname normlization (Destructive)
// www.foo => www.foo ('foo' seems TLD)
// www.foo.bar => foo.bar
// www.10.20 => www.10.20 (Invalid hostname)
// NOTE:
// 'www' is mostly used as traditional hostname of WWW server.
// 'www.foo.bar' may be identical with 'foo.bar'.
function host_normalize($host = '')
{
if (! is_string($host)) return '';
$host = strtolower($host);
$matches = array();
if (preg_match('/^www\.(.+\.[a-z]+)$/', $host, $matches)) {
return $matches[1];
} else {
return $host;
}
}
// Port normalization: Suppress the (redundant) default port
// HTTP://example.org:80/ => http://example.org/
// HTTP://example.org:8080/ => http://example.org:8080/
// HTTPS://example.org:443/ => https://example.org/
function port_normalize($port, $scheme, $scheme_normalize = FALSE)
{
// Schemes that users _maybe_ want to add protocol-handlers
// to their web browsers. (and attackers _maybe_ want to use ...)
// Reference: http://www.iana.org/assignments/port-numbers
static $array = array(
// scheme => default port
'ftp' => 21,
'ssh' => 22,
'telnet' => 23,
'smtp' => 25,
'tftp' => 69,
'gopher' => 70,
'finger' => 79,
'http' => 80,
'pop3' => 110,
'sftp' => 115,
'nntp' => 119,
'imap' => 143,
'irc' => 194,
'wais' => 210,
'https' => 443,
'nntps' => 563,
'rsync' => 873,
'ftps' => 990,
'telnets' => 992,
'imaps' => 993,
'ircs' => 994,
'pop3s' => 995,
'mysql' => 3306,
);
// intval() converts '0-1' to '0', so preg_match() rejects these invalid ones
if (! is_numeric($port) || $port < 0 || preg_match('/[^0-9]/i', $port))
return '';
$port = intval($port);
if ($scheme_normalize) $scheme = scheme_normalize($scheme);
if (isset($array[$scheme]) && $port == $array[$scheme])
$port = ''; // Ignore the defaults
return $port;
}
// Path normalization
// http://example.org => http://example.org/
// http://example.org#hoge => http://example.org/#hoge
// http://example.org/path/a/b/./c////./d => http://example.org/path/a/b/c/d
// http://example.org/path/../../a/../back => http://example.org/back
function path_normalize($path = '', $divider = '/', $add_root = TRUE)
{
if (! is_string($divider)) return is_string($path) ? $path : '';
if ($add_root) {
$first_div = & $divider;
} else {
$first_div = '';
}
if (! is_string($path) || $path == '') return $first_div;
if (strpos($path, $divider, strlen($path) - strlen($divider)) === FALSE) {
$last_div = '';
} else {
$last_div = & $divider;
}
$array = explode($divider, $path);
// Remove paddings ('//' and '/./')
foreach(array_keys($array) as $key) {
if ($array[$key] == '' || $array[$key] == '.') {
unset($array[$key]);
}
}
// Remove back-tracks ('/../')
$tmp = array();
foreach($array as $value) {
if ($value == '..') {
array_pop($tmp);
} else {
array_push($tmp, $value);
}
}
$array = & $tmp;
if (empty($array)) {
return $first_div;
} else {
return $first_div . implode($divider, $array) . $last_div;
}
}
// DirectoryIndex normalize (Destructive and rough)
// TODO: sample.en.ja.html.gz => sample.html
function file_normalize($file = 'index.html.en')
{
static $simple_defaults = array(
'default.htm' => TRUE,
'default.html' => TRUE,
'default.asp' => TRUE,
'default.aspx' => TRUE,
'index' => TRUE, // Some system can omit the suffix
);
static $content_suffix = array(
// index.xxx, sample.xxx
'htm' => TRUE,
'html' => TRUE,
'shtml' => TRUE,
'jsp' => TRUE,
'php' => TRUE,
'php3' => TRUE,
'php4' => TRUE,
'pl' => TRUE,
'py' => TRUE,
'rb' => TRUE,
'cgi' => TRUE,
'xml' => TRUE,
);
static $language_suffix = array(
// Reference: Apache 2.0.59 'AddLanguage' default
'ca' => TRUE,
'cs' => TRUE, // cs
'cz' => TRUE, // cs
'de' => TRUE,
'dk' => TRUE, // da
'el' => TRUE,
'en' => TRUE,
'eo' => TRUE,
'es' => TRUE,
'et' => TRUE,
'fr' => TRUE,
'he' => TRUE,
'hr' => TRUE,
'it' => TRUE,
'ja' => TRUE,
'ko' => TRUE,
'ltz' => TRUE,
'nl' => TRUE,
'nn' => TRUE,
'no' => TRUE,
'po' => TRUE,
'pt' => TRUE,
'pt-br' => TRUE,
'ru' => TRUE,
'sv' => TRUE,
'zh-cn' => TRUE,
'zh-tw' => TRUE,
// Reference: Apache 2.0.59 default 'index.html' variants
'ee' => TRUE,
'lb' => TRUE,
'var' => TRUE,
);
static $charset_suffix = array(
// Reference: Apache 2.0.59 'AddCharset' default
'iso8859-1' => TRUE, // ISO-8859-1
'latin1' => TRUE, // ISO-8859-1
'iso8859-2' => TRUE, // ISO-8859-2
'latin2' => TRUE, // ISO-8859-2
'cen' => TRUE, // ISO-8859-2
'iso8859-3' => TRUE, // ISO-8859-3
'latin3' => TRUE, // ISO-8859-3
'iso8859-4' => TRUE, // ISO-8859-4
'latin4' => TRUE, // ISO-8859-4
'iso8859-5' => TRUE, // ISO-8859-5
'latin5' => TRUE, // ISO-8859-5
'cyr' => TRUE, // ISO-8859-5
'iso-ru' => TRUE, // ISO-8859-5
'iso8859-6' => TRUE, // ISO-8859-6
'latin6' => TRUE, // ISO-8859-6
'arb' => TRUE, // ISO-8859-6
'iso8859-7' => TRUE, // ISO-8859-7
'latin7' => TRUE, // ISO-8859-7
'grk' => TRUE, // ISO-8859-7
'iso8859-8' => TRUE, // ISO-8859-8
'latin8' => TRUE, // ISO-8859-8
'heb' => TRUE, // ISO-8859-8
'iso8859-9' => TRUE, // ISO-8859-9
'latin9' => TRUE, // ISO-8859-9
'trk' => TRUE, // ISO-8859-9
'iso2022-jp'=> TRUE, // ISO-2022-JP
'jis' => TRUE, // ISO-2022-JP
'iso2022-kr'=> TRUE, // ISO-2022-KR
'kis' => TRUE, // ISO-2022-KR
'iso2022-cn'=> TRUE, // ISO-2022-CN
'cis' => TRUE, // ISO-2022-CN
'big5' => TRUE,
'cp-1251' => TRUE, // ru, WINDOWS-1251
'win-1251' => TRUE, // ru, WINDOWS-1251
'cp866' => TRUE, // ru
'koi8-r' => TRUE, // ru, KOI8-r
'koi8-ru' => TRUE, // ru, KOI8-r
'koi8-uk' => TRUE, // ru, KOI8-ru
'ua' => TRUE, // ru, KOI8-ru
'ucs2' => TRUE, // ru, ISO-10646-UCS-2
'ucs4' => TRUE, // ru, ISO-10646-UCS-4
'utf8' => TRUE,
// Reference: Apache 2.0.59 default 'index.html' variants
'euc-kr' => TRUE,
'gb2312' => TRUE,
);
// May uncompress by web browsers on the fly
// Must be at the last of the filename
// Reference: Apache 2.0.59 'AddEncoding'
static $encoding_suffix = array(
'z' => TRUE,
'gz' => TRUE,
);
if (! is_string($file)) return '';
$_file = strtolower($file);
if (isset($simple_defaults[$_file])) return '';
// Roughly removing language/character-set/encoding suffixes
// References:
// * Apache 2 document about 'Content-negotiaton', 'mod_mime' and 'mod_negotiation'
// http://httpd.apache.org/docs/2.0/content-negotiation.html
// http://httpd.apache.org/docs/2.0/mod/mod_mime.html
// http://httpd.apache.org/docs/2.0/mod/mod_negotiation.html
// * http://www.iana.org/assignments/character-sets
// * RFC3066: Tags for the Identification of Languages
// http://www.ietf.org/rfc/rfc3066.txt
// * ISO 639: codes of 'language names'
$suffixes = explode('.', $_file);
$body = array_shift($suffixes);
if ($suffixes) {
// Remove the last .gz/.z
$last_key = end(array_keys($suffixes));
if (isset($encoding_suffix[$suffixes[$last_key]])) {
unset($suffixes[$last_key]);
}
}
// Cut language and charset suffixes
foreach($suffixes as $key => $value){
if (isset($language_suffix[$value]) || isset($charset_suffix[$value])) {
unset($suffixes[$key]);
}
}
if (empty($suffixes)) return $body;
// Index.xxx
$count = count($suffixes);
reset($suffixes);
$current = current($suffixes);
if ($body == 'index' && $count == 1 && isset($content_suffix[$current])) return '';
return $file;
}
// Sort query-strings if possible (Destructive and rough)
// [OK] &&&&f=d&b&d&c&a=0dd => a=0dd&b&c&d&f=d
// [OK] nothing==&eg=dummy&eg=padding&eg=foobar => eg=foobar
function query_normalize($string = '', $equal = TRUE, $equal_cutempty = TRUE, $stortolower = TRUE)
{
if (! is_string($string)) return '';
if ($stortolower) $string = strtolower($string);
$array = explode('&', $string);
// Remove '&' paddings
foreach(array_keys($array) as $key) {
if ($array[$key] == '') {
unset($array[$key]);
}
}
// Consider '='-sepalated input and paddings
if ($equal) {
$equals = $not_equals = array();
foreach ($array as $part) {
if (strpos($part, '=') === FALSE) {
$not_equals[] = $part;
} else {
list($key, $value) = explode('=', $part, 2);
$value = ltrim($value, '=');
if (! $equal_cutempty || $value != '') {
$equals[$key] = $value;
}
}
}
$array = & $not_equals;
foreach ($equals as $key => $value) {
$array[] = $key . '=' . $value;
}
unset($equals);
}
natsort($array);
return implode('&', $array);
}
// ---------------------
// Area pickup
// Pickup all of markup areas
function area_pickup($string = '', $method = array())
{
$area = array();
if (empty($method)) return $area;
// Anchor tag pair by preg_match and preg_match_all()
// [OK]
// [OK] Good site!
// [OK] test
// [OK] visit http://nasty.example.com/
// [OK] discount foobar
// [NG] visit http://ng.example.com _not_ended_
$regex = '#]*\bhref\b[^>]*>.*?]*(>)#is';
if (isset($method['area_anchor'])) {
$areas = array();
$count = isset($method['asap']) ?
preg_match($regex, $string) :
preg_match_all($regex, $string, $areas);
if (! empty($count)) $area['area_anchor'] = $count;
}
if (isset($method['uri_anchor'])) {
$areas = array();
preg_match_all($regex, $string, $areas, PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
foreach(array_keys($areas) as $_area) {
$areas[$_area] = array(
$areas[$_area][0][1], // Area start ()
$areas[$_area][1][1], // Area end ()
);
}
if (! empty($areas)) $area['uri_anchor'] = $areas;
}
// phpBB's "BBCode" pair by preg_match and preg_match_all()
// [OK] [url][/url]
// [OK] [url]http://nasty.example.com/[/url]
// [OK] [link]http://nasty.example.com/[/link]
// [OK] [url=http://nasty.example.com]visit http://nasty.example.com/[/url]
// [OK] [link http://nasty.example.com/]buy something[/link]
$regex = '#\[(url|link)\b[^\]]*\].*?\[/\1\b[^\]]*(\])#is';
if (isset($method['area_bbcode'])) {
$areas = array();
$count = isset($method['asap']) ?
preg_match($regex, $string) :
preg_match_all($regex, $string, $areas, PREG_SET_ORDER);
if (! empty($count)) $area['area_bbcode'] = $count;
}
if (isset($method['uri_bbcode'])) {
$areas = array();
preg_match_all($regex, $string, $areas, PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
foreach(array_keys($areas) as $_area) {
$areas[$_area] = array(
$areas[$_area][0][1], // Area start ([url])
$areas[$_area][2][1], // Area end ([/url])
);
}
if (! empty($areas)) $area['uri_bbcode'] = $areas;
}
// Various Wiki syntax
// [text_or_uri>text_or_uri]
// [text_or_uri:text_or_uri]
// [text_or_uri|text_or_uri]
// [text_or_uri->text_or_uri]
// [text_or_uri text_or_uri] // MediaWiki
// MediaWiki: [http://nasty.example.com/ visit http://nasty.example.com/]
return $area;
}
// If in doubt, it's a little doubtful
// if (Area => inside <= Area) $brief += -1
function area_measure($areas, & $array, $belief = -1, $a_key = 'area', $o_key = 'offset')
{
if (! is_array($areas) || ! is_array($array)) return;
$areas_keys = array_keys($areas);
foreach(array_keys($array) as $u_index) {
$offset = isset($array[$u_index][$o_key]) ?
intval($array[$u_index][$o_key]) : 0;
foreach($areas_keys as $a_index) {
if (isset($array[$u_index][$a_key])) {
$offset_s = intval($areas[$a_index][0]);
$offset_e = intval($areas[$a_index][1]);
// [Area => inside <= Area]
if ($offset_s < $offset && $offset < $offset_e) {
$array[$u_index][$a_key] += $belief;
}
}
}
}
}
// ---------------------
// Spam-uri pickup
// Preprocess: Removing uninterest part for URI detection
function spam_uri_removing_hocus_pocus($binary = '', $method = array())
{
$length = 4 ; // 'http'(1) and '://'(2) and 'fqdn'(1)
if (is_array($method)) {
// ''(1) or ''(4)
// '[uri'(4) or ']'(1) or '[/uri]'(6)
if (isset($method['area_anchor']) || isset($method['uri_anchor']) ||
isset($method['area_bbcode']) || isset($method['uri_bbcode']))
$length = 1; // Seems not effective
}
// Removing sequential spaces and too short lines
$binary = strings($binary, $length, TRUE, FALSE); // Multibyte NOT needed
// Remove words (has no '<>[]:') between spaces
$binary = preg_replace('/[ \t][\w.,()\ \t]+[ \t]/', ' ', $binary);
return $binary;
}
// Preprocess: Domain exposure callback (See spam_uri_pickup_preprocess())
// http://victim.example.org/?foo+site:nasty.example.com+bar
// => http://nasty.example.com/?refer=victim.example.org
// NOTE: 'refer=' is not so good for (at this time).
// Consider about using IP address of the victim, try to avoid that.
function _preg_replace_callback_domain_exposure($matches = array())
{
$result = '';
// Preserve the victim URI as a complicity or ...
if (isset($matches[5])) {
$result =
$matches[1] . '://' . // scheme
$matches[2] . '/' . // victim.example.org
$matches[3]; // The rest of all (before victim)
}
// Flipped URI
if (isset($matches[4])) {
$result =
$matches[1] . '://' . // scheme
$matches[4] . // nasty.example.com
'/?refer=' . strtolower($matches[2]) . // victim.example.org
' ' . $result;
}
return $result;
}
// Preprocess: rawurldecode() and adding space(s) and something
// to detect/count some URIs _if possible_
// NOTE: It's maybe danger to var_dump(result). [e.g. 'javascript:']
// [OK] http://victim.example.org/?site:nasty.example.org
// [OK] http://victim.example.org/nasty.example.org
// [OK] http://victim.example.org/go?http%3A%2F%2Fnasty.example.org
// [OK] http://victim.example.org/http://nasty.example.org
function spam_uri_pickup_preprocess($string = '', $method = array())
{
if (! is_string($string)) return '';
$string = spam_uri_removing_hocus_pocus(rawurldecode($string), $method);
//var_dump(htmlspecialchars($string));
// Domain exposure (simple)
// http://victim.example.org/nasty.example.org/path#frag
// => http://nasty.example.org/?refer=victim.example.org and original
$string = preg_replace(
'#h?ttp://' .
'(' .
'ime\.(?:nu|st)/' . '|' . // 2ch.net
'link\.toolbot\.com/' . '|' .
'urlx\.org/' . '|' .
'big5.51job.com/gate/big5/' . '|' .
'big5.china.com/gate/big5/' . '|' .
'big5.xinhuanet.com/gate/big5/' . '|' .
'bhomiyo.com/en.xliterate/' . '|' .
'google.com/translate_c\?u=(?:http://)?' . '|' .
'web.archive.org/web/2[^/]*/(?:http://)?' .
')' .
'([a-z0-9.%_-]+\.[a-z0-9.%_-]+)' . // nasty.example.org
'#i',
'http://$2/?refer=$1 $0', // Preserve $0 or remove?
$string
);
// Domain exposure (site:) See _preg_replace_callback_domain_exposure()
$string = preg_replace_callback(
array(
'#(h?ttp)://' . // 1:Scheme
// 2:Host
'(' .
'(?:[a-z0-9_.-]+\.)?[a-z0-9_-]+\.[a-z0-9_-]+' .
// Something Google: http://www.google.com/supported_domains
// AltaVista: http://es.altavista.com/web/results?q=site%3Anasty.example.org+foobar
// Live Search: search.live.com
// MySpace: http://sads.myspace.com/Modules/Search/Pages/Search.aspx?_snip_&searchString=site:nasty.example.org
// (also searchresults.myspace.com)
// alltheweb.com
// search.bbc.co.uk
// search.orange.co.uk
// ...
')' .
'/' .
'([a-z0-9?=&.%_/\'\\\+-]+)' . // 3:path/?query=foo+bar+
'\bsite:([a-z0-9.%_-]+\.[a-z0-9.%_-]+)' . // 4:site:nasty.example.com
'()' . // 5:Preserve or remove?
'#i',
),
'_preg_replace_callback_domain_exposure',
$string
);
// URI exposure (uriuri => uri uri)
$string = preg_replace(
array(
'#(?