<?php
-// $Id: spam_pickup.php,v 1.59 2007/09/13 13:02:53 henoheno Exp $
+// $Id: spam_pickup.php,v 1.66 2009/01/02 09:30:50 henoheno Exp $
// Copyright (C) 2006-2007 PukiWiki Developers Team
// License: GPL v2 or (at your option) any later version
//
'[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<>"\'\[\]/\#]+)*/+)?' . // 5: Directory path
'([^\s<>"\'\[\]\#?]+)?' . // 6: File?
'(?:\?([^\s<>"\'\[\]\#]+))?' . // 7: Query string
'(?:\#([a-z0-9._~%!$&\'()*+,;=:@-]*))?' . // 8: Fragment
$string, $array, PREG_SET_ORDER | PREG_OFFSET_CAPTURE
);
- // Format the $array
+ // Reformat the $array
static $parts = array(
1 => 'scheme', 2 => 'userinfo', 3 => 'host', 4 => 'port',
5 => 'path', 6 => 'file', 7 => 'query', 8 => 'fragment'
);
- $default = array('');
+ $default = array(0 => '', 1 => -1);
foreach(array_keys($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
+ $_uri[$part] = $_uri[$part][0]; // Remove offsets
}
}
// Normalize an array of URI arrays
// NOTE: Give me the uri_pickup() results
-function uri_pickup_normalize(& $pickups, $destructive = TRUE)
+function uri_pickup_normalize(& $pickups, $destructive = TRUE, $pathfile = FALSE)
{
if (! is_array($pickups)) return $pickups;
}
}
+ if ($pathfile) {
+ return uri_pickup_normalize_pathfile($pickups);
+ } else {
+ return $pickups;
+ }
+}
+
+// Normalize: 'path' + 'file' = 'path' (Similar structure using PHP's "parse_url()" function)
+// NOTE: In some case, 'file' DOES NOT mean _filename_.
+// [EXAMPLE] http://example.com/path/to/directory-accidentally-not-ended-with-slash
+function uri_pickup_normalize_pathfile(& $pickups)
+{
+ if (! is_array($pickups)) return $pickups;
+
+ foreach (array_keys($pickups) as $key) {
+ $_key = & $pickups[$key];
+ if (isset($_key['path'], $_key['file'])) {
+ $_key['path'] = $_key['path'] . $_key['file'];
+ unset($_key['file']);
+ }
+ }
+
return $pickups;
}
// www.foo.bar => foo.bar
// www.10.20 => www.10.20 (Invalid hostname)
// NOTE:
-// 'www' is mostly used as traditional hostname of WWW server.
-// 'www.foo.bar' may be identical with 'foo.bar'.
+// 'www' is basically traditional hostname for WWW server.
+// In these case, 'www.foo.bar' MAY be identical with 'foo.bar'.
function host_normalize($host = '')
{
if (! is_string($host)) return '';
$host = strtolower($host);
+
$matches = array();
if (preg_match('/^www\.(.+\.[a-z]+)$/', $host, $matches)) {
return $matches[1];
// ---------------------
// Spam-uri pickup
-// Preprocess: Removing uninterest part for URI detection
+// Preprocess: Removing/Modifying uninterest part for URI detection
function spam_uri_removing_hocus_pocus($binary = '', $method = array())
{
+ $from = $to = array();
+
+ // Remove sequential spaces and too short lines
$length = 4 ; // 'http'(1) and '://'(2) and 'fqdn'(1)
if (is_array($method)) {
// '<a'(2) or 'href='(5) or '>'(1) or '</a>'(4)
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/Replace quoted-spaces within tags
+ $from[] = '#(<\w+ [^<>]*?\w ?= ?")([^"<>]*? [^"<>]*)("[^<>]*?>)#ie';
+ $to[] = "'$1' . str_replace(' ' , '%20' , trim('$2')) . '$3'";
+
// Remove words (has no '<>[]:') between spaces
- $binary = preg_replace('/[ \t][\w.,()\ \t]+[ \t]/', ' ', $binary);
+ $from[] = '/[ \t][\w.,()\ \t]+[ \t]/';
+ $to[] = ' ';
- return $binary;
+ return preg_replace($from, $to, $binary);
}
// Preprocess: Domain exposure callback (See spam_uri_pickup_preprocess())
);
$string = spam_uri_removing_hocus_pocus($string, $method);
- //var_dump(htmlspecialchars($string));
// Domain exposure (simple)
// http://victim.example.org/nasty.example.org/path#frag
$string = preg_replace(
'#h?ttp://' .
'(' .
+ 'a9\.com/' . '|' .
+ 'aboutus\.org/' . '|' .
+ 'alexa\.com/data/details\?url=' . '|' .
'ime\.(?:nu|st)/' . '|' . // 2ch.net
'link\.toolbot\.com/' . '|' .
'urlx\.org/' . '|' .
// ...
')' .
'/' .
- '([a-z0-9?=&.%_/\'\\\+-]+)' . // 3:path/?query=foo+bar+
+ //TODO: Specify URL-enable characters
+ '([a-z0-9?=&.,%_/\'\\\+-]+)' . // 3:path/?query=foo+bar+
'(?:\b|%20)site:([a-z0-9.%_-]+\.[a-z0-9.%_-]+)' . // 4:site:nasty.example.com
'()' . // 5:Preserve or remove?
'#i',
}
// Remove 'offset's for area_measure()
- foreach(array_keys($array) as $key)
+ foreach(array_keys($array) as $key) {
unset($array[$key]['area']['offset']);
+ }
return $array;
}