<?php
-// $Id: spam.php,v 1.133 2007/04/28 03:00:43 henoheno Exp $
+// $Id: spam.php,v 1.148 2007/05/05 04:59:31 henoheno Exp $
// Copyright (C) 2006-2007 PukiWiki Developers Team
// License: GPL v2 or (at your option) any later version
//
}
}
-// Very roughly strings(1)
-function strings($binary = '')
+// Remove redundant values from array()
+function array_unique_recursive($array = array())
{
- // http://www.pcre.org/pcre.txt
- return preg_replace('/[[:^graph:]]+/', "\n", $binary);
+ if (! is_array($array)) return $array;
+
+ $tmp = array();
+ foreach($array as $key => $value){
+ if (is_array($value)) {
+ $array[$key] = array_unique_recursive($value);
+ } else {
+ if (isset($tmp[$value])) {
+ unset($array[$key]);
+ } else {
+ $tmp[$value] = TRUE;
+ }
+ }
+ }
+
+ return $array;
+}
+
+// Roughly strings(1) using PCRE
+// This function is useful to:
+// * Reduce the size of data, from removing unprintable binary data
+// * Detect _bare_strings_ from binary data
+// References:
+// http://www.freebsd.org/cgi/man.cgi?query=strings (Man-page of GNU strings)
+// http://www.pcre.org/pcre.txt
+function strings($binary = '', $min_len = 4, $ignore_space = FALSE)
+{
+ if ($ignore_space) {
+ $binary = preg_replace(
+ array(
+ '/(?:[^[:graph:] \t\n]|[\r])+/s',
+ '/[ \t]{2,}/',
+ '/^[ \t]/m',
+ '/[ \t]$/m',
+ ),
+ array(
+ "\n",
+ ' ',
+ '',
+ ''
+ ),
+ $binary);
+ } else {
+ $binary = preg_replace('/(?:[^[:graph:][:space:]]|[\r])+/s', "\n", $binary);
+ }
+
+ if ($min_len > 1) {
+ $min_len = min(1024, intval($min_len));
+ $binary =
+ implode("\n",
+ preg_grep('/^.{' . $min_len . ',}/S',
+ explode("\n", $binary)
+ )
+ );
+ }
+
+ return $binary;
}
// [OK] <a href="http://nasty.example.com">visit http://nasty.example.com/</a>
// [OK] <a href=\'http://nasty.example.com/\' >discount foobar</a>
// [NG] <a href="http://ng.example.com">visit http://ng.example.com _not_ended_
- $regex = '#<a\b[^>]*\bhref\b[^>]*>.*?</a\b[^>]*(>)#i';
+ $regex = '#<a\b[^>]*\bhref\b[^>]*>.*?</a\b[^>]*(>)#is';
if (isset($method['area_anchor'])) {
$areas = array();
$count = isset($method['asap']) ?
// [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 = '#\[(url|link)\b[^\]]*\].*?\[/\1\b[^\]]*(\])#i';
+ $regex = '#\[(url|link)\b[^\]]*\].*?\[/\1\b[^\]]*(\])#is';
if (isset($method['area_bbcode'])) {
$areas = array();
$count = isset($method['asap']) ?
// 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
-// TODO: link.toolbot.com, urlx.org
function spam_uri_pickup_preprocess($string = '')
{
if (! is_string($string)) return '';
$string = rawurldecode($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(
+ '#http://' .
+ '(' .
+ 'ime\.nu' . '|' . // 2ch.net
+ 'ime\.st' . '|' . // 2ch.net
+ 'link\.toolbot\.com' . '|' .
+ 'urlx\.org' .
+ ')' .
+ '/([a-z0-9.%_-]+\.[a-z0-9.%_-]+)#i', // nasty.example.org
+ 'http://$2/?refer=$1 $0', // Preserve $0 or remove?
+ $string
+ );
+
// Domain exposure (See _preg_replace_callback_domain_exposure())
$string = preg_replace_callback(
array(
}
}
-
// Blocklist metrics: Separate $host, to $blocked and not blocked
function blocklist_distiller(& $hosts, $keys = array('goodhost', 'badhost'), $asap = FALSE)
{
return $blocked;
}
-function is_badhost($hosts = array(), $asap = TRUE, & $remains)
+// Simple example for badhost (not used now)
+function is_badhost($hosts = array(), $asap = TRUE, $bool = TRUE)
{
- $remains = array();
-
- if (! is_array($hosts)) $hosts = array($hosts);
- foreach(array_keys($hosts) as $key) {
- if (! is_string($hosts[$key])) {
- unset($hosts[$key]);
- }
- }
- if (empty($hosts)) return $result;
-
$list = get_blocklist('list');
$blocked = blocklist_distiller($hosts, array_keys($list), $asap);
foreach($list as $key=>$type){
- if (! $type) {
- unset($blocked[$key]); // Ignore goodhost etc
- }
+ if (! $type) unset($blocked[$key]); // Ignore goodhost etc
}
- $remains = $hosts;
- return $blocked;
+ return $bool ? ! empty($blocked) : $blocked;
}
// Simple/fast spam check
function check_uri_spam($target = '', $method = array())
{
- if (! is_array($method) || empty($method)) {
- $method = check_uri_spam_method();
- }
+ // Return value
$progress = array(
+ 'method' => array(
+ // Theme to do => Dummy, optional value, or optional array()
+ //'quantity' => 8,
+ //'uniqhost' => TRUE,
+ //'non_uniqhost'=> 3,
+ //'non_uniquri' => 3,
+ //'badhost' => TRUE,
+ //'area_anchor' => 0,
+ //'area_bbcode' => 0,
+ //'uri_anchor' => 0,
+ //'uri_bbcode' => 0,
+ ),
'sum' => array(
- 'quantity' => 0,
- 'uniqhost' => 0,
- 'non_uniqhost'=> 0,
- 'non_uniquri' => 0,
- 'badhost' => 0,
- 'area_anchor' => 0,
- 'area_bbcode' => 0,
- 'uri_anchor' => 0,
- 'uri_bbcode' => 0,
+ // Theme => Volume found (int)
+ ),
+ 'is_spam' => array(
+ // Flag. If someting defined here,
+ // one or more spam will be included
+ // in this report
+ ),
+ 'blocked' => array(
+ // Hosts blocked
+ //'category' => array(
+ // 'host',
+ //)
+ ),
+ 'hosts' => array(
+ // Hosts not blocked
),
- 'is_spam' => array(),
- 'method' => & $method,
- 'remains' => array(),
- 'error' => array(),
);
+
+ // Aliases
$sum = & $progress['sum'];
$is_spam = & $progress['is_spam'];
- $remains = & $progress['remains'];
- $error = & $progress['error'];
+ $progress['method'] = & $method; // Argument
+ $blocked = & $progress['blocked'];
+ $hosts = & $progress['hosts'];
$asap = isset($method['asap']);
- // Recurse
+ // Init
+ if (! is_array($method) || empty($method)) {
+ $method = check_uri_spam_method();
+ }
+ foreach(array_keys($method) as $key) {
+ if (! isset($sum[$key])) $sum[$key] = 0;
+ }
+
if (is_array($target)) {
foreach($target as $str) {
- // Recurse
- $_progress = check_uri_spam($str, $method);
- $_sum = & $_progress['sum'];
- $_is_spam = & $_progress['is_spam'];
- $_remains = & $_progress['remains'];
- $_error = & $_progress['error'];
+ if (! is_string($str)) continue;
+
+ $_progress = check_uri_spam($str, $method); // Recurse
+
+ // Merge $sum
+ $_sum = & $_progress['sum'];
foreach (array_keys($_sum) as $key) {
- $sum[$key] += $_sum[$key];
- }
- foreach (array_keys($_is_spam) as $key) {
- if (is_array($_is_spam[$key])) {
- // Marge keys (badhost)
- foreach(array_keys($_is_spam[$key]) as $_key) {
- if (! isset($is_spam[$key][$_key])) {
- $is_spam[$key][$_key] = $_is_spam[$key][$_key];
- } else {
- $is_spam[$key][$_key] += $_is_spam[$key][$_key];
- }
- }
+ if (! isset($sum[$key])) {
+ $sum[$key] = & $_sum[$key];
} else {
- $is_spam[$key] = TRUE;
+ $sum[$key] += $_sum[$key];
}
}
- foreach ($_remains as $key=>$value) {
- foreach ($value as $_key=>$_value) {
- if (is_int($_key)) {
- $remains[$key][] = $_value;
- } else {
- $remains[$key][$_key] = $_value;
- }
- }
+
+ // Merge $is_spam
+ $_is_spam = & $_progress['is_spam'];
+ foreach (array_keys($_is_spam) as $key) {
+ $is_spam[$key] = TRUE;
+ if ($asap) break;
}
- if (! empty($_error)) $error += $_error;
if ($asap && $is_spam) break;
+
+ // Merge $blocked
+ $blocked = array_merge_leaves($blocked, $_progress['blocked']);
+
+ // Merge $hosts
+ $hosts = array_merge_leaves($hosts, $_progress['hosts']);
}
+
+ // Recount $sum['badhost']
+ $sum['badhost'] = array_count_leaves($blocked);
+
return $progress;
}
// URI: Pickup
$pickups = uri_pickup_normalize(spam_uri_pickup($target, $method));
- //$remains['uri_pickup'] = & $pickups;
// Return if ...
if (empty($pickups)) return $progress;
if ($asap && $is_spam) return $progress;
// Host: Uniqueness (uniq / non-uniq)
- $hosts = array();
foreach ($pickups as $pickup) $hosts[] = & $pickup['host'];
$hosts = array_unique($hosts);
- //$remains['uniqhost'] = & $hosts;
$sum['uniqhost'] += count($hosts);
if ((! $asap || ! $is_spam) && isset($method['non_uniqhost'])) {
$sum['non_uniqhost'] = $sum['quantity'] - $sum['uniqhost'];
// Return if ...
if ($asap && $is_spam) return $progress;
- // URI: Bad host
+ // URI: Bad host (Separate good/bad hosts from $hosts)
if ((! $asap || ! $is_spam) && isset($method['badhost'])) {
- $__remains = array();
- $badhost = is_badhost($hosts, $asap, $__remains);
- if (! $asap) {
- if ($__remains) {
- $remains['badhost'] = array();
- foreach ($__remains as $value) {
- $remains['badhost'][$value] = TRUE;
- }
- }
- }
- unset($__remains);
- if (! empty($badhost)) {
-
- //var_dump($badhost); // BADHOST detail
- $sum['badhost'] += array_count_leaves($badhost);
- foreach(array_keys($badhost) as $keys) {
- $is_spam['badhost'][$keys] =
- array_count_leaves($badhost[$keys]);
- }
- unset($badhost);
+ // is_badhost()
+ $list = get_blocklist('list');
+ $blocked = blocklist_distiller($hosts, array_keys($list), $asap);
+ foreach($list as $key=>$type){
+ if (! $type) unset($blocked[$key]); // Ignore goodhost etc
}
+ unset($list);
+
+ if (! empty($blocked)) $is_spam['badhost'] = TRUE;
}
return $progress;
return $count;
}
+// Merge two leaves' value
+function array_merge_leaves(& $array1, & $array2, $unique_values = TRUE)
+{
+ // All NUMERIC keys are always renumbered from 0
+ $array = array_merge_recursive($array1, $array2);
+
+ // Redundant values (and keys) are vanished
+ if ($unique_values) $array = array_unique_recursive($array);
+
+ return $array;
+}
+
+
// ---------------------
// Reporting
$method = & $progress['method'];
if (isset($progress['sum'])) {
foreach ($progress['sum'] as $key => $value) {
- if (isset($method[$key])) {
+ if (isset($method[$key]) && $value) {
$tmp[] = $key . '(' . $value . ')';
}
}
return implode(', ', $tmp);
}
+function summarize_detail_badhost($progress = array())
+{
+ if (! isset($progress['is_spam']['badhost'])) return '';
+
+ return var_export($progress['blocked'], TRUE);
+}
+
+function summarize_detail_newtral($progress = array())
+{
+ if (empty($progress['hosts'])) return '';
+
+ return var_export($progress['hosts'], TRUE);
+}
+
+
// ---------------------
// Exit
if (! $asap) {
$summary['METRICS'] = summarize_spam_progress($progress);
}
- if (isset($progress['is_spam']['badhost'])) {
- $badhost = array();
- foreach($progress['is_spam']['badhost'] as $glob=>$number) {
- $badhost[] = $glob . '(' . $number . ')';
- }
- $summary['DETAIL_BADHOST'] = implode(', ', $badhost);
- }
- if (! $asap && $progress['remains']['badhost']) {
- $count = count($progress['remains']['badhost']);
- $summary['DETAIL_NEUTRAL_HOST'] = $count .
- ' (' .
- preg_replace(
- '/[^, a-z0-9.-]/i', '',
- implode(', ', array_keys($progress['remains']['badhost']))
- ) .
- ')';
- }
+
+ $tmp = summarize_detail_badhost($progress);
+ if ($tmp != '') $summary['DETAIL_BADHOST'] = $tmp;
+
+ $tmp = summarize_detail_newtral($progress);
+ if (! $asap && $tmp != '') $summary['DETAIL_NEUTRAL_HOST'] = $tmp;
+
$summary['COMMENT'] = $action;
$summary['PAGE'] = '[blocked] ' . (is_pagename($page) ? $page : '');
$summary['URI'] = get_script_uri() . '?' . rawurlencode($page);