<?php
-// $Id: spam.php,v 1.131 2007/04/22 08:04:19 henoheno Exp $
+// $Id: spam.php,v 1.176 2007/06/14 14:57:37 henoheno Exp $
// Copyright (C) 2006-2007 PukiWiki Developers Team
// License: GPL v2 or (at your option) any later version
//
}
}
+// ----
+
+// Very roughly, shrink the lines of var_export()
+// NOTE: If the same data exists, it must be corrupted.
+function var_export_shrink($expression, $return = FALSE, $ignore_numeric_keys = FALSE)
+{
+ $result =preg_replace(
+ // Remove a newline and spaces
+ '# => \n *array \(#', ' => array (',
+ var_export($expression, TRUE)
+ );
+
+ if ($ignore_numeric_keys) {
+ $result =preg_replace(
+ // Remove numeric keys
+ '#^( *)[0-9]+ => #m', '$1',
+ $result
+ );
+ }
+
+ if ($return) {
+ return $result;
+ } else {
+ echo $result;
+ return NULL;
+ }
+}
+
+// Remove redundant values from array()
+function array_unique_recursive($array = array())
+{
+ 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;
+}
+
+// Renumber all numeric keys from 0
+function array_renumber_numeric_keys(& $array)
+{
+ if (! is_array($array)) return $array;
+
+ $count = -1;
+ $tmp = array();
+ foreach($array as $key => $value){
+ if (is_array($value)) array_renumber_numeric_keys($array[$key]); // Recurse
+ if (is_numeric($key)) $tmp[$key] = ++$count;
+ }
+ array_rename_keys($array, $tmp);
+
+ 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 {
+ // Remove "\0" etc. Preserve readable spaces if possible.
+ $binary = preg_replace('/(?:[^[:graph:][:space:]]|[\r])+/s', "\n", $binary);
+ }
+
+ if ($min_len > 1) {
+ $min_len = min(1024, intval($min_len));
+ $regex = '/^.{' . $min_len . ',}/S';
+ if (is_array($binary)) {
+ foreach(array_keys($binary) as $key) {
+ $binary[$key] = implode("\n", preg_grep($regex, explode("\n", $binary[$key])));
+ }
+ } else {
+ $binary = implode("\n", preg_grep($regex, explode("\n", $binary)));
+ }
+ }
+
+ return $binary;
+}
+
+// Reverse $string with specified delimiter
+function delimiter_reverse($string = 'foo.bar.example.com', $from_delim = '.', $to_delim = '.')
+{
+ if (! is_string($string) || ! is_string($from_delim) || ! is_string($to_delim))
+ return $string;
+
+ // com.example.bar.foo
+ return implode($to_delim, array_reverse(explode($from_delim, $string)));
+}
+
+
// ---------------------
// URI pickup
preg_match_all(
// scheme://userinfo@host:port/path/or/pathinfo/maybefile.and?query=string#fragment
// Refer RFC3986 (Regex below is not strict)
- '#(\b[a-z][a-z0-9.+-]{1,8}):/+' . // 1: Scheme
+ '#(\b[a-z][a-z0-9.+-]{1,8}):[/\\\]+' . // 1: Scheme
'(?:' .
'([^\s<>"\'\[\]/\#?@]*)' . // 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.-]+' . // hostname(FQDN) : foo.example.org
+ '[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
// [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(
+ '#h?ttp://' .
+ '(' .
+ '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 (gate-big5)
+ // http://victim.example.org/gate/big5/nasty.example.org/path
+ // => http://nasty.example.org/?refer=victim.example.org and original
+ $string = preg_replace(
+ '#h?ttp://' .
+ '(' .
+ 'big5.51job.com' . '|' .
+ 'big5.china.com' . '|' .
+ 'big5.xinhuanet.com' . '|' .
+ ')' .
+ '/gate/big5' .
+ '/([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())
$string = preg_replace_callback(
array(
function get_blocklist($list = '')
{
- static $regexs;
+ static $regexes;
- if (! isset($regexs)) {
- $regexs = array();
+ if ($list === NULL) {
+ $regexes = NULL; // Unset
+ return array();
+ }
+
+ if (! isset($regexes)) {
+ $regexes = array();
if (file_exists(SPAM_INI_FILE)) {
$blocklist = array();
include(SPAM_INI_FILE);
// 'IANA-examples' => '#^(?:.*\.)?example\.(?:com|net|org)$#',
// );
if (isset($blocklist['list'])) {
- $regexs['list'] = & $blocklist['list'];
+ $regexes['list'] = & $blocklist['list'];
} else {
// Default
$blocklist['list'] = array(
if (! isset($blocklist[$_list])) continue;
foreach ($blocklist[$_list] as $key => $value) {
if (is_array($value)) {
- $regexs[$_list][$key] = array();
+ $regexes[$_list][$key] = array();
foreach($value as $_key => $_value) {
- get_blocklist_add($regexs[$_list][$key], $_key, $_value);
+ get_blocklist_add($regexes[$_list][$key], $_key, $_value);
}
} else {
- get_blocklist_add($regexs[$_list], $key, $value);
+ get_blocklist_add($regexes[$_list], $key, $value);
}
}
unset($blocklist[$_list]);
}
}
- if ($list == '') {
- return $regexs; // ALL
- } else if (isset($regexs[$list])) {
- return $regexs[$list];
- } else {
+ if ($list === '') {
+ return $regexes; // ALL
+ } else if (isset($regexes[$list])) {
+ return $regexes[$list];
+ } else {
return array();
}
}
}
}
-function is_badhost($hosts = array(), $asap = TRUE, & $remains)
+// Blocklist metrics: Separate $host, to $blocked and not blocked
+function blocklist_distiller(& $hosts, $keys = array('goodhost', 'badhost'), $asap = FALSE)
{
- $result = array();
if (! is_array($hosts)) $hosts = array($hosts);
- foreach(array_keys($hosts) as $key) {
- if (! is_string($hosts[$key])) {
- unset($hosts[$key]);
- }
- }
- if (empty($hosts)) return $result;
-
- foreach(get_blocklist('list') as $key=>$value){
- if ($value) {
- foreach (get_blocklist($key) as $label => $regex) {
- if (is_array($regex)) {
- $result[$label] = array();
- foreach($regex as $_label => $_regex) {
- if (is_badhost_avail($_label, $_regex, $hosts, $result[$label]) && $asap) {
- break;
- }
- }
- if (empty($result[$label])) unset($result[$label]);
- } else {
- if (is_badhost_avail($label, $regex, $hosts, $result) && $asap) {
- break;
+ if (! is_array($keys)) $keys = array($keys);
+
+ $list = get_blocklist('list');
+ $blocked = array();
+
+ foreach($keys as $key){
+ foreach (get_blocklist($key) as $label => $regex) {
+ if (is_array($regex)) {
+ foreach($regex as $_label => $_regex) {
+ $group = preg_grep($_regex, $hosts);
+ if ($group) {
+ $hosts = array_diff($hosts, $group);
+ $blocked[$key][$label][$_label] = $group;
+ if ($asap && $list[$key]) break;
}
}
+ } else {
+ $group = preg_grep($regex, $hosts);
+ if ($group) {
+ $hosts = array_diff($hosts, $group);
+ $blocked[$key][$label] = $group;
+ if ($asap && $list[$key]) break;
+ }
}
- } else {
- foreach (get_blocklist($key) as $regex) {
- $hosts = preg_grep_invert($regex, $hosts);
- }
- if (empty($hosts)) return $result;
}
}
- $remains = $hosts;
- return $result;
-}
-
-// Subroutine for is_badhost()
-function is_badhost_avail($label = '*.example.org', $regex = '/^.*\.example\.org$/', & $hosts, & $result)
-{
- $group = preg_grep($regex, $hosts);
- if ($group) {
-
- // DEBUG var_dump($group); // badhost detail
-
- $result[$label] = & $group;
- $hosts = array_diff($hosts, $result[$label]);
- return TRUE;
- } else {
- return FALSE;
- }
+ return $blocked;
}
// Default (enabled) methods and thresholds (for content insertion)
// Simple/fast spam check
function check_uri_spam($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 only
+ $blocked = array_merge_recursive($blocked, $_progress['blocked']);
+ $hosts = array_merge_recursive($hosts, $_progress['hosts']);
}
+
+ // Unique values
+ $blocked = array_unique_recursive($blocked);
+ $hosts = array_unique_recursive($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 $progress;
}
-// Count leaves
-function array_count_leaves($array = array(), $count_empty_array = FALSE)
+// Count leaves (A leaf = value that is not an array, or an empty array)
+function array_count_leaves($array = array(), $count_empty = FALSE)
{
- if (! is_array($array) || (empty($array) && $count_empty_array))
- return 1;
+ if (! is_array($array) || (empty($array) && $count_empty)) return 1;
// Recurse
- $result = 0;
+ $count = 0;
foreach ($array as $part) {
- $result += array_count_leaves($part, $count_empty_array);
+ $count += array_count_leaves($part, $count_empty);
}
- return $result;
+ return $count;
+}
+
+// An array-leaves to a flat array
+function array_flat_leaves($array, $unique = TRUE)
+{
+ if (! is_array($array)) return $array;
+
+ $tmp = array();
+ foreach(array_keys($array) as $key) {
+ if (is_array($array[$key])) {
+ // Recurse
+ foreach(array_flat_leaves($array[$key]) as $_value) {
+ $tmp[] = $_value;
+ }
+ } else {
+ $tmp[] = & $array[$key];
+ }
+ }
+
+ return $unique ? array_values(array_unique($tmp)) : $tmp;
+}
+
+// An array() to an array leaf
+function array_leaf($array = array('A', 'B', 'C.D'), $stem = FALSE, $edge = TRUE)
+{
+ if (! is_array($array)) return $array;
+
+ $leaf = array();
+ $tmp = & $leaf;
+ foreach($array as $arg) {
+ if (! is_string($arg) && ! is_int($arg)) continue;
+ $tmp[$arg] = array();
+ $parent = & $tmp;
+ $tmp = & $tmp[$arg];
+ }
+ if ($stem) {
+ $parent[key($parent)] = & $edge;
+ } else {
+ $parent = key($parent);
+ }
+
+ return $leaf; // array('A' => array('B' => 'C.D'))
}
+
// ---------------------
// Reporting
-// TODO: Don't show unused $method!
// Summarize $progress (blocked only)
function summarize_spam_progress($progress = array(), $blockedonly = FALSE)
{
$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($progress['blocked']) || empty($progress['blocked'])) return '';
+
+ // Flat per group
+ $blocked = array();
+ foreach($progress['blocked'] as $list => $lvalue) {
+ foreach($lvalue as $group => $gvalue) {
+ $flat = implode(', ', array_flat_leaves($gvalue));
+ if ($flat === $group) {
+ $blocked[$list][] = $flat;
+ } else {
+ $blocked[$list][$group] = $flat;
+ }
+ }
+ }
+
+ // Shrink per list
+ // From: 'A-1' => array('ie.to')
+ // To: 'A-1' => 'ie.to'
+ foreach($blocked as $list => $lvalue) {
+ if (is_array($lvalue) &&
+ count($lvalue) == 1 &&
+ is_numeric(key($lvalue))) {
+ $blocked[$list] = current($lvalue);
+ }
+ }
+
+ return var_export_shrink($blocked, TRUE, TRUE);
+}
+
+function summarize_detail_newtral($progress = array())
+{
+ if (! isset($progress['hosts']) ||
+ ! is_array($progress['hosts']) ||
+ empty($progress['hosts'])) return '';
+
+ $result = '';
+ if (FALSE) {
+ // Sort by domain
+ $tmp = array();
+ foreach($progress['hosts'] as $value) {
+ $tmp[delimiter_reverse($value)] = $value;
+ }
+ ksort($tmp, SORT_STRING);
+ $result = count($tmp) . ' (' .implode(', ', $tmp) . ')';
+ } else {
+ $tmp = array();
+ foreach($progress['hosts'] as $value) {
+ $tmp = array_merge_recursive(
+ $tmp,
+ array_leaf(explode('.', delimiter_reverse($value)), TRUE, $value)
+ );
+ }
+
+//var_dump($tmp);
+// TODO: IP address 1.2.3.4 => "0"-3-2-1 by array_shrinkbranch_leaves()
+
+ array_shrinkbranch_leaves($tmp, '.', TRUE); // "domain.tld"
+ array_joinbranch_leaf($tmp, '.', 0, TRUE);
+ foreach($tmp as $key => $value) {
+ if (is_array($value)) {
+ ksort($tmp[$key], SORT_STRING);
+ $tmp[$key] = implode(', ', array_flat_leaves($value));
+ }
+ }
+ ksort($tmp, SORT_STRING);
+
+ $result = var_export_shrink($tmp, TRUE, TRUE);
+ }
+
+ return $result;
+}
+
+// array('F' => array('B' => array('C' => array('d' => array('' => 'foobar')))))
+// to
+// array('F.B.C.d.' => 'foobar')
+function array_joinbranch_leaf(& $array, $delim = '.', $limit = 0, $reverse = FALSE)
+{
+ $result = array();
+ if (! is_array($array)) return $result; // Nothing to do
+
+ $limit = max(0, intval($limit));
+ $cstack = array();
+
+ foreach(array_keys($array) as $key) {
+ $kstack = array();
+ $k = -1;
+
+ $single = array($key => & $array[$key]); // Keep it single
+ $cursor = & $single;
+ while(is_array($cursor) && count($cursor) == 1) { // Once
+ ++$k;
+ $kstack[] = key($cursor);
+ $cursor = & $cursor[$kstack[$k]];
+ if ($limit != 0 && $k == $limit) break;
+ }
+
+ // Relink
+ if ($k != 0) {
+ if ($reverse) $kstack = array_reverse($kstack);
+ $joinkey = implode($delim, $kstack);
+
+ unset($array[$key]);
+ $array[$joinkey] = & $cursor;
+ $result[$joinkey] = $k + 1; // Key seems not an single array => joined length
+ }
+ }
+
+ return $result;
+}
+
+
+// array('A' => array('B' => 'C')) to
+// array('A.B' => 'C')
+// array(
+// 'A' => array(
+// 'B' => array(
+// 'C' => array(
+// 'D' => '1'
+// ),
+// ),
+// ),
+// 'G' => array(
+// 'H' => '2'
+// ),
+// )
+// to
+// array (
+// 'G.H' => '2',
+// 'A.B.C.D' => '1',
+// )
+function array_shrinkbranch_leaves(& $array, $delim = '.', $reverse = FALSE, $recurse = FALSE)
+{
+ $result = 0;
+ if (! is_array($array) || empty($array)) return $result;
+
+ foreach(array_keys($array) as $key) {
+ $branch = & $array[$key];
+ if (! is_array($branch) || empty($branch)) continue;
+
+ foreach(array_keys($branch) as $bkey) {
+ $joinkey = $reverse ?
+ $bkey . $delim . $key :
+ $key . $delim . $bkey;
+ $array[$joinkey] = & $branch[$bkey];
+ unset($array[$key]);
+ ++$result;
+ }
+ }
+
+ // Rescan (Recurse)
+ if ($recurse && $result) {
+ $result = array_shrinkbranch_leaves($array, $delim, $reverse, $recurse);
+ }
+
+ return $result; // Tell me how many
+}
+//$a = array (
+// 'edu' => array (
+// 'berkeley' => array (
+// 'polisci' => array (
+// '' => 'polisci.berkeley.edu',
+// ),
+// ),
+// 'cmich' => array (
+// 'rso' => array (
+// '' => 'rso.cmich.edu',
+// ),
+// ),
+// ),
+//);
+//array_shrinkbranch_leaves($a, '.', TRUE);
+//var_export($a);
+
+//$a = array (
+// '4' => array (
+// '5' => array (
+// '6' => array (
+// '' => '7.8.9',
+// ),
+// ),
+// ),
+//);
+//array_shrinkbranch_leaves($a, '.', TRUE);
+//var_export($a);
+
+
+
+// Check responsibility-root of the FQDN
+// 'foo.bar.example.com' => 'example.com' (.com has the last whois for it)
+// 'foo.bar.example.au' => 'example.au' (.au has the last whois for it)
+// 'foo.bar.example.edu.au' => 'example.edu.au' (.edu.au has the last whois for it)
+// 'foo.bar.example.act.edu.au' => 'example.act.edu.au' (.act.edu.au has the last whois for it)
+function whois_responsibility($fqdn = 'foo.bar.example.com', $implicit = TRUE)
+{
+ // Domains who have 2nd and/or 3rd level domains
+ static $domain = array(
+
+ // ccTLD: Australia
+ // http://www.auda.org.au/
+ // NIC : http://www.aunic.net/
+ // Whois: http://www.ausregistry.com.au/
+ 'au' => array(
+ // .au Second Level Domains
+ // http://www.auda.org.au/domains/
+ 'asn' => TRUE,
+ 'com' => TRUE,
+ 'conf' => TRUE,
+ 'csiro' => TRUE,
+ 'edu' => array( // http://www.domainname.edu.au/
+ 'act' => TRUE,
+ 'nt' => TRUE,
+ 'nsw' => TRUE,
+ 'qld' => TRUE,
+ 'sa' => TRUE,
+ 'tas' => TRUE,
+ 'vic' => TRUE,
+ 'wa' => TRUE,
+ ),
+ 'gov' => array(
+ 'act' => TRUE, // Australian Capital Territory
+ 'nt' => TRUE, // Northern Territory
+ 'nsw' => TRUE, // New South Wales
+ 'qld' => TRUE, // Queensland
+ 'sa' => TRUE, // South Australia
+ 'tas' => TRUE, // Tasmania
+ 'vic' => TRUE, // Victoria
+ 'wa' => TRUE, // Western Australia
+ ),
+ 'id' => TRUE,
+ 'net' => TRUE,
+ 'org' => TRUE,
+ 'info' => TRUE,
+ ),
+
+ // ccTLD: China
+ // NIC : http://www.cnnic.net.cn/en/index/
+ // Whois: http://ewhois.cnnic.cn/
+ 'cn' => array(
+ // Provisional Administrative Rules for Registration of Domain Names in China
+ // http://www.cnnic.net.cn/html/Dir/2003/11/27/1520.htm
+
+ // Organizational
+ 'ac' => TRUE,
+ 'com' => TRUE,
+ 'edu' => TRUE,
+ 'gov' => TRUE,
+ 'net' => TRUE,
+ 'org' => TRUE,
+
+ // Geographic
+ 'ah' => TRUE,
+ 'bj' => TRUE,
+ 'cq' => TRUE,
+ 'fj' => TRUE,
+ 'gd' => TRUE,
+ 'gs' => TRUE,
+ 'gx' => TRUE,
+ 'gz' => TRUE,
+ 'ha' => TRUE,
+ 'hb' => TRUE,
+ 'he' => TRUE,
+ 'hi' => TRUE,
+ 'hk' => TRUE,
+ 'hl' => TRUE,
+ 'hn' => TRUE,
+ 'jl' => TRUE,
+ 'js' => TRUE,
+ 'jx' => TRUE,
+ 'ln' => TRUE,
+ 'mo' => TRUE,
+ 'nm' => TRUE,
+ 'nx' => TRUE,
+ 'qh' => TRUE,
+ 'sc' => TRUE,
+ 'sd' => TRUE,
+ 'sh' => TRUE,
+ 'sn' => TRUE,
+ 'sx' => TRUE,
+ 'tj' => TRUE,
+ 'tw' => TRUE,
+ 'xj' => TRUE,
+ 'xz' => TRUE,
+ 'yn' => TRUE,
+ 'zj' => TRUE,
+ ),
+
+ // ccTLD: South Korea
+ // NIC : http://www.nic.or.kr/english/
+ // Whois: http://whois.nida.or.kr/english/
+ 'kr' => array(
+ // .kr domain policy [appendix 1] : Qualifications for Second Level Domains
+ // http://domain.nida.or.kr/eng/policy.jsp
+
+ // Organizational
+ 'co' => TRUE,
+ 'ne ' => TRUE,
+ 'or ' => TRUE,
+ 're ' => TRUE,
+ 'pe' => TRUE,
+ 'go ' => TRUE,
+ 'mil' => TRUE,
+ 'ac' => TRUE,
+ 'hs' => TRUE,
+ 'ms' => TRUE,
+ 'es' => TRUE,
+ 'sc' => TRUE,
+ 'kg' => TRUE,
+
+ // Geographic
+ 'seoul' => TRUE,
+ 'busan' => TRUE,
+ 'daegu' => TRUE,
+ 'incheon' => TRUE,
+ 'gwangju' => TRUE,
+ 'daejeon' => TRUE,
+ 'ulsan' => TRUE,
+ 'gyeonggi' => TRUE,
+ 'gangwon' => TRUE,
+ 'chungbuk' => TRUE,
+ 'chungnam' => TRUE,
+ 'jeonbuk' => TRUE,
+ 'jeonnam' => TRUE,
+ 'gyeongbuk' => TRUE,
+ 'gyeongnam' => TRUE,
+ 'jeju' => TRUE,
+ ),
+
+ // ccTLD: Japan
+ // NIC : http://jprs.co.jp/en/
+ // Whois: http://whois.jprs.jp/en/
+ 'jp' => array(
+ // Guide to JP Domain Name
+ // http://jprs.co.jp/en/jpdomain.html
+
+ // Organizational
+ 'ac' => TRUE,
+ 'ad' => TRUE,
+ 'co' => TRUE,
+ 'go' => TRUE,
+ 'gr' => TRUE,
+ 'lg' => TRUE,
+ 'ne' => TRUE,
+ 'or' => TRUE,
+
+ // Geographic
+ //
+ // Examples for 3rd level domains
+ //'kumamoto' => array(
+ // // http://www.pref.kumamoto.jp/link/list.asp#4
+ // 'amakusa' => TRUE,
+ // 'hitoyoshi' => TRUE,
+ // 'jonan' => TRUE,
+ // 'kumamoto' => TRUE,
+ // ...
+ //),
+ 'aichi' => TRUE,
+ 'akita' => TRUE,
+ 'aomori' => TRUE,
+ 'chiba' => TRUE,
+ 'ehime' => TRUE,
+ 'fukui' => TRUE,
+ 'fukuoka' => TRUE,
+ 'fukushima' => TRUE,
+ 'gifu' => TRUE,
+ 'gunma' => TRUE,
+ 'hiroshima' => TRUE,
+ 'hokkaido' => TRUE,
+ 'hyogo' => TRUE,
+ 'ibaraki' => TRUE,
+ 'ishikawa' => TRUE,
+ 'iwate' => TRUE,
+ 'kagawa' => TRUE,
+ 'kagoshima' => TRUE,
+ 'kanagawa' => TRUE,
+ 'kawasaki' => TRUE,
+ 'kitakyushu'=> TRUE,
+ 'kobe' => TRUE,
+ 'kochi' => TRUE,
+ 'kumamoto' => TRUE,
+ 'kyoto' => TRUE,
+ 'mie' => TRUE,
+ 'miyagi' => TRUE,
+ 'miyazaki' => TRUE,
+ 'nagano' => TRUE,
+ 'nagasaki' => TRUE,
+ 'nagoya' => TRUE,
+ 'nara' => TRUE,
+ 'niigata' => TRUE,
+ 'oita' => TRUE,
+ 'okayama' => TRUE,
+ 'okinawa' => TRUE,
+ 'osaka' => TRUE,
+ 'saga' => TRUE,
+ 'saitama' => TRUE,
+ 'sapporo' => TRUE,
+ 'sendai' => TRUE,
+ 'shiga' => TRUE,
+ 'shimane' => TRUE,
+ 'shizuoka' => TRUE,
+ 'tochigi' => TRUE,
+ 'tokushima' => TRUE,
+ 'tokyo' => TRUE,
+ 'tottori' => TRUE,
+ 'toyama' => TRUE,
+ 'wakayama' => TRUE,
+ 'yamagata' => TRUE,
+ 'yamaguchi' => TRUE,
+ 'yamanashi' => TRUE,
+ 'yokohama' => TRUE,
+ ),
+
+ // ccTLD: Ukraine
+ // NIC : http://www.nic.net.ua/
+ // Whois: http://whois.com.ua/
+ 'ua' => array(
+ // policy for alternative 2nd level domain names (a2ld)
+ // http://www.nic.net.ua/doc/a2ld
+ // http://whois.com.ua/
+ 'cherkassy' => TRUE, // www.cherkassy.ua
+ 'chernigov' => TRUE,
+ 'chernovtsy' => TRUE,
+ 'ck' => TRUE,
+ 'cn' => TRUE,
+ 'com' => TRUE,
+ 'crimea' => TRUE,
+ 'cv' => TRUE,
+ 'dn' => TRUE,
+ 'dnepropetrovsk' => TRUE,
+ 'donetsk' => TRUE,
+ 'dp' => TRUE,
+ 'edu' => TRUE,
+ 'gov' => TRUE,
+ 'if' => TRUE,
+ 'ivano-frankivsk' => TRUE,
+ 'kh' => TRUE,
+ 'kharkov' => TRUE,
+ 'kherson' => TRUE,
+ 'kiev' => TRUE,
+ 'kirovograd' => TRUE,
+ 'km' => TRUE,
+ 'kr' => TRUE,
+ 'ks' => TRUE,
+ 'lg' => TRUE,
+ 'lugansk' => TRUE,
+ 'lutsk' => TRUE,
+ 'lviv' => TRUE,
+ 'mk' => TRUE,
+ 'net' => TRUE,
+ 'nikolaev' => TRUE,
+ 'od' => TRUE,
+ 'odessa' => TRUE,
+ 'org' => TRUE,
+ 'pl' => TRUE,
+ 'poltava' => TRUE,
+ 'rovno' => TRUE,
+ 'rv' => TRUE,
+ 'sebastopol' => TRUE,
+ 'sumy' => TRUE,
+ 'te' => TRUE,
+ 'ternopil' => TRUE,
+ 'uz' => TRUE,
+ 'uzhgorod' => TRUE,
+ 'vinnica' => TRUE,
+ 'vn' => TRUE,
+ 'zaporizhzhe' => TRUE,
+ 'zhitomir' => TRUE,
+ 'zp' => TRUE,
+ 'zt' => TRUE,
+ ),
+
+ // ccTLD: United Kingdom
+ // NIC : http://www.nic.uk/
+ 'uk' => array(
+ // Second Level Domains
+ // http://www.nic.uk/registrants/aboutdomainnames/sld/
+ 'co' => TRUE,
+ 'ltd' => TRUE,
+ 'me' => TRUE,
+ 'net' => TRUE,
+ 'nic' => TRUE,
+ 'org' => TRUE,
+ 'plc' => TRUE,
+ 'sch' => TRUE,
+
+ // Delegated Second Level Domains
+ // http://www.nic.uk/registrants/aboutdomainnames/sld/delegated/
+ 'ac' => TRUE,
+ 'gov' => TRUE,
+ 'mil' => TRUE,
+ 'mod' => TRUE,
+ 'nhs' => TRUE,
+ 'police' => TRUE,
+ ),
+
+ // ccTLD: United States of America
+ // NIC : http://nic.us/
+ // Whois: http://whois.us/
+ 'us' => array(
+ // RFC1480
+
+ // United States Postal Service: State abbreviations (for postal codes)
+ // http://www.usps.com/ncsc/lookups/abbreviations.html
+ 'ak' => TRUE, // Alaska
+ 'al' => TRUE, // Alabama
+ 'ar' => TRUE, // Arkansas
+ 'as' => TRUE, // American samoa
+ 'az' => TRUE, // Arizona
+ 'ca' => TRUE, // California
+ 'co' => TRUE, // Colorado
+ 'ct' => TRUE, // Connecticut
+ 'dc' => TRUE, // District of Columbia
+ 'de' => TRUE, // Delaware
+ 'fl' => TRUE, // Florida
+ 'fm' => TRUE, // Federated states of Micronesia
+ 'ga' => TRUE, // Georgia
+ 'gu' => TRUE, // Guam
+ 'hi' => TRUE, // Hawaii
+ 'ia' => TRUE, // Iowa
+ 'id' => TRUE, // Idaho
+ 'il' => TRUE, // Illinois
+ 'in' => TRUE, // Indiana
+ 'ks' => TRUE, // Kansas
+ 'ky' => TRUE, // Kentucky
+ 'la' => TRUE, // Louisiana
+ 'ma' => TRUE, // Massachusetts
+ 'md' => TRUE, // Maryland
+ 'me' => TRUE, // Maine
+ 'mh' => TRUE, // Marshall Islands
+ 'mi' => TRUE, // Michigan
+ 'mn' => TRUE, // Minnesota
+ 'mo' => TRUE, // Missouri
+ 'mp' => TRUE, // Northern mariana islands
+ 'ms' => TRUE, // Mississippi
+ 'mt' => TRUE, // Montana
+ 'nc' => TRUE, // North Carolina
+ 'nd' => TRUE, // North Dakota
+ 'ne' => TRUE, // Nebraska
+ 'nh' => TRUE, // New Hampshire
+ 'nj' => TRUE, // New Jersey
+ 'nm' => TRUE, // New Mexico
+ 'nv' => TRUE, // Nevada
+ 'ny' => TRUE, // New York
+ 'oh' => TRUE, // Ohio
+ 'ok' => TRUE, // Oklahoma
+ 'or' => TRUE, // Oregon
+ 'pa' => TRUE, // Pennsylvania
+ 'pr' => TRUE, // Puerto Rico
+ 'pw' => TRUE, // Palau
+ 'ri' => TRUE, // Rhode Island
+ 'sc' => TRUE, // South Carolina
+ 'sd' => TRUE, // South Dakota
+ 'tn' => TRUE, // Tennessee
+ 'tx' => TRUE, // Texas
+ 'ut' => TRUE, // Utah
+ 'va' => TRUE, // Virginia
+ 'vi' => TRUE, // Virgin Islands
+ 'vt' => TRUE, // Vermont
+ 'wa' => TRUE, // Washington
+ 'wi' => TRUE, // Wisconsin
+ 'wv' => TRUE, // West Virginia
+ 'wy' => TRUE, // Wyoming
+
+ // Others
+ 'dni',
+ 'fed',
+ 'isa',
+ 'kids',
+ 'nsn',
+ ),
+ );
+
+ if (! is_string($fqdn)) return '';
+
+ $result = array();
+ $dcursor = & $domain;
+ $array = array_reverse(explode('.', $fqdn));
+ $i = 0;
+ while(TRUE) {
+ $acursor = $array[$i];
+ if (is_array($dcursor) && isset($dcursor[$acursor])) {
+ $result[] = & $array[$i];
+ $dcursor = & $dcursor[$acursor];
+ } else {
+ if (isset($acursor)) {
+ $result[] = & $array[$i]; // Whois servers must know this subdomain
+ }
+ break;
+ }
+ ++$i;
+ }
+
+ // Implicit responsibility: Top-Level-Domains must not be yours
+ // 'bar.foo.something' => 'foo.something'
+ if ($implicit && count($result) == 1 && count($array) > 1) {
+ $result[] = & $array[1];
+ }
+
+ return $result ? implode('.', array_reverse($result)) : '';
+}
+
+
// ---------------------
// Exit
+// Freeing memories
+function spam_dispose()
+{
+ get_blocklist(NULL);
+}
+
// Common bahavior for blocking
// NOTE: Call this function from various blocking feature, to disgueise the reason 'why blocked'
function spam_exit($mode = '', $data = array())
{
+
+ $exit = TRUE;
switch ($mode) {
- case '': echo("\n"); break;
+ case '':
+ echo("\n");
+ break;
case 'dump':
echo('<pre>' . "\n");
echo htmlspecialchars(var_export($data, TRUE));
break;
};
- // Force exit
- exit;
+ if ($exit) exit; // Force exit
}
{
$progress = check_uri_spam($target, $method);
- if (! empty($progress['is_spam'])) {
- // Mail to administrator(s)
+ if (empty($progress['is_spam'])) {
+ spam_dispose();
+ } else {
+ $target = string($target, 0); // Removing "\0" etc
pkwk_spamnotify($action, $page, $target, $progress, $method);
-
- // Exit
spam_exit($exitmode, $progress);
}
}
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);
OSDN Git Service
