X-Git-Url: http://git.osdn.net/view?a=blobdiff_plain;f=spam%2Fspam.php;h=0b2d91a2422dfde21a685d84f722e31f29d42e61;hb=cc65330c5c3ac80f3d24348774a28af01849a86a;hp=9c54d3d4f44ed438a7b466a73744d2547d5be0d3;hpb=a47b7cf6e562d552289d1915c04e0e2be2fc0f5a;p=pukiwiki%2Fpukiwiki_sandbox.git
diff --git a/spam/spam.php b/spam/spam.php
index 9c54d3d..0b2d91a 100644
--- a/spam/spam.php
+++ b/spam/spam.php
@@ -1,5 +1,5 @@
1) {
+ $min_len = min(1024, intval($min_len));
+ $binary =
+ implode("\n",
+ preg_grep('/^.{' . $min_len . ',}/S',
+ explode("\n", $binary)
+ )
+ );
+ }
+
+ return $binary;
+}
+
+
// ---------------------
// URI pickup
@@ -110,9 +152,9 @@ function uri_pickup_normalize(& $pickups, $destructive = TRUE)
if ($destructive) {
foreach (array_keys($pickups) as $key) {
$_key = & $pickups[$key];
- $_key['scheme'] = isset($_key['scheme']) ? scheme_normalize($_key['scheme']) : '';
+ $_key['scheme'] = isset($_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) : '';
+ $_key['port'] = isset($_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']) : '';
@@ -121,14 +163,13 @@ function uri_pickup_normalize(& $pickups, $destructive = TRUE)
} else {
foreach (array_keys($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']) : '';
+ $_key['scheme'] = isset($_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 $pickups;
}
@@ -217,7 +258,7 @@ function area_pickup($string = '', $method = array())
// [OK] visit http://nasty.example.com/
// [OK] discount foobar
// [NG] visit http://ng.example.com _not_ended_
- $regex = '#]*\bhref\b[^>]*>.*?]*(>)#i';
+ $regex = '#]*\bhref\b[^>]*>.*?]*(>)#is';
if (isset($method['area_anchor'])) {
$areas = array();
$count = isset($method['asap']) ?
@@ -243,7 +284,7 @@ function area_pickup($string = '', $method = array())
// [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']) ?
@@ -332,15 +373,32 @@ function _preg_replace_callback_domain_exposure($matches = array())
// Preprocess: rawurldecode() and adding space(s) and something
// to detect/count some URIs _if possible_
// NOTE: It's maybe danger to var_dump(result). [e.g. 'javascript:']
+// [OK] http://victim.example.org/?site:nasty.example.org
+// [OK] http://victim.example.org/nasty.example.org
// [OK] http://victim.example.org/go?http%3A%2F%2Fnasty.example.org
// [OK] http://victim.example.org/http://nasty.example.org
-// TODO: link.toolbot.com, urlx.org
function spam_uri_pickup_preprocess($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())
$string = preg_replace_callback(
array(
@@ -742,6 +800,7 @@ function file_normalize($file = 'index.html.en')
// [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('&', $string);
@@ -807,6 +866,8 @@ function generate_glob_regex($string = '', $divider = '/')
// 23 => ']',
);
+ if (! is_string($string)) return '';
+
$string = str_replace($from, $mid, $string); // Hide
$string = preg_quote($string, $divider);
$string = str_replace($mid, $to, $string); // Unhide
@@ -839,6 +900,8 @@ function is_ip($string = '')
// TODO: IPv4, CIDR?, IPv6
function generate_host_regex($string = '', $divider = '/')
{
+ if (! is_string($string)) return '';
+
if (mb_strpos($string, '.') === FALSE)
return generate_glob_regex($string, $divider);
@@ -874,8 +937,17 @@ function get_blocklist($list = '')
// '*.blogspot.com', // Blog services's subdomains (only)
// 'IANA-examples' => '#^(?:.*\.)?example\.(?:com|net|org)$#',
// );
- foreach(array('goodhost', 'badhost') as $_list) {
- if (! isset($blocklist[$list])) continue;
+ if (isset($blocklist['list'])) {
+ $regexs['list'] = & $blocklist['list'];
+ } else {
+ // Default
+ $blocklist['list'] = array(
+ 'goodhost' => FALSE,
+ 'badhost' => TRUE,
+ );
+ }
+ foreach(array_keys($blocklist['list']) as $_list) {
+ if (! isset($blocklist[$_list])) continue;
foreach ($blocklist[$_list] as $key => $value) {
if (is_array($value)) {
$regexs[$_list][$key] = array();
@@ -886,6 +958,7 @@ function get_blocklist($list = '')
get_blocklist_add($regexs[$_list], $key, $value);
}
}
+ unset($blocklist[$_list]);
}
}
}
@@ -907,53 +980,55 @@ function get_blocklist_add(& $array, $key = 0, $value = '*.example.org')
} else {
$array[$value] = '/^' . generate_host_regex($value, '/') . '$/i';
}
-}
+}
-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('goodhost') as $regex) {
- $hosts = preg_grep_invert($regex, $hosts);
- }
- if (empty($hosts)) return $result;
-
- $tmp = array();
- foreach (get_blocklist('badhost') as $label => $regex) {
- if (is_array($regex)) {
- $result[$label] = array();
- foreach($regex as $_label => $_regex) {
- if (is_badhost_avail($_label, $_regex, $hosts, $result[$label]) && $asap) break;
+ if (! is_array($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;
+ }
}
- if (empty($result[$label])) unset($result[$label]);
- } else {
- if (is_badhost_avail($label, $regex, $hosts, $result) && $asap) break;
}
}
- $remains = $hosts;
-
- return $result;
+ return $blocked;
}
-// Subroutine for is_badhost()
-function is_badhost_avail($label = '*.example.org', $regex = '/^.*\.example\.org$/', & $hosts, & $result)
+// Simple example for badhost (not used now)
+function is_badhost($hosts = array(), $asap = TRUE, $bool = TRUE)
{
- $group = preg_grep($regex, $hosts);
- if ($group) {
- $result[$label] = & $group;
- $hosts = array_diff($hosts, $result[$label]);
- return TRUE;
- } else {
- return FALSE;
+ $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
}
+
+ return $bool ? ! empty($blocked) : $blocked;
}
+
// Default (enabled) methods and thresholds (for content insertion)
function check_uri_spam_method($times = 1, $t_area = 0, $rule = TRUE)
{
@@ -994,59 +1069,84 @@ function check_uri_spam_method($times = 1, $t_area = 0, $rule = TRUE)
// Simple/fast spam check
function check_uri_spam($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
+ ),
+ 'remains' => array(
),
- 'is_spam' => array(),
- 'method' => & $method,
- 'remains' => array(),
- 'error' => array(),
);
+
+ // Aliases
$sum = & $progress['sum'];
$is_spam = & $progress['is_spam'];
+ $progress['method'] = & $method; // Argument
$remains = & $progress['remains'];
- $error = & $progress['error'];
$asap = isset($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];
+ if (! isset($sum[$key])) {
+ $sum[$key] = & $_sum[$key];
+ } else {
+ $sum[$key] += $_sum[$key];
+ }
}
+
+ // Merge $is_spam
+ $_is_spam = & $_progress['is_spam'];
foreach (array_keys($_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];
+ $is_spam[$key][$_key] = & $_is_spam[$key][$_key];
} else {
$is_spam[$key][$_key] += $_is_spam[$key][$_key];
}
}
} else {
$is_spam[$key] = TRUE;
+ if ($asap) break;
}
}
- foreach ($_remains as $key=>$value) {
+ if ($asap && $is_spam) break;
+
+ // Merge $remains
+ foreach ($_progress['remains'] as $key=>$value) {
foreach ($value as $_key=>$_value) {
if (is_int($_key)) {
$remains[$key][] = $_value;
@@ -1055,8 +1155,6 @@ function check_uri_spam($target = '', $method = array())
}
}
}
- if (! empty($_error)) $error += $_error;
- if ($asap && $is_spam) break;
}
return $progress;
}
@@ -1091,8 +1189,7 @@ function check_uri_spam($target = '', $method = array())
if ($asap && $is_spam) return $progress;
// URI: Pickup
- $pickups = spam_uri_pickup($target, $method);
- //$remains['uri_pickup'] = & $pickups;
+ $pickups = uri_pickup_normalize(spam_uri_pickup($target, $method));
// Return if ...
if (empty($pickups)) return $progress;
@@ -1140,8 +1237,6 @@ function check_uri_spam($target = '', $method = array())
// URI: Uniqueness (and removing non-uniques)
if ((! $asap || ! $is_spam) && isset($method['non_uniquri'])) {
- uri_pickup_normalize($pickups);
-
$uris = array();
foreach (array_keys($pickups) as $key) {
$uris[$key] = uri_pickup_implode($pickups[$key]);
@@ -1168,7 +1263,6 @@ function check_uri_spam($target = '', $method = array())
$hosts = array();
foreach ($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'] - $sum['uniqhost'];
@@ -1180,45 +1274,56 @@ function check_uri_spam($target = '', $method = array())
// Return if ...
if ($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;
- }
+
+ // 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 (! $asap && $hosts) {
+ $remains['badhost'] = array();
+ foreach ($hosts 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 $keys) {
+
+ if (! empty($blocked)) {
+
+ //var_dump($blocked); // BADHOST detail
+
+ $sum['badhost'] += array_count_leaves($blocked);
+ foreach(array_keys($blocked) as $keys) {
$is_spam['badhost'][$keys] =
- array_count_leaves($badhost[$keys]);
+ array_count_leaves($blocked[$keys]);
}
- unset($badhost);
}
}
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 $count;
+}
+
+// Merge two leaves
+function array_merge_leaves($array1 = array(), $array2 = array())
+{
+ return array_merge_recursive($array1, $array2);
}
// ---------------------
@@ -1245,6 +1350,26 @@ function summarize_spam_progress($progress = array(), $blockedonly = FALSE)
return implode(', ', $tmp);
}
+function summarize_detail_badhost($progress = array())
+{
+ if (! isset($progress['is_spam']['badhost'])) return '';
+
+ $badhost = array();
+ foreach($progress['is_spam']['badhost'] as $glob=>$number) {
+ $badhost[] = $glob . '(' . $number . ')';
+ }
+ return implode(', ', $badhost);
+}
+
+function summarize_detail_newtral($progress = array())
+{
+ if (! isset($progress['remains']['badhost'])) return '';
+
+ return count($progress['remains']['badhost']) .
+ ' (' . implode(', ', array_keys($progress['remains']['badhost'])) . ')';
+}
+
+
// ---------------------
// Exit
@@ -1300,23 +1425,13 @@ function pkwk_spamnotify($action, $page, $target = array('title' => ''), $progre
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);