<?php
-// $Id: spam.php,v 1.59 2006/12/04 12:53:40 henoheno Exp $
-// Copyright (C) 2006 PukiWiki Developers Team
+// $Id: spam.php,v 1.122 2007/03/04 03:59:37 henoheno Exp $
+// Copyright (C) 2006-2007 PukiWiki Developers Team
// License: GPL v2 or (at your option) any later version
-
+//
// Functions for Concept-work of spam-uri metrics
+//
+// (PHP 4 >= 4.3.0): preg_match_all(PREG_OFFSET_CAPTURE): $method['uri_XXX'] related feature
if (! defined('SPAM_INI_FILE')) define('SPAM_INI_FILE', 'spam.ini.php');
// ---------------------
+// Compat etc
+
+// (PHP 4 >= 4.2.0): var_export(): mail-reporting and dump related
+if (! function_exists('var_export')) {
+ function var_export() {
+ return 'var_export() is not found on this server' . "\n";
+ }
+}
+
+// (PHP 4 >= 4.2.0): preg_grep() enables invert option
+function preg_grep_invert($pattern = '//', $input = array())
+{
+ static $invert;
+ if (! isset($invert)) $invert = defined('PREG_GREP_INVERT');
+
+ if ($invert) {
+ return preg_grep($pattern, $input, PREG_GREP_INVERT);
+ } else {
+ $result = preg_grep($pattern, $input);
+ if ($result) {
+ return array_diff($input, preg_grep($pattern, $input));
+ } else {
+ return $input;
+ }
+ }
+}
+
+// ---------------------
// URI pickup
// Return an array of URIs in the $string
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
- '[^\s<>"\'\[\]:/\#?]+' . // FQDN: foo.example.org
+ '[a-z0-9.-]+' . // hostname(FQDN) : foo.example.org
')' .
'(?::([0-9]*))?' . // 4: Port
'((?:/+[^\s<>"\'\[\]/\#]+)*/+)?' . // 5: Directory path or path-info
'#i',
$string, $array, PREG_SET_ORDER | PREG_OFFSET_CAPTURE
);
- //var_dump(recursive_map('htmlspecialchars', $array));
// Shrink $array
static $parts = array(
if ($normalize) {
$_uri['scheme'] = scheme_normalize($_uri['scheme']);
if ($_uri['scheme'] === '') {
- unset ($array[$uri]);
+ unset($array[$uri]);
continue;
}
$_uri['host'] = strtolower($_uri['host']);
$_uri['fragment']
);
}
+
+ // Area offset for area_measure()
$_uri['area']['offset'] = $offset;
}
}
// ---------------------
+// Area pickup
+
+// Pickup all of markup areas
+function area_pickup($string = '', $method = array())
+{
+ $area = array();
+ if (empty($method)) return $area;
+
+ // Anchor tag pair by preg_match and preg_match_all()
+ // [OK] <a href></a>
+ // [OK] <a href= >Good site!</a>
+ // [OK] <a href= "#" >test</a>
+ // [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_
+ $regex = '#<a\b[^>]*\bhref\b[^>]*>.*?</a\b[^>]*(>)#i';
+ if (isset($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 (<a href>)
+ $areas[$_area][1][1], // Area end (</a>)
+ );
+ }
+ 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 = '#\[(url|link)\b[^\]]*\].*?\[/\1\b[^\]]*(\])#i';
+ 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] += $belief;
+ }
+ }
+ }
+ }
+}
+
+// ---------------------
// Spam-uri pickup
// Domain exposure callback (See spam_uri_pickup_preprocess())
// NOTE: It's maybe danger to var_dump(result). [e.g. 'javascript:']
// [OK] http://victim.example.org/go?http%3A%2F%2Fnasty.example.org
// [OK] http://victim.example.org/http://nasty.example.org
+// TODO: link.toolbot.com, urlx.org
function spam_uri_pickup_preprocess($string = '')
{
if (! is_string($string)) return '';
// Domain exposure (See _preg_replace_callback_domain_exposure())
$string = preg_replace_callback(
array(
- // Something Google: http://www.google.com/supported_domains
- '#(http)://([a-z0-9.]+\.google\.[a-z]{2,3}(?:\.[a-z]{2})?)/' .
- '([a-z0-9?=&.%_+-]+)' . // ?query=foo+
- '\bsite:([a-z0-9.%_-]+)' . // site:nasty.example.com
+ '#(http)://' .
+ '(' .
+ // Something Google: http://www.google.com/supported_domains
+ '(?:[a-z0-9.]+\.)?google\.[a-z]{2,3}(?:\.[a-z]{2})?' .
+ '|' .
+ // AltaVista
+ '(?:[a-z0-9.]+\.)?altavista.com' .
+
+ ')' .
+ '/' .
+ '([a-z0-9?=&.%_/\'\\\+-]+)' . // path/?query=foo+bar+
+ '\bsite:([a-z0-9.%_-]+\.[a-z0-9.%_-]+)' . // site:nasty.example.com
//'()' . // Preserve or remove?
'#i',
),
}
// Main function of spam-uri pickup
-function spam_uri_pickup($string = '', $area = array())
+function spam_uri_pickup($string = '', $method = array())
{
- if (! is_array($area) || empty($area)) {
- $area = array(
- 'anchor' => TRUE,
- 'bbcode' => TRUE,
- );
+ if (! is_array($method) || empty($method)) {
+ $method = check_uri_spam_method();
}
$string = spam_uri_pickup_preprocess($string);
$array = uri_pickup($string);
- // Area elevation for '(especially external)link' intension
+ // Area elevation of URIs, for '(especially external)link' intension
if (! empty($array)) {
-
- $area_shadow = array();
- foreach(array_keys($array) as $key){
- $area_shadow[$key] = & $array[$key]['area'];
- $area_shadow[$key]['anchor'] = 0;
- $area_shadow[$key]['bbcode'] = 0;
- }
-
- // Anchor tags by preg_match_all()
- // [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_
- // [NG] <a href= >Good site!</a> <a href= "#" >test</a>
- if (isset($area['anchor'])) {
- $areas = array();
- preg_match_all('#<a\b[^>]*href[^>]*>.*?</a\b[^>]*(>)#i',
- $string, $areas, PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
- //var_dump(recursive_map('htmlspecialchars', $areas));
- foreach(array_keys($areas) as $_area) {
- $areas[$_area] = array(
- $areas[$_area][0][1], // Area start (<a href>)
- $areas[$_area][1][1], // Area end (</a>)
- );
+ $_method = array();
+ if (isset($method['uri_anchor'])) $_method['uri_anchor'] = & $method['uri_anchor'];
+ if (isset($method['uri_bbcode'])) $_method['uri_bbcode'] = & $method['uri_bbcode'];
+ $areas = area_pickup($string, $_method, TRUE);
+ if (! empty($areas)) {
+ $area_shadow = array();
+ foreach (array_keys($array) as $key) {
+ $area_shadow[$key] = & $array[$key]['area'];
+ foreach (array_keys($_method) as $_key) {
+ $area_shadow[$key][$_key] = 0;
+ }
}
- area_measure($areas, $area_shadow, 1, 'anchor');
- }
-
- // phpBB's "BBCode" by preg_match_all()
- // [url]http://nasty.example.com/[/url]
- // [link]http://nasty.example.com/[/link]
- // [url=http://nasty.example.com]visit http://nasty.example.com/[/url]
- // [link http://nasty.example.com/]buy something[/link]
- // ?? [url=][/url]
- if (isset($area['bbcode'])) {
- $areas = array();
- preg_match_all('#\[(url|link)\b[^\]]*\].*?\[/\1\b[^\]]*(\])#i',
- $string, $areas, PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
- //var_dump(recursive_map('htmlspecialchars', $areas));
- foreach(array_keys($areas) as $_area) {
- $areas[$_area] = array(
- $areas[$_area][0][1], // Area start ([url])
- $areas[$_area][2][1], // Area end ([/url])
- );
+ foreach (array_keys($_method) as $_key) {
+ if (isset($areas[$_key])) {
+ area_measure($areas[$_key], $area_shadow, 1, $_key);
+ }
}
- area_measure($areas, $area_shadow, 1, 'bbcode');
}
-
- // Various Wiki syntax
- // [text_or_uri>text_or_uri]
- // [text_or_uri:text_or_uri]
- // [text_or_uri|text_or_uri]
- // [text_or_uri->text_or_uri]
- // [text_or_uri text_or_uri] // MediaWiki
- // MediaWiki: [http://nasty.example.com/ visit http://nasty.example.com/]
-
- // Remove 'offset's for area_measure()
- foreach(array_keys($array) as $key)
- unset($array[$key]['area']['offset']);
}
+ // Remove 'offset's for area_measure()
+ foreach(array_keys($array) as $key)
+ unset($array[$key]['area']['offset']);
+
return $array;
}
-// If in doubt, it's a little doubtful
-function area_measure($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;
- }
- }
- }
- }
-}
// ---------------------
// Normalization
return $path;
}
-// File normalize (Destructive and rough)
-function file_normalize($string = '')
+// DirectoryIndex normalize (Destructive and rough)
+function file_normalize($string = 'index.html.en')
{
static $array = array(
- 'index' => TRUE,
+ 'index' => TRUE, // Some system can omit the suffix
'index.htm' => TRUE,
'index.html' => TRUE,
'index.shtml' => TRUE,
'index.php' => TRUE,
'index.php3' => TRUE,
'index.php4' => TRUE,
- 'index.pl' => TRUE,
- 'index.py' => TRUE,
- 'index.rb' => TRUE,
+ //'index.pl' => TRUE,
+ //'index.py' => TRUE,
+ //'index.rb' => TRUE,
'index.cgi' => TRUE,
'default.htm' => TRUE,
'default.html' => TRUE,
'default.aspx' => TRUE,
);
- if (isset($array[strtolower($string)])) {
+ // Content-negothiation filter:
+ // Roughly removing ISO 639 -like
+ // 2-letter suffixes (See RFC3066)
+ $matches = array();
+ if (preg_match('/(.*)\.[a-z][a-z](?:-[a-z][a-z])?$/i', $string, $matches)) {
+ $_string = $matches[1];
+ } else {
+ $_string = & $string;
+ }
+
+ if (isset($array[strtolower($_string)])) {
return '';
} else {
return $string;
// ---------------------
// Part One : Checker
+// Rough implementation of globbing
+//
+// USAGE: $regex = '/^' . generate_glob_regex('*.txt', '/') . '$/i';
+//
function generate_glob_regex($string = '', $divider = '/')
{
static $from = array(
}
}
-// TODO: Ignore list
-// TODO: preg_grep() ?
-// TODO: Multi list
-function is_badhost($hosts = '', $asap = TRUE)
+function get_blocklist($list = '')
{
- static $regex;
-
- if (! isset($regex)) {
- $regex = array();
- $regex['badhost'] = array();
-
- // Sample
- if (TRUE) {
- $blocklist['badhost'] = array(
- //'*', // Deny all uri
- //'10.20.*.*', // 10.20.example.com also matches
- //'*.blogspot.com', // Blog services subdomains
- //array('blogspot.com', '*.blogspot.com')
- );
- foreach ($blocklist['badhost'] as $part) {
- $regex['badhost'][] = '/^' . generate_glob_regex($part) . '$/i';
- }
- }
+ static $regexs;
- // Load
+ if (! isset($regexs)) {
+ $regexs = array();
if (file_exists(SPAM_INI_FILE)) {
$blocklist = array();
- require(SPAM_INI_FILE);
- foreach ($blocklist['badhost'] as $part) {
- $regex['badhost'][] = '/^' . generate_glob_regex($part) . '$/i';
+ include(SPAM_INI_FILE);
+ // $blocklist['badhost'] = array(
+ // '*.blogspot.com', // Blog services's subdomains (only)
+ // 'IANA-examples' => '#^(?:.*\.)?example\.(?:com|net|org)$#',
+ // );
+ foreach(array('goodhost', 'badhost') as $_list) {
+ if (! isset($blocklist[$list])) continue;
+ foreach ($blocklist[$_list] as $key => $value) {
+ if (is_string($key)) {
+ $regexs[$_list][$key] = $value;
+ } else {
+ $regexs[$_list][$value] =
+ '/^(?:www\.)?' . generate_glob_regex($value, '/') . '$/i';
+ }
+ }
}
}
}
- //var_dump($regex);
- $result = 0;
+ if ($list == '') {
+ return $regexs; // ALL
+ } else if (isset($regexs[$list])) {
+ return $regexs[$list];
+ } else {
+ return array();
+ }
+}
+
+function is_badhost($hosts = array(), $asap = TRUE, & $remains)
+{
+ $result = array();
if (! is_array($hosts)) $hosts = array($hosts);
+ foreach(array_keys($hosts) as $key) {
+ if (! is_string($hosts[$key])) unset($hosts[$key]);
+ }
+ if (empty($hosts)) return $result;
- foreach($hosts as $host) {
- if (! is_string($host)) $host = '';
+ foreach (get_blocklist('goodhost') as $regex) {
+ $hosts = preg_grep_invert($regex, $hosts);
+ }
+ if (empty($hosts)) return $result;
- // badhost
- foreach ($regex['badhost'] as $_regex) {
- if (preg_match($_regex, $host)) {
- ++$result;
- if ($asap) {
- return $result;
- } else {
- break;
- }
- }
+ $tmp = array();
+ foreach (get_blocklist('badhost') as $label => $regex) {
+ $result[$label] = preg_grep($regex, $hosts);
+ if (empty($result[$label])) {
+ unset($result[$label]);
+ } else {
+ $hosts = array_diff($hosts, $result[$label]);
+ if ($asap) break;
}
}
+ $remains = $hosts;
+
return $result;
}
-// Default (enabled) methods and thresholds
+// Default (enabled) methods and thresholds (for content insertion)
function check_uri_spam_method($times = 1, $t_area = 0, $rule = TRUE)
{
$times = intval($times);
$t_area = intval($t_area);
- // Thresholds
- $method = array(
- 'quantity' => 8 * $times, // Allow N URIs
- 'non_uniq' => 3 * $times, // Allow N duped (and normalized) URIs
- );
-
- // Areas
- $area = array(
- //'total' => $t_area, // Allow N areas total, enabled below
- 'anchor' => $t_area, // Inside <a href> HTML tag
- 'bbcode' => $t_area, // Inside [url] or [link] BBCode
+ $positive = array(
+ // Thresholds
+ 'quantity' => 8 * $times, // Allow N URIs
+ 'non_uniqhost' => 3 * $times, // Allow N duped (and normalized) Hosts
+ //'non_uniquri'=> 3 * $times, // Allow N duped (and normalized) URIs
+
+ // Areas
+ 'area_anchor' => $t_area, // Using <a href> HTML tag
+ 'area_bbcode' => $t_area, // Using [url] or [link] BBCode
+ //'uri_anchor' => $t_area, // URI inside <a href> HTML tag
+ //'uri_bbcode' => $t_area, // URI inside [url] or [link] BBCode
);
-
- // Rules
- $rules = array(
- 'badhost' => TRUE, // Check badhost
- );
-
- // Remove unused
- foreach (array_keys($method) as $key) {
- if ($method[$key] < 0) unset($method[$key]);
+ if ($rule) {
+ $bool = array(
+ // Rules
+ //'asap' => TRUE, // Quit or return As Soon As Possible
+ 'uniqhost' => TRUE, // Show uniq host (at block notification mail)
+ 'badhost' => TRUE, // Check badhost
+ );
+ } else {
+ $bool = array();
}
- foreach (array_keys($area) as $key) {
- if ($area[$key] < 0) unset($area[$key]);
+
+ // Remove non-$positive values
+ foreach (array_keys($positive) as $key) {
+ if ($positive[$key] < 0) unset($positive[$key]);
}
- $area = empty($area) ? array() : array('area' => $area);
- $rules = $rule ? $rules : array();
- return $method + $area + $rules;
+ return $positive + $bool;
}
-// TODO: Simplify $progress data structure
-// TODO: Simplify. !empty(['_action']) just means $is_spam
// Simple/fast spam check
-function check_uri_spam($target = '', $method = array(), $asap = TRUE)
+function check_uri_spam($target = '', $method = array())
{
- $is_spam = FALSE;
- $progress = array(
- 'quantity' => 0,
- 'area' => array(
- 'total' => 0,
- 'anchor' => 0,
- 'bbcode' => 0,
- ),
- 'non_uniq' => 0,
- 'uniqhost' => 0,
- 'badhost' => 0,
- '_action' => array(),
- );
-
if (! is_array($method) || empty($method)) {
$method = check_uri_spam_method();
}
+ $progress = array(
+ '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,
+ ),
+ 'is_spam' => array(),
+ 'method' => & $method,
+ 'remains' => array(),
+ );
+ $sum = & $progress['sum'];
+ $is_spam = & $progress['is_spam'];
+ $remains = & $progress['remains'];
+ $asap = isset($method['asap']);
+ // Recurse
if (is_array($target)) {
- // Recurse
foreach($target as $str) {
- list($_is_spam, $_progress) = check_uri_spam($str, $method, $asap);
- $is_spam = $is_spam || $_is_spam;
- $progress['quantity'] += $_progress['quantity'];
- $progress['area']['total'] += $_progress['area']['total'];
- $progress['area']['anchor'] += $_progress['area']['anchor'];
- $progress['area']['bbcode'] += $_progress['area']['bbcode'];
- $progress['non_uniq'] += $_progress['non_uniq'];
- $progress['uniqhost'] += $_progress['uniqhost'];
- $progress['badhost'] += $_progress['badhost'];
- foreach($_progress['_action'] as $key => $value) {
- if (is_array($value)) {
- foreach(array_keys($value) as $_key) {
- $progress['_action'][$key][$_key] = TRUE;
+ // Recurse
+ $_progress = check_uri_spam($str, $method);
+ $_sum = & $_progress['sum'];
+ $_is_spam = & $_progress['is_spam'];
+ $_remains = & $_progress['remains'];
+ foreach (array_keys($_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];
+ }
}
} else {
- $progress['_action'][$key] = TRUE;
+ $is_spam[$key] = TRUE;
}
}
- if ($is_spam && $asap) break;
- }
- } else {
- $pickups = spam_uri_pickup($target);
- if (! empty($pickups)) {
- $progress['quantity'] += count($pickups);
-
- // URI quantity
- if ((! $is_spam || ! $asap) && isset($method['quantity']) &&
- $progress['quantity'] > $method['quantity']) {
- $is_spam = TRUE;
- $progress['_action']['quantity'] = TRUE;
- }
- //var_dump($method['quantity'], $is_spam);
-
- // Using invalid area
- if ((! $is_spam || ! $asap) && isset($method['area'])) {
- foreach($pickups as $pickup) {
- foreach ($pickup['area'] as $key => $value) {
- if ($key == 'offset') continue;
- $progress['area']['total'] += $value;
- $progress['area'][$key] += $value;
- if (isset($method['area']['total']) &&
- $progress['area']['total'] > $method['area']['total']) {
- $is_spam = TRUE;
- $progress['_action']['area']['total'] = TRUE;
- if ($is_spam && $asap) break;
- }
- if(isset($method['area'][$key]) &&
- $progress['area'][$key] > $method['area'][$key]) {
- $is_spam = TRUE;
- $progress['_action']['area'][$key] = TRUE;
- if ($is_spam && $asap) break;
- }
+ foreach ($_remains as $key=>$value) {
+ foreach ($value as $_key=>$_value) {
+ if (is_int($_key)) {
+ $remains[$key][] = $_value;
+ } else {
+ $remains[$key][$_key] = $_value;
}
- if ($is_spam && $asap) break;
}
}
- //var_dump($method['area'], $is_spam);
+ if ($asap && $is_spam) break;
+ }
+ return $progress;
+ }
+
+ // Area: There's HTML anchor tag
+ if ((! $asap || ! $is_spam) && isset($method['area_anchor'])) {
+ $key = 'area_anchor';
+ $_asap = isset($method['asap']) ? array('asap' => TRUE) : array();
+ $result = area_pickup($target, array($key => TRUE) + $_asap);
+ if ($result) {
+ $sum[$key] = $result[$key];
+ if (isset($method[$key]) && $sum[$key] > $method[$key]) {
+ $is_spam[$key] = TRUE;
+ }
+ }
+ }
+
+ // Area: There's 'BBCode' linking tag
+ if ((! $asap || ! $is_spam) && isset($method['area_bbcode'])) {
+ $key = 'area_bbcode';
+ $_asap = isset($method['asap']) ? array('asap' => TRUE) : array();
+ $result = area_pickup($target, array($key => TRUE) + $_asap);
+ if ($result) {
+ $sum[$key] = $result[$key];
+ if (isset($method[$key]) && $sum[$key] > $method[$key]) {
+ $is_spam[$key] = TRUE;
+ }
+ }
+ }
+
+ // Return if ...
+ if ($asap && $is_spam) return $progress;
+ // URI: Pickup
+ $pickups = spam_uri_pickup($target, $method);
+ //$remains['uri_pickup'] = & $pickups;
- // URI uniqueness (and removing non-uniques)
- if ((! $is_spam || ! $asap) && isset($method['non_uniq'])) {
+ // Return if ...
+ if (empty($pickups)) return $progress;
- // Destructive normalize of URIs
- uri_array_normalize($pickups);
+ // URI: Check quantity
+ $sum['quantity'] += count($pickups);
+ // URI quantity
+ if ((! $asap || ! $is_spam) && isset($method['quantity']) &&
+ $sum['quantity'] > $method['quantity']) {
+ $is_spam['quantity'] = TRUE;
+ }
- $uris = array();
- foreach (array_keys($pickups) as $key) {
- $uris[$key] = uri_array_implode($pickups[$key]);
- }
- $count = count($uris);
- $uris = array_unique($uris);
- $progress['non_uniq'] += $count - count($uris);
- if ($progress['non_uniq'] > $method['non_uniq']) {
- $is_spam = TRUE;
- $progress['_action']['non_uniq'] = TRUE;
+ // URI: used inside HTML anchor tag pair
+ if ((! $asap || ! $is_spam) && isset($method['uri_anchor'])) {
+ $key = 'uri_anchor';
+ foreach($pickups as $pickup) {
+ if (isset($pickup['area'][$key])) {
+ $sum[$key] += $pickup['area'][$key];
+ if(isset($method[$key]) &&
+ $sum[$key] > $method[$key]) {
+ $is_spam[$key] = TRUE;
+ if ($asap && $is_spam) break;
}
- if (! $asap || ! $is_spam) {
- foreach (array_diff(array_keys($pickups),
- array_keys($uris)) as $remove) {
- unset($pickups[$remove]);
- }
+ if ($asap && $is_spam) break;
+ }
+ }
+ }
+
+ // URI: used inside 'BBCode' pair
+ if ((! $asap || ! $is_spam) && isset($method['uri_bbcode'])) {
+ $key = 'uri_bbcode';
+ foreach($pickups as $pickup) {
+ if (isset($pickup['area'][$key])) {
+ $sum[$key] += $pickup['area'][$key];
+ if(isset($method[$key]) &&
+ $sum[$key] > $method[$key]) {
+ $is_spam[$key] = TRUE;
+ if ($asap && $is_spam) break;
}
- unset($uris);
+ if ($asap && $is_spam) break;
}
- //var_dump($method['non_uniq'], $is_spam);
+ }
+ }
+
+ // URI: Uniqueness (and removing non-uniques)
+ if ((! $asap || ! $is_spam) && isset($method['non_uniquri'])) {
- // Unique host
- $hosts = array();
- foreach ($pickups as $pickup) {
- $hosts[] = & $pickup['host'];
+ // Destructive normalize of URIs
+ uri_array_normalize($pickups);
+
+ $uris = array();
+ foreach (array_keys($pickups) as $key) {
+ $uris[$key] = uri_array_implode($pickups[$key]);
+ }
+ $count = count($uris);
+ $uris = array_unique($uris);
+ $sum['non_uniquri'] += $count - count($uris);
+ if ($sum['non_uniquri'] > $method['non_uniquri']) {
+ $is_spam['non_uniquri'] = TRUE;
+ }
+ if (! $asap || ! $is_spam) {
+ foreach (array_diff(array_keys($pickups),
+ array_keys($uris)) as $remove) {
+ unset($pickups[$remove]);
}
- $hosts = array_unique($hosts);
- $progress['uniqhost'] += count($hosts);
- //var_dump($method['uniqhost'], $is_spam);
-
- // Bad host
- if ((! $is_spam || ! $asap) && isset($method['badhost'])) {
- $count = is_badhost($hosts, $asap);
- $progress['badhost'] += $count;
- if ($count !== 0) {
- $progress['_action']['badhost'] = TRUE;
- $is_spam = TRUE;
+ }
+ unset($uris);
+ }
+
+ // Return if ...
+ 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'] - $sum['uniqhost'];
+ if ($sum['non_uniqhost'] > $method['non_uniqhost']) {
+ $is_spam['non_uniqhost'] = TRUE;
+ }
+ }
+
+ // Return if ...
+ if ($asap && $is_spam) return $progress;
+
+ // URI: Bad host
+ if ((! $asap || ! $is_spam) && isset($method['badhost'])) {
+ $__remains = array();
+ if ($asap) {
+ $badhost = is_badhost($hosts, $asap, $__remains);
+ } else {
+ $badhost = is_badhost($hosts, $asap, $__remains);
+ if ($__remains) {
+ $remains['badhost'] = array();
+ foreach ($__remains as $value) {
+ $remains['badhost'][$value] = TRUE;
}
}
- //var_dump($method['badhost'], $is_spam);
+ }
+ unset($__remains);
+ if (! empty($badhost)) {
+ $sum['badhost'] += array_count_leaves($badhost);
+ foreach(array_keys($badhost) as $keys) {
+ $is_spam['badhost'][$keys] =
+ array_count_leaves($badhost[$keys]);
+ }
+ unset($badhost);
}
}
- return array($is_spam, $progress);
+ return $progress;
+}
+
+// Count leaves
+function array_count_leaves($array = array(), $count_empty_array = FALSE)
+{
+ if (! is_array($array) || (empty($array) && $count_empty_array))
+ return 1;
+
+ // Recurse
+ $result = 0;
+ foreach ($array as $part) {
+ $result += array_count_leaves($part, $count_empty_array);
+ }
+ return $result;
}
// ---------------------
// Reporting
-// TODO: Show all
+// TODO: Don't show unused $method!
// Summarize $progress (blocked only)
-function summarize_check_uri_spam_progress($progress = array(), $shownum = TRUE)
+function summarize_spam_progress($progress = array(), $blockedonly = FALSE)
{
- //$list = check_uri_spam_method();
-
- $tmp = array();
- foreach (array_keys($progress['_action']) as $_action) {
- if (is_array($progress['_action'][$_action])) {
- foreach (array_keys($progress['_action'][$_action]) as $_area) {
- $tmp[] = $_action . '=>' . $_area .
- ($shownum ? '(' . $progress[$_action][$_area] . ')' : '');
+ if ($blockedonly) {
+ $tmp = array_keys($progress['is_spam']);
+ } else {
+ $tmp = array();
+ $method = & $progress['method'];
+ if (isset($progress['sum'])) {
+ foreach ($progress['sum'] as $key => $value) {
+ if (isset($method[$key])) {
+ $tmp[] = $key . '(' . $value . ')';
+ }
}
- } else {
- $tmp[] = $_action .
- ($shownum ? '(' . $progress[$_action] . ')' : '');
}
}
// Common bahavior for blocking
// NOTE: Call this function from various blocking feature, to disgueise the reason 'why blocked'
-function spam_exit()
+function spam_exit($mode = '', $data = array())
{
- die("\n");
+ switch ($mode) {
+ case '': echo("\n"); break;
+ case 'dump':
+ echo('<pre>' . "\n");
+ echo htmlspecialchars(var_export($data, TRUE));
+ echo('</pre>' . "\n");
+ break;
+ };
+
+ // Force exit
+ exit;
}
// TODO: Record them
// Simple/fast spam filter ($target: 'a string' or an array())
-function pkwk_spamfilter($action, $page, $target = array('title' => ''), $method = array(), $asap = FALSE)
+function pkwk_spamfilter($action, $page, $target = array('title' => ''), $method = array(), $exitmode = '')
{
- global $notify;
-
- list($is_spam, $progress) = check_uri_spam($target, $method, $asap);
+ $progress = check_uri_spam($target, $method);
- if ($is_spam) {
+ if (! empty($progress['is_spam'])) {
// Mail to administrator(s)
- if ($notify) pkwk_spamnotify($action, $page, $target, $progress);
- // End
- spam_exit();
+ pkwk_spamnotify($action, $page, $target, $progress, $method);
+
+ // Exit
+ spam_exit($exitmode, $progress);
}
}
// PukiWiki original
// Mail to administrator(s)
-function pkwk_spamnotify($action, $page, $target = array('title' => ''), $progress = array())
+function pkwk_spamnotify($action, $page, $target = array('title' => ''), $progress = array(), $method = array())
{
- global $notify_subject;
-
- $footer['ACTION'] = 'Blocked by: ' . summarize_check_uri_spam_progress($progress);
- $footer['COMMENT'] = $action;
- $footer['PAGE'] = '[blocked] ' . $page;
- $footer['URI'] = get_script_uri() . '?' . rawurlencode($page);
- $footer['USER_AGENT'] = TRUE;
- $footer['REMOTE_ADDR'] = TRUE;
- pkwk_mail_notify($notify_subject, var_export($target, TRUE), $footer);
+ global $notify, $notify_subject;
+
+ if (! $notify) return;
+
+ $asap = isset($method['asap']);
+
+ $summary['ACTION'] = 'Blocked by: ' . summarize_spam_progress($progress, TRUE);
+ 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']))
+ ) .
+ ')';
+ }
+ $summary['COMMENT'] = $action;
+ $summary['PAGE'] = '[blocked] ' . (is_pagename($page) ? $page : '');
+ $summary['URI'] = get_script_uri() . '?' . rawurlencode($page);
+ $summary['USER_AGENT'] = TRUE;
+ $summary['REMOTE_ADDR'] = TRUE;
+ pkwk_mail_notify($notify_subject, var_export($target, TRUE), $summary, TRUE);
}
?>