<?php
-// $Id: spam_pickup.php,v 1.5 2006/10/28 13:39:28 henoheno Exp $
+// $Id: spam_pickup.php,v 1.39 2007/02/17 13:43:59 henoheno Exp $
// Concept-work of spam-uri metrics
-// Copyright (C) 2006 PukiWiki Developers Team
+// Copyright (C) 2006-2007 PukiWiki Developers Team
// License: GPL v2 or (at your option) any later version
error_reporting(E_ALL); // Debug purpose
-// Return an array of normalized/parsed URIs in the $string
-// [OK] http://nasty.example.org#nasty_string
-function spam_pickup($string = '')
-{
- // Not available for user@password, IDN
- $array = array();
- preg_match_all(
- // Refer RFC3986
- '#(\b[a-z][a-z0-9.+-]{1,8})://' . // 1: Scheme
- '(' .
- // 2: 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]*))?' . // 3: Port
- '((?:/+[^\s<>"\'\[\]/]+)*/)?' . // 4: Directory path or path-info
- '([^\s<>"\'\[\]]+)?' . // 5: File and query string
- '#i',
- $string, $array, PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
- // Tests: ftp://dfshodfs:80/dfsdfs
- //var_dump(recursive_map('htmlspecialchars', $array));
- // Shrink $array
- $_port = 3;
- $_path = 4;
- foreach(array_keys($array) as $uri) {
- unset($array[$uri][0]); // Matched string itself
- $offset = $array[$uri][1][1]; // [1][1] = scheme's offset
- foreach(array_keys($array[$uri]) as $part) {
- // Remove offsets (with normalization)
- $array[$uri][$part] =
- strtolower(urldecode($array[$uri][$part][0]));
- }
-
- if (! isset($array[$uri][$_port])) $array[$uri][$_port] = '';
- if (! isset($array[$uri][$_path])) $array[$uri][$_path] = '';
- $array[$uri][$_path] = path_normalize($array[$uri][$_path]);
- $array[$uri]['offset'] = $offset;
- $array[$uri]['area'] = 0;
- }
-
- // Area elevation for '(especially external)link' intension
- if (! empty($array)) {
- // Anchor tags by preg_match_all()
- // [OK] <a href="http://nasty.example.com">visit http://nasty.example.com/</a>
- // [NG] <a href="http://ng.example.com">visit http://ng.example.com _not_ended_
- // [??] <a href= >Good site!</a> <a href= "#" >test</a>
- $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], // [0][1] = Area start (<a href>)
- $areas[$area][1][1], // [1][1] = Area end (</a>)
- );
- }
- area_measure($areas, $array);
-
- // 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/]
-
- // 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]
- $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], // [0][1] = Area start ([url])
- $areas[$area][2][1], // [4][1] = Area end ([/url])
- );
- }
- area_measure($areas, $array);
-
- // Remove 'offset's for area_measure()
- foreach(array_keys($array) as $key)
- unset($array[$key]['offset']);
- }
-
- return $array;
-}
-
-// Path normalization
-// example.org => example.org/
-// example.org#hoge -> example.org/#hoge
-// example.org/path/a/b/./c////./d -> example.org/path/a/b/c/d
-// Not available: path/../../../back
-function path_normalize($path = '')
-{
- if (! is_string($path) || $path == '') {
- $path = '/';
- } else {
- $path = trim($path);
- $last = ($path[strlen($path) - 1] == '/') ? '/' : '';
- $array = explode('/', $path);
- foreach(array_keys($array) as $key) {
- if ($array[$key] == '' || $array[$key] == '.')
- unset($array[$key]);
- }
- $path = '/' . implode('/', $array) . $last;
- }
- return $path;
-}
-
-// If in doubt, it's a little doubtful
-function area_measure($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[$a_index][1]);
- if ($offset_s < $offset && $offset < $offset_e) {
- $array[$u_index][$a_key] += $belief;
- }
- }
- }
- }
-}
-
-function show_form($string)
-{
- $base = basename(__FILE__);
- $string = htmlspecialchars($string);
- print <<< EOF
-<form action="$base" method="post">
- <textarea name="msg" rows="8" cols="80">$string</textarea><br />
- <input type="submit" name="write" value="Submit" />
-</form>
-<br/>
-EOF;
-}
-
+require('spam.php');
// Recursive array_map()
// e.g. Sanitilze ALL values (Debug purpose): var_dump(recursive_map('htmlspecialchars', $array));
return $array;
}
+function show_form($string, $asap = FALSE, $progress = TRUE, $pickup = TRUE)
+{
+ $string = htmlspecialchars($string);
+ $asap = $asap ? ' checked' : '';
+ $progress = $progress ? ' checked' : '';
+ $pickup = $pickup ? ' checked' : '';
+ $base = basename(__FILE__);
+
+ return <<< EOF
+<form action="$base" method="post">
+
+ <textarea name="msg" rows="8" cols="80">$string</textarea><br />
+
+ <input type="checkbox" name="asap" id="asap" value="on"$asap>
+ <label for="asap">asap</label><br />
+
+ <input type="checkbox" name="progress" id="progress" value="on"$progress>
+ <label for="progress">Show \$progress</label><br />
+
+ <input type="checkbox" name="pickup" id="pickup" value="on"$pickup>
+ <label for="pickup">Show pickuped URIs</label><br />
+
+ <input type="submit" name="write" value="Submit" />
+
+</form>
+EOF;
+
+}
+
// ---- Show form and result
echo basename(__FILE__) . '<br />';
-$msg = isset($_POST['msg']) ? $_POST['msg'] : '';
-show_form($msg);
+
+$msg = isset($_POST['msg']) ? $_POST['msg'] : '';
+$asap = isset($_POST['asap']) ? TRUE : FALSE;
+$prog = isset($_POST['progress']) ? TRUE : FALSE;
+$pickup = isset($_POST['pickup']) ? TRUE : FALSE;
+
+echo show_form($msg, $asap, $prog, $pickup);
+echo '<br/>';
+
echo '<pre>';
-$results = spam_pickup($msg);
-// Measure
-$count = count($results);
-$area = 0;
-foreach($results as $result)
- if (isset($result['area']))
- $area += $result['area'];
-$average = $count ? ($area / $count) : 'NULL';
+// -----------------------------------------------------
+ $spam = array();
+
+ // Threshold and rules for insertion (default)
+ $spam['method']['_default'] = array(
+ '_comment' => '_default',
+ 'quantity' => 8,
+ //'non_uniquri' => 3,
+ 'non_uniqhost' => 3,
+ 'area_anchor' => 0,
+ 'area_bbcode' => 0,
+ 'uniqhost' => TRUE,
+ 'badhost' => TRUE,
+ //'asap' => TRUE, // Stop as soon as possible (quick but less-info)
+ );
+
+ // For editing
+ // NOTE:
+ // Any thresholds may LOCK your contents by
+ // "posting one URL" many times.
+ // Any rules will lock contents that have NG things already.
+ $spam['method']['edit'] = array(
+ // Supposed_by_you(n) * Edit_form_spec(2) * Margin(1.5)
+ '_comment' => 'edit',
+ //'quantity' => 60 * 3,
+ //'non_uniquri' => 5 * 3,
+ //'non_uniqhost' => 50 * 3,
+ //'area_anchor' => 30 * 3,
+ //'area_bbcode' => 15 * 3,
+ 'uniqhost' => TRUE,
+ 'badhost' => TRUE,
+ //'asap' => TRUE,
+ );
+
+
+$method = & $spam['method']['_default'];
+//$method = & $spam['method']['edit'];
+//$method = check_uri_spam_method();
+//var_dump($method);
+// -----------------------------------------------------
+
+if ($asap) $method['asap'] = TRUE;
+
+$progress = check_uri_spam(array('a', $msg, 'b'), $method);
+
+if (! empty($progress)) {
+
+
+ if (empty($progress['is_spam'])) {
+ var_dump('ACTION: Seems not a spam');
+ } else {
+ var_dump('ACTION: Blocked by ' . summarize_spam_progress($progress, TRUE));
+
+ if (! $asap) var_dump('METRICS: ' . summarize_spam_progress($progress));
+
+ $action = 'Blocked by: ' . summarize_spam_progress($progress, TRUE);
+ if (isset($progress['is_spam']['badhost'])) {
+ $badhost = array();
+ foreach($progress['is_spam']['badhost'] as $glob=>$number) {
+ $badhost[] = $glob . '(' . $number . ')';
+ }
+ var_dump('DETAIL_BADHOST: ' . implode(', ', $badhost));
+ //var_dump($progress['is_spam']['badhost']);
+ }
+ }
+
+ if (isset($progress['remains']['badhost'])) {
+ $count = count($progress['remains']['badhost']);
+ var_dump('DETAIL_NEUTRAL_HOST: ' . $count .
+ ' (' .
+ preg_replace(
+ '/[^, a-z0-9.-]/i', '',
+ implode(', ', array_keys($progress['remains']['badhost']))
+ ) .
+ ')'
+ );
+ }
+
+ if ($prog) var_dump($progress);
+}
+
+if ($pickup) {
+ $results = spam_uri_pickup($msg);
+ $results = uri_array_normalize($results, TRUE);
+ var_dump('$results', $results);
+}
-echo "TOTAL = $count URIs, AREA_TOTAL = $area, AREA_AVERAGE = " . $average . "</br >" . "</br >";
-var_dump($results);
echo '</pre>';
?>