<?php
-// $Id: spam_pickup.php,v 1.3 2006/10/25 17:58:48 teanan Exp $
+// $Id: spam_pickup.php,v 1.49 2007/05/12 09:19:49 henoheno Exp $
// Concept-work of spam-uri metrics
-// Copyright (C) 2006 PukiWiki Developer Team
+// Copyright (C) 2006-2007 PukiWiki Developers Team
// License: GPL v2 or (at your option) any later version
error_reporting(E_ALL); // Debug purpose
-// TODO: Use 'm' multi-line regex option
+require('spam.php');
-// Return an array of normalized/parsed URIs in the $string
-// [OK] http://nasty.example.org#nasty_string
-function spam_pickup($string = '')
+// Recursive array_map()
+// e.g. Sanitilze ALL values (Debug purpose): var_dump(recursive_map('htmlspecialchars', $array));
+function recursive_map($func, $array)
{
- // Picup external URIs: scheme:+//+fqdn(/path)
- // Not available for IPv6 host, user@password, port
- $array = array();
- preg_match_all(
- '#(https?|\b[a-z0-9]{3,6})' . // 1:Scheme
- ':?//' . // "//" or "://"
- '([^\s<>"\'\[\]\#/]+)' . // 2:Host (FQDN or IPv4 address)
- '(?::[a-z0-9]*)?' . // Port
- '((?:/[^\s<>"\'\[\]/]+)*/)?' . // 3:Directory
- '([^\s<>"\'\[\]]+)?' . // 4:Path and Query string
- '#i', $string, $array, PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
- // Shrink $array
- $_path = 3;
- foreach(array_keys($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 (is_array($array)) {
+ if (! empty($array)) {
+ $array = array_map('recursive_map',
+ array_fill(0, count($array), $func), $array);
}
- // example.org => example.org/
- if (! isset($array[$uri][$_path])) $array[$uri][$_path] = '/';
- $array[$uri]['offset'] = $offset;
- $array[$uri]['area'] = 0;
+ } else {
+ $array = $func($array);
}
-
- // 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_
- $areas = array();
- preg_match_all('#<a\b[^>]*href[^>]*>.*?</a\b[^>]*(>)#i',
- $string, $areas, PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
- foreach(array_keys($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]
- $areas = array();
- preg_match_all('#\[(url|link)\b[^\]]*\].*?\[/\1\b[^\]]*(\])#i',
- $string, $areas, PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
- foreach(array_keys($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 $array;
}
-// If in doubt, it's a little doubtful
-function area_measure($areas, &$array, $belief = -1, $a_key = 'area', $o_key = 'offset')
+function show_form($string, $asap = FALSE, $progress = TRUE, $pickup = TRUE)
{
- if (! is_array($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;
- }
- }
- }
- }
-}
+ $string = htmlspecialchars($string);
+ $asap = $asap ? ' checked' : '';
+ $progress = $progress ? ' checked' : '';
+ $pickup = $pickup ? ' checked' : '';
+ $base = basename(__FILE__);
-function show_form($string)
-{
- $base = basename(__FILE__);
- $string = htmlspecialchars($string);
- print <<< EOF
+ 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"$pickup>
+ <label for="pickup">Show pickuped URIs</label><br />
+
<input type="submit" name="write" value="Submit" />
+
</form>
-<br/>
EOF;
+
}
// ---- Show form and result
echo basename(__FILE__) . '<br />';
-$msg = isset($_POST['msg']) ? $_POST['msg'] : '';
-show_form($msg);
-echo '<pre>';
-$results = spam_pickup($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/>';
+
+
+// -----------------------------------------------------
+ $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 http://foobaA.example.com',
+ $msg,
+ 'b http://foobarB.example.com'
+ ), $method);
+
+if (! empty($progress)) {
+
+
+ if (empty($progress['is_spam'])) {
+ echo 'ACTION: Seems not a spam';
+ echo '<br />';
+ } else {
+ echo 'ACTION: Blocked by ' . summarize_spam_progress($progress, TRUE);
+ echo '<br />';
+
+ if (! $asap) {
+ echo 'METRICS: ' . summarize_spam_progress($progress) . '<br />' . "\n";
+ }
-// Measure
-$count = count($results);
-$area = 0;
-foreach($results as $result)
- if (isset($result['area']))
- $area += $result['area'];
+ $action = 'Blocked by: ' . summarize_spam_progress($progress, TRUE);
-echo "TOTAL = $count URIs, AREA_TOTAL = $area, AREA_AVERAGE = " . ($area / $count) . "</br >" . "</br >";
-var_dump($results);
-echo '</pre>';
+ $tmp = summarize_detail_badhost($progress);
+ if ($tmp != '') {
+ echo 'DETAIL_BADHOST: ' .
+ str_replace(' ', ' ', nl2br(htmlspecialchars($tmp). "\n"));
+ }
+ }
+ $tmp = summarize_detail_newtral($progress);
+ if (! $asap && $tmp != '') {
+ echo 'DETAIL_NEUTRAL_HOST: ' .
+ str_replace(' ', ' ', nl2br(htmlspecialchars($tmp). "\n"));
+ }
+
+ if ($prog) {
+ echo '<pre>';
+ echo '$progress:' . "\n";
+ echo htmlspecialchars(var_export($progress, TRUE));
+ echo '</pre>';
+ }
+}
+if ($pickup) {
+ echo '<pre>';
+ $results = uri_pickup_normalize(spam_uri_pickup($msg));
+ echo '$results:' . "\n";
+ echo htmlspecialchars(var_export($results, TRUE));
+ echo '</pre>';
+}
?>