[Project] Keyword Checker (for links)

KeyDog · 2010-08-31 12:58

KeyDog
Global Moderator
Offline

Registered: 2008-12-12
Posts: 1,693

Topic: [Project] Keyword Checker (for links)

I've noticed that in addition to certain URLs coming up again and again there are also of course keywords that stay the same, and IMO, are rather unique.... like ACAI, VIAGRA etc - these kind of keywords I want blocked - just like URL Checker by Grez blocks URLs.

If anyone can mod this manifest to catch keywords from a list "keywords.csv" I think it could be usefull to people trying to block links ever appearing that have those keywords. After all, without good keywords, no-one will click on the links to spam sites.... [It's important only links with the keywords are blocked - not in general of course in posts as hit ratio would give loads of false positives]

Here's the manifest for doing the same with URLs (URL Checker manifest.xml)
PS: Opinions against or for it are welcome
The list can be obtained here

[code=xml]
<install>
<![CDATA[
if (!$forum_db->table_exists('url_spam'))
{
$schema = array(
'FIELDS' => array(
'user_id' => array(
'datatype' => 'int(7)',
'allow_null' => false
),
'ip' => array(
'datatype' => 'VARCHAR(15)',
'allow_null' => false,
'default' => '\'0.0.0.0\'',
),
'url' => array(
'datatype' => 'VARCHAR(150)',
'allow_null' => true,
'default' => '\'\''
),
'time' => array(
'datatype' => 'DATETIME',
'allow_null' => false
),
'type' => array(
'datatype' => 'VARCHAR(4)',
'allow_null' => false
),
),
'PRIMARY KEY' => array('url', 'type')
);
$forum_db->create_table('url_spam', $schema);
}
]]>
</install>

<uninstall>
<![CDATA[
$forum_db->drop_table('url_spam');
]]>
</uninstall>
<hooks>
<hook id="po_end_validation"><![CDATA[
$pattern = array();
$pattern[] = '#\[url\]([^\[]*?)\[/url\]#e';
$pattern[] = '#\[url=([^\[]+?)\](.*?)\[/url\]#e';

$replace = array();
$replace[] = 'urlcheck(\'"$1"\', $errors)';
$replace[] = 'urlcheck(\'"$1"\', $errors, \'$2\')';

$message = preg_replace($pattern, $replace, $message);

if(!empty($text)) {
return ''.$text.'';
} else {
return ''.$url.'';
}
}
]]></hook>
<hook id="pf_change_details_signature_validation"><![CDATA[
$pattern = array();
$pattern[] = '#\[url\]([^\[]*?)\[/url\]#e';
$pattern[] = '#\[url=([^\[]+?)\](.*?)\[/url\]#e';

$replace = array();
$replace[] = 'urlcheck(\'$1\', $errors)';
$replace[] = 'urlcheck(\'$1\', $errors, \'$2\')';

if ($forum_config['o_make_links'] == '1')
{
if (!defined('FORUM_PARSER_LOADED')) {
require FORUM_ROOT.'include/parser.php';
}
$_POST['signature'] = do_clickable($_POST['signature']);
}

$_POST['signature'] = preg_replace($pattern, $replace, $_POST['signature']);

function urlcheck($url, &$errors, $text = "") {
global $ext_info, $forum_user, $forum_db, $forum_url;

$filename = $ext_info['path'].'/spam.csv';
$handle = fopen($filename, "r");
$content = fread($handle, filesize($filename));

if(strpos($content, $url)) {
if (file_exists($ext_info['path'].'/lang/'.$forum_user['language'].'/'.$ext_info['id'].'.php'))
include $ext_info['path'].'/lang/'.$forum_user['language'].'/'.$ext_info['id'].'.php';
else
include $ext_info['path'].'/lang/English/'.$ext_info['id'].'.php';

$url = str_replace('"', '', $url);
$query = array(
'SELECT' => 'url',
'FROM' => 'url_spam',
'WHERE' => 'url = \''.$url.'\' AND type = \'sig\''
);
$result = $forum_db->query_build($query) or error(__FILE__, __LINE__);
if($forum_db->num_rows($result) == 0) {
$query = array(
'INSERT' => 'user_id, ip, url, time, type',
'INTO' => 'url_spam',
'VALUES' => '\''.$forum_user['id'].'\', \''.get_remote_address().'\', \''.$url.'\', NOW(), \'sig\''
);
$forum_db->query_build($query) or error(__FILE__, __LINE__);
}

if($forum_user['num_posts'] < 5 && $forum_user['group_id'] != 2) {
delete_user($forum_user['id'], true);
message($lang_url_spam['User deleted']);
} else {
$errors[] = sprintf($lang_url_spam['URL disallowed'], $url, $url);
}
} else {
$url = str_replace('"', '', $url);
}

if(!empty($text)) {
return ''.$text.'';
} else {
return ''.$url.'';
}
}
]]></hook>
<hook id="ain_items_end"><![CDATA[
?>
<div class="ct-set group-item<?php echo ++$forum_page['item_count'] ?>">
<div class="ct-box">
<h3 class="ct-legend hn"><span>URL spam checker</span></h3>
<ul class="data-list">

<?php
$query = array(
'SELECT' => 'us.user_id, u.username, us.ip, us.url, DATE_FORMAT(us.time, \'%Y-%m-%d\') AS date, us.type',
'FROM' => 'url_spam AS us',
'JOINS' => array(
array(
'LEFT JOIN' => 'users AS u',
'ON' => 'u.id = us.user_id',
),
),
'ORDER BY' => 'time DESC',
);
$result = $forum_db->query_build($query) or error(__FILE__, __LINE__);
?>
<li><table>
<tr><th style="width: 100px;">User</th><th style="width: 70px;">IP</th><th style="width: 300px;">URL</th><th style="width: 70px;">Date</th><th style="width: 50px;">Type</th></tr>
<?php
while($spam = $forum_db->fetch_assoc($result)) {
echo "<tr>";
echo "<td>".((!empty($spam['username'])) ? "<a href=\"".forum_link($forum_url['user'], $spam['user_id'])."\">".forum_htmlencode($spam['username'])."</a>" : "DELETED")."</td>";
echo "<td>".$spam['ip']."</td>";
echo "<td><small>".$spam['url']."</small></td>";
echo "<td>".$spam['date']."</td>";
echo "<td>".$spam['type']."</td>";
echo "</tr>";
}
?>
</table>
<center><i>Last 25 entries. Ordered by date.</i></center></li>
</ul>
</div>
</div>
<?php
]]></hooks>
[/code]

[Project] Keyword Checker (for links)

Posts: 1

1 Topic by KeyDog 2010-08-31 12:58

Topic: [Project] Keyword Checker (for links)

Posts: 1