From da2348fbbe27facf4cfbf3fad800b3ccf98b4136 Mon Sep 17 00:00:00 2001 From: Evan Prodromou Date: Sat, 14 Feb 2009 21:55:25 -0500 Subject: [PATCH 1/3] Optionally ignore some notice sources for public page We optionally ignore some notice sources from the public page. Typically these are automatic notice sources like twitterfeed that don't usually represent the community on the site very well. --- classes/Notice.php | 4 +++- config.php.sample | 8 ++++++++ lib/common.php | 3 ++- 3 files changed, 13 insertions(+), 2 deletions(-) diff --git a/classes/Notice.php b/classes/Notice.php index 6db59c96ef..b8cd2bd7f2 100644 --- a/classes/Notice.php +++ b/classes/Notice.php @@ -136,10 +136,12 @@ class Notice extends Memcached_DataObject $notice->profile_id = $profile_id; $blacklist = common_config('public', 'blacklist'); + $autosource = common_config('public', 'autosource'); # Blacklisted are non-false, but not 1, either - if ($blacklist && in_array($profile_id, $blacklist)) { + if (($blacklist && in_array($profile_id, $blacklist)) || + ($source && $autosource && in_array($source, $autosource))) { $notice->is_local = -1; } else { $notice->is_local = $is_local; diff --git a/config.php.sample b/config.php.sample index a2c5801f45..3fa898e1be 100644 --- a/config.php.sample +++ b/config.php.sample @@ -107,6 +107,14 @@ $config['sphinx']['port'] = 3312; #$config['public']['blacklist'][] = 123; #$config['public']['blacklist'][] = 2307; +#Mark certain notice sources as automatic and thus not +#appropriate for public feed +#$config['public]['autosource'][] = 'twitterfeed'; +#$config['public]['autosource'][] = 'rssdent'; +#$config['public]['autosource'][] = 'Ping.Fm'; +#$config['public]['autosource'][] = 'HelloTxt'; +#$config['public]['autosource'][] = 'Updating.Me'; + #Do notice broadcasts offline #If you use this, you must run the six offline daemons in the #background. See the README for details. diff --git a/lib/common.php b/lib/common.php index 7bfd14c429..4fc749ca06 100644 --- a/lib/common.php +++ b/lib/common.php @@ -106,7 +106,8 @@ $config = array('server' => null), 'public' => array('localonly' => true, - 'blacklist' => array()), + 'blacklist' => array(), + 'autosource' => array()), 'theme' => array('server' => null), 'throttle' => From 1af6423738ab702b39f81746e0ff2aeeacb4fea4 Mon Sep 17 00:00:00 2001 From: Zach Copley Date: Mon, 16 Feb 2009 18:55:03 +0000 Subject: [PATCH 2/3] Whoops: forgot to include facebookqueuehandler.php with the my original Facebook queue handler patch set! --- scripts/facebookqueuehandler.php | 71 ++++++++++++++++++++++++++++++++ 1 file changed, 71 insertions(+) create mode 100755 scripts/facebookqueuehandler.php diff --git a/scripts/facebookqueuehandler.php b/scripts/facebookqueuehandler.php new file mode 100755 index 0000000000..c6859cb219 --- /dev/null +++ b/scripts/facebookqueuehandler.php @@ -0,0 +1,71 @@ +#!/usr/bin/env php +. + */ + +# Abort if called from a web server +if (isset($_SERVER) && array_key_exists('REQUEST_METHOD', $_SERVER)) { + print "This script must be run from the command line\n"; + exit(); +} + +define('INSTALLDIR', realpath(dirname(__FILE__) . '/..')); +define('LACONICA', true); + +require_once(INSTALLDIR . '/lib/common.php'); +require_once(INSTALLDIR . '/lib/facebookutil.php'); +require_once(INSTALLDIR . '/lib/queuehandler.php'); + +set_error_handler('common_error_handler'); + +class FacebookQueueHandler extends QueueHandler +{ + + function transport() + { + return 'facebook'; + } + + function start() + { + $this->log(LOG_INFO, "INITIALIZE"); + return true; + } + + function handle_notice($notice) + { + return facebookBroadcastNotice($notice); + } + + function finish() + { + } + +} + +ini_set("max_execution_time", "0"); +ini_set("max_input_time", "0"); +set_time_limit(0); + +mb_internal_encoding('UTF-8'); + +$id = ($argc > 1) ? $argv[1] : null; + +$handler = new FacebookQueueHandler($id); + +$handler->runOnce(); From cc38809764c63dbed24300cdcf68d9f667f0dfad Mon Sep 17 00:00:00 2001 From: Evan Prodromou Date: Mon, 16 Feb 2009 16:34:07 -0500 Subject: [PATCH 3/3] First version of blogspam.net plugin Added a plugin for blogspam.net. The service is kind of aggressive, so by default I turn a lot of the tests off. But worth a look. --- plugins/BlogspamNetPlugin.php | 144 ++++++++++++++++++++++++++++++++++ 1 file changed, 144 insertions(+) create mode 100644 plugins/BlogspamNetPlugin.php diff --git a/plugins/BlogspamNetPlugin.php b/plugins/BlogspamNetPlugin.php new file mode 100644 index 0000000000..d9372bcd56 --- /dev/null +++ b/plugins/BlogspamNetPlugin.php @@ -0,0 +1,144 @@ +. + * + * @category Plugin + * @package Laconica + * @author Evan Prodromou + * @copyright 2009 Control Yourself, Inc. + * @license http://www.fsf.org/licensing/licenses/agpl-3.0.html GNU Affero General Public License version 3.0 + * @link http://laconi.ca/ + */ + +if (!defined('LACONICA')) { + exit(1); +} + +define('BLOGSPAMNETPLUGIN_VERSION', '0.1'); + +/** + * Plugin to check submitted notices with blogspam.net + * + * When new notices are saved, we check their text with blogspam.net (or + * a compatible service). + * + * Blogspam.net is supposed to catch blog comment spam, and I found that + * some of its tests (min/max size, bayesian match) gave a lot of false positives. + * So, I've turned those tests off by default. This may not get as many + * hits, but it's better than nothing. + * + * @category Plugin + * @package Laconica + * @author Evan Prodromou + * @license http://www.fsf.org/licensing/licenses/agpl-3.0.html GNU Affero General Public License version 3.0 + * @link http://laconi.ca/ + * + * @see Event + */ + +class BlogspamNetPlugin extends Plugin +{ + var $baseUrl = 'http://test.blogspam.net:8888/'; + + function __construct($url=null) + { + parent::__construct(); + if ($url) { + $this->baseUrl = $url; + } + } + + function onStartNoticeSave($notice) + { + $args = $this->testArgs($notice); + common_debug("Blogspamnet args = " . print_r($args, TRUE)); + $request = xmlrpc_encode_request('testComment', array($args)); + $context = stream_context_create(array('http' => array('method' => "POST", + 'header' => + "Content-Type: text/xml\r\n". + "User-Agent: " . $this->userAgent(), + 'content' => $request))); + $file = file_get_contents($this->baseUrl, false, $context); + $response = xmlrpc_decode($file); + if (xmlrpc_is_fault($response)) { + throw new ServerException("$response[faultString] ($response[faultCode])", 500); + } else { + common_debug("Blogspamnet results = " . $response); + if (preg_match('/^ERROR(:(.*))?$/', $response, $match)) { + throw new ServerException(sprintf(_("Error from %s: %s"), $this->baseUrl, $match[2]), 500); + } else if (preg_match('/^SPAM(:(.*))?$/', $response, $match)) { + throw new ClientException(sprintf(_("Spam checker results: %s"), $match[2]), 400); + } else if (preg_match('/^OK$/', $response)) { + // don't do anything + } else { + throw new ServerException(sprintf(_("Unexpected response from %s: %s"), $this->baseUrl, $response), 500); + } + } + return true; + } + + function testArgs($notice) + { + $args = array(); + $args['comment'] = $notice->content; + $args['ip'] = $this->getClientIP(); + + if (isset($_SERVER) && array_key_exists('HTTP_USER_AGENT', $_SERVER)) { + $args['agent'] = $_SERVER['HTTP_USER_AGENT']; + } + + $profile = $notice->getProfile(); + + if ($profile && $profile->homepage) { + $args['link'] = $profile->homepage; + } + + if ($profile && $profile->fullname) { + $args['name'] = $profile->fullname; + } else { + $args['name'] = $profile->nickname; + } + + $args['site'] = common_root_url(); + $args['version'] = $this->userAgent(); + + $args['options'] = "max-size=140,min-size=0,min-words=0,exclude=bayasian"; + + return $args; + } + + function getClientIP() + { + if (isset($_SERVER) && array_key_exists('REQUEST_METHOD', $_SERVER)) { + // Note: order matters here; use proxy-forwarded stuff first + foreach (array('HTTP_X_FORWARDED_FOR', 'CLIENT-IP', 'REMOTE_ADDR') as $k) { + if (isset($_SERVER[$k])) { + return $_SERVER[$k]; + } + } + } + return '127.0.0.1'; + } + + function userAgent() + { + return 'BlogspamNetPlugin/'.BLOGSPAMNETPLUGIN_VERSION . ' Laconica/' . LACONICA_VERSION; + } +}