Remove unused contacts

src/Worker/Cron.php

@@ -117,6 +117,8 @@ class Cron
 
 			Worker::add(PRIORITY_LOW, 'CleanItemUri');
 
+			Worker::add(PRIORITY_LOW, 'RemoveUnusedContacts');
+
 			// check upstream version?
 			Worker::add(PRIORITY_LOW, 'CheckVersion');
 

src/Worker/RemoveUnusedContacts.php

@@ -0,0 +1,59 @@
+<?php
+/**
+ * @copyright Copyright (C) 2020, Friendica
+ *
+ * @license GNU AGPL version 3 or any later version
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as
+ * published by the Free Software Foundation, either version 3 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program.  If not, see <https://www.gnu.org/licenses/>.
+ *
+ */
+
+namespace Friendica\Worker;
+
+use Friendica\Core\Logger;
+use Friendica\Core\Protocol;
+use Friendica\Database\DBA;
+use Friendica\Model\Photo;
+
+/**
+ * Removes public contacts that aren't in use
+ */
+class RemoveUnusedContacts {
+	public static function execute() {
+		$condition = ["`uid` = ? AND NOT `self` AND NOT `nurl` IN (SELECT `nurl` FROM `contact` WHERE `uid` != ?)
+			AND (NOT `network` IN (?, ?, ?, ?, ?, ?) OR (`archive` AND `success_update` < UTC_TIMESTAMP() - INTERVAL ? DAY))
+			AND NOT `id` IN (SELECT `author-id` FROM `item`) AND NOT `id` IN (SELECT `owner-id` FROM `item`)
+			AND NOT `id` IN (SELECT `causer-id` FROM `item`) AND NOT `id` IN (SELECT `cid` FROM `post-tag`)
+			AND NOT `id` IN (SELECT `contact-id` FROM `item`) AND NOT `id` IN (SELECT `author-id` FROM `thread`)
+			AND NOT `id` IN (SELECT `owner-id` FROM `thread`) AND NOT `id` IN (SELECT `contact-id` FROM `thread`)
+			AND NOT `id` IN (SELECT `contact-id` FROM `post-user`) AND NOT `id` IN (SELECT `cid` FROM `user-contact`) 
+			AND NOT `id` IN (SELECT `cid` FROM `event`) AND NOT `id` IN (SELECT `contact-id` FROM `group_member`)",
+			0, 0, Protocol::DFRN, Protocol::DIASPORA, Protocol::OSTATUS, Protocol::FEED, Protocol::MAIL, Protocol::ACTIVITYPUB, 365];
+
+		$total = DBA::count('contact', $condition);
+		Logger::notice('Starting removal', ['total' => $total]);
+		$count = 0;
+		$contacts = DBA::select('contact', ['id', 'uid'], $condition);
+		while ($contact = DBA::fetch($contacts)) {
+			if (Photo::delete(['uid' => $contact['uid'], 'contact-id' => $contact['id']])) {
+				DBA::delete('contact', ['id' => $contact['id']]);
+				if ((++$count % 1000) == 0) {
+					Logger::notice('In removal', ['count' => $count, 'total' => $total]);
+				}
+			}
+		}
+		DBA::close($contacts);
+		Logger::notice('Removal done', ['count' => $count, 'total' => $total]);
+	}
+}

src/Worker/UpdateContacts.php

@@ -74,8 +74,9 @@ class UpdateContacts
 
 		$count = 0;
 		foreach ($ids as $id) {
-			Worker::add(PRIORITY_LOW, "UpdateContact", $id);
-			++$count;
+			if (Worker::add(PRIORITY_LOW, "UpdateContact", $id)) {
+				++$count;
+			}
 		}
 
 		Logger::info('Initiated update for federated contacts', ['count' => $count]);

src/Worker/UpdateGServers.php

@@ -62,12 +62,15 @@ class UpdateGServers
 			// There are duplicated "url" but not "nurl". So we check both addresses instead of just overwriting them,
 			// since that would mean loosing data.
 			if (!empty($gserver['url'])) {
-				Worker::add(PRIORITY_LOW, 'UpdateGServer', $gserver['url']);
+				if (Worker::add(PRIORITY_LOW, 'UpdateGServer', $gserver['url'])) {
+					$count++;
+				}
 			}
 			if (!empty($gserver['nurl']) && ($gserver['nurl'] != Strings::normaliseLink($gserver['url']))) {
-				Worker::add(PRIORITY_LOW, 'UpdateGServer', $gserver['nurl']);
+				if (Worker::add(PRIORITY_LOW, 'UpdateGServer', $gserver['nurl'])) {
+					$count++;
+				}
 			}
-			$count++;
 		}
 		DBA::close($gservers);
 		Logger::info('Updated servers', ['count' => $count]);