From 7d85b79814e7e91a88f23d7c7e752a0bcfc83ff4 Mon Sep 17 00:00:00 2001 From: Evan Prodromou Date: Mon, 3 May 2010 14:00:12 -0400 Subject: [PATCH] Database tables to cache expensive query data We need to bundle counts of notices and users by date. This can be expensive for large sites. So, new tables are added to cache the results of these queries, which don't change after the date is over. --- plugins/Sitemap/SitemapPlugin.php | 36 +++++++ plugins/Sitemap/Sitemap_notice_count.php | 125 +++++++++++++++++++++++ plugins/Sitemap/Sitemap_user_count.php | 121 ++++++++++++++++++++++ 3 files changed, 282 insertions(+) create mode 100644 plugins/Sitemap/Sitemap_notice_count.php create mode 100644 plugins/Sitemap/Sitemap_user_count.php diff --git a/plugins/Sitemap/SitemapPlugin.php b/plugins/Sitemap/SitemapPlugin.php index 6fc7021049..831694efc6 100644 --- a/plugins/Sitemap/SitemapPlugin.php +++ b/plugins/Sitemap/SitemapPlugin.php @@ -121,4 +121,40 @@ class SitemapPlugin extends Plugin 'index' => '[1-9][0-9]*')); return true; } + + /** + * Database schema setup + * + * We cache some data persistently to avoid overlong queries. + * + * @see Sitemap_user_count + * @see Sitemap_notice_count + * + * @return boolean hook value; true means continue processing, false means stop. + */ + + function onCheckSchema() + { + $schema = Schema::get(); + + // For storing user-submitted flags on profiles + + $schema->ensureTable('sitemap_user_count', + array(new ColumnDef('registration_date', 'date', null, + true, 'PRI'), + new ColumnDef('user_count', 'integer'), + new ColumnDef('created', 'datetime', + null, false), + new ColumnDef('modified', 'timestamp'))); + + $schema->ensureTable('sitemap_notice_count', + array(new ColumnDef('notice_date', 'date', null, + true, 'PRI'), + new ColumnDef('notice_count', 'integer'), + new ColumnDef('created', 'datetime', + null, false), + new ColumnDef('modified', 'timestamp'))); + + return true; + } } diff --git a/plugins/Sitemap/Sitemap_notice_count.php b/plugins/Sitemap/Sitemap_notice_count.php new file mode 100644 index 0000000000..72bb2b9d41 --- /dev/null +++ b/plugins/Sitemap/Sitemap_notice_count.php @@ -0,0 +1,125 @@ + + * @license http://www.fsf.org/licensing/licenses/agpl.html AGPLv3 + * @link http://status.net/ + * + * StatusNet - the distributed open-source microblogging tool + * Copyright (C) 2010, StatusNet, Inc. + * + * 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 . + */ + +if (!defined('STATUSNET')) { + exit(1); +} + +require_once INSTALLDIR . '/classes/Memcached_DataObject.php'; + +/** + * Data class for counting notices by date + * + * We make a separate sitemap for each notice posted by date. + * To save ourselves some (not inconsiderable) processing effort, + * we cache this data in the sitemap_notice_count table. Each + * row represents a day since the site has been started, with a count + * of notices posted on that day. Since, after the end of the day, + * this number doesn't change, it's a good candidate for persistent caching. + * + * @category Data + * @package StatusNet + * @author Evan Prodromou + * @license http://www.fsf.org/licensing/licenses/agpl.html AGPLv3 + * @link http://status.net/ + * + * @see DB_DataObject + */ + +class Sitemap_notice_count extends Memcached_DataObject +{ + public $__table = 'sitemap_notice_count'; // table name + + public $notice_date; // date primary_key not_null + public $notice_count; // int(4) + public $created; + public $modified; + + /** + * Get an instance by key + * + * This is a utility method to get a single instance with a given key value. + * + * @param string $k Key to use to lookup (usually 'notice_id' for this class) + * @param mixed $v Value to lookup + * + * @return Sitemap_notice_count object found, or null for no hits + * + */ + + function staticGet($k, $v=null) + { + return Memcached_DataObject::staticGet('Sitemap_notice_count', $k, $v); + } + + /** + * return table definition for DB_DataObject + * + * DB_DataObject needs to know something about the table to manipulate + * instances. This method provides all the DB_DataObject needs to know. + * + * @return array array of column definitions + */ + + function table() + { + return array('notice_date' => DB_DATAOBJECT_DATE + DB_DATAOBJECT_NOTNULL, + 'notice_count' => DB_DATAOBJECT_INT, + 'created' => DB_DATAOBJECT_STR + DB_DATAOBJECT_DATE + DB_DATAOBJECT_TIME + DB_DATAOBJECT_NOTNULL, + 'modified' => DB_DATAOBJECT_STR + DB_DATAOBJECT_DATE + DB_DATAOBJECT_TIME + DB_DATAOBJECT_NOTNULL); + } + + /** + * return key definitions for DB_DataObject + * + * DB_DataObject needs to know about keys that the table has; this function + * defines them. + * + * @return array key definitions + */ + + function keys() + { + return array('notice_date' => 'K'); + } + + /** + * return key definitions for Memcached_DataObject + * + * Our caching system uses the same key definitions, but uses a different + * method to get them. + * + * @return array key definitions + */ + + function keyTypes() + { + return $this->keys(); + } +} diff --git a/plugins/Sitemap/Sitemap_user_count.php b/plugins/Sitemap/Sitemap_user_count.php new file mode 100644 index 0000000000..1a7a6577d5 --- /dev/null +++ b/plugins/Sitemap/Sitemap_user_count.php @@ -0,0 +1,121 @@ + + * @license http://www.fsf.org/licensing/licenses/agpl.html AGPLv3 + * @link http://status.net/ + * + * StatusNet - the distributed open-source microblogging tool + * Copyright (C) 2010, StatusNet, Inc. + * + * 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 . + */ + +if (!defined('STATUSNET')) { + exit(1); +} + +require_once INSTALLDIR . '/classes/Memcached_DataObject.php'; + +/** + * Data class for counting users by date + * + * We make a separate sitemap for each user registered by date. + * To save ourselves some processing effort, we cache this data + * + * @category Action + * @package StatusNet + * @author Evan Prodromou + * @license http://www.fsf.org/licensing/licenses/agpl.html AGPLv3 + * @link http://status.net/ + * + * @see DB_DataObject + */ + +class Sitemap_user_count extends Memcached_DataObject +{ + public $__table = 'sitemap_user_count'; // table name + + public $registration_date; // date primary_key not_null + public $user_count; // int(4) + public $created; + public $modified; + + /** + * Get an instance by key + * + * This is a utility method to get a single instance with a given key value. + * + * @param string $k Key to use to lookup (usually 'user_id' for this class) + * @param mixed $v Value to lookup + * + * @return Sitemap_user_count object found, or null for no hits + * + */ + + function staticGet($k, $v=null) + { + return Memcached_DataObject::staticGet('Sitemap_user_count', $k, $v); + } + + /** + * return table definition for DB_DataObject + * + * DB_DataObject needs to know something about the table to manipulate + * instances. This method provides all the DB_DataObject needs to know. + * + * @return array array of column definitions + */ + + function table() + { + return array('registration_date' => DB_DATAOBJECT_DATE + DB_DATAOBJECT_NOTNULL, + 'user_count' => DB_DATAOBJECT_INT, + 'created' => DB_DATAOBJECT_STR + DB_DATAOBJECT_DATE + DB_DATAOBJECT_TIME + DB_DATAOBJECT_NOTNULL, + 'modified' => DB_DATAOBJECT_STR + DB_DATAOBJECT_DATE + DB_DATAOBJECT_TIME + DB_DATAOBJECT_NOTNULL); + } + + /** + * return key definitions for DB_DataObject + * + * DB_DataObject needs to know about keys that the table has; this function + * defines them. + * + * @return array key definitions + */ + + function keys() + { + return array('registration_date' => 'K'); + } + + /** + * return key definitions for Memcached_DataObject + * + * Our caching system uses the same key definitions, but uses a different + * method to get them. + * + * @return array key definitions + */ + + function keyTypes() + { + return $this->keys(); + } +} -- 2.39.5