sphinx integration for notices and profiles

darcs-hash:20081123185136-099f7-ca600dbdfff09762802e1cc58cf8777c45a24402.gz

diff --git a/actions/noticesearch.php b/actions/noticesearch.php
index f8dad2a20eecd39f123ae82dc08a9f58d8cfd48e..96e4d777fa3aca01c86ea3a0d39285535e014ffd 100644 (file)
--- a/actions/noticesearch.php
+++ b/actions/noticesearch.php
@@ -20,7 +20,6 @@
 if (!defined('LACONICA')) { exit(1); }
 
 require_once(INSTALLDIR.'/lib/searchaction.php');
-define('NOTICES_PER_PAGE', 20);
 
 # XXX common parent for people and content search?
 
@@ -42,13 +41,17 @@ class NoticesearchAction extends SearchAction {
                $q = strtolower($q);
 
         $search_engine = $notice->getSearchEngine('identica_notices');
-        $search_engine->query($q);
 
+        $search_engine->set_sort_mode('chron');
                # Ask for an extra to see if there's more.
                $search_engine->limit((($page-1)*NOTICES_PER_PAGE), NOTICES_PER_PAGE + 1);
 
-               $cnt = $notice->find();
-
+        if (false === $search_engine->query($q)) {
+            $cnt = 0;
+        }
+        else {
+               $cnt = $notice->find();
+        }
                if ($cnt > 0) {
                        $terms = preg_split('/[\s,]+/', $q);
                        common_element_start('ul', array('id' => 'notices'));

diff --git a/actions/noticesearchrss.php b/actions/noticesearchrss.php
index f598d833eb8fd35981616d5d69679ff201629092..0f38515a020f8145fbbd2cd19153e66c60f70f4b 100644 (file)
--- a/actions/noticesearchrss.php
+++ b/actions/noticesearchrss.php
@@ -39,19 +39,12 @@ class NoticesearchrssAction extends Rss10Action {
                # lcase it for comparison
                $q = strtolower($q);
 
-               if(common_config('db','type')=='mysql') {
-                       $notice->whereAdd('MATCH(content) against (\''.addslashes($q).'\')');
-               } else {
-                       $notice->whereAdd('to_tsvector(\'english\',content) @@ plainto_tsquery(\''.addslashes($q).'\')');
-               }
-               $notice->orderBy('created DESC, notice.id DESC');
-
-               # Ask for an extra to see if there's more.
-
-               if ($limit != 0) {
-                       $notice->limit(0, $limit);
-               }
+        $search_engine = $notice->getSearchEngine('identica_notices');
+        $search_engine->set_sort_mode('chron');
 
+               if (!$limit) $limit = 20;
+        $search_engine->limit(0, $limit, true);
+        $search_engine->query($q);
                $notice->find();
 
                while ($notice->fetch()) {

diff --git a/actions/peoplesearch.php b/actions/peoplesearch.php
index fa54dc9f7cd69f176c2a0e1fc923b7d86d156c5b..f4886fd835650085497c9941bfd40f775fb8a733 100644 (file)
--- a/actions/peoplesearch.php
+++ b/actions/peoplesearch.php
@@ -20,7 +20,7 @@
 if (!defined('LACONICA')) { exit(1); }
 
 require_once(INSTALLDIR.'/lib/searchaction.php');
-require_once(INSTALLDIR.'/lib/profilelist.php');
+define('PROFILES_PER_PAGE', 10);
 
 class PeoplesearchAction extends SearchAction {
 
@@ -42,13 +42,15 @@ class PeoplesearchAction extends SearchAction {
 
         $search_engine = $profile->getSearchEngine('identica_people');
 
-        $search_engine->query($q);
-
+        $search_engine->set_sort_mode('chron');
                # Ask for an extra to see if there's more.
         $search_engine->limit((($page-1)*PROFILES_PER_PAGE), PROFILES_PER_PAGE + 1);
-
-               $cnt = $profile->find();
-
+        if (false === $search_engine->query($q)) {
+            $cnt = 0;
+        }
+        else {
+                   $cnt = $profile->find();
+        }
                if ($cnt > 0) {
                        $terms = preg_split('/[\s,]+/', $q);
                        $results = new PeopleSearchResults($profile, $terms);

diff --git a/classes/SearchEngines.php b/classes/SearchEngines.php
index 253e0028be9d91d22ec21e44dfb2347cb06a6cfa..96fd7da90da95376a63b0124798354e0d5c1b35a 100644 (file)
--- a/classes/SearchEngines.php
+++ b/classes/SearchEngines.php
@@ -20,54 +20,85 @@
 if (!defined('LACONICA')) { exit(1); }
 
 class SearchEngine {
-    protected $profile;
+    protected $target;
     protected $table;
 
-    function __construct($profile, $table) {
-        $this->profile = $profile;
+    function __construct($target, $table) {
+        $this->target = $target;
         $this->table = $table;
     }
 
     function query($q) {
     }
 
-    function limit($offset, $count) {
-        return $this->profile->limit($offset, $count);
+    function limit($offset, $count, $rss = false) {
+        return $this->target->limit($offset, $count);
+    }
+
+    function set_sort_mode($mode) {
+        if ('chron' === $mode)
+            return $this->target->orderBy('created desc');
     }
 }
 
 class SphinxSearch extends SearchEngine {
     private $sphinx;
 
-    function __construct($profile, $table) {
-        parent::__construct($profile, $table);
+    function __construct($target, $table) {
+        parent::__construct($target, $table);
         $this->sphinx = new SphinxClient;
         $this->sphinx->setServer(common_config('sphinx', 'server'), common_config('sphinx', 'port'));
     }
 
-    function limit($offset, $count) {
-        $this->sphinx->setLimits($offset, $count);
-        $this->profile->limit($offset, $count);
+    function limit($offset, $count, $rss = false) {
+        //FIXME without LARGEST_POSSIBLE, the most recent results aren't returned
+        //      this probably has a large impact on performance
+        $LARGEST_POSSIBLE = 1e6; 
+
+        if ($rss) {
+            $this->sphinx->setLimits($offset, $count, $count, $LARGEST_POSSIBLE);
+        }
+        else {
+            // return at most 50 pages of results
+            $this->sphinx->setLimits($offset, $count, 50 * ($count - 1), $LARGEST_POSSIBLE);
+        }
+
+        return $this->target->limit(0, $count);
     }
 
     function query($q) {
         $result = $this->sphinx->query($q, $this->table);
         if (!isset($result['matches'])) return false;
         $id_set = join(', ', array_keys($result['matches']));
-        return $this->profile->whereAdd("id in ($id_set)");
+        $this->target->whereAdd("id in ($id_set)");
+        return true;
      }
+
+    function set_sort_mode($mode) {
+        if ('chron' === $mode) {
+            $this->sphinx->SetSortMode(SPH_SORT_ATTR_DESC, 'created_ts');
+            return $this->target->orderBy('created desc');
+        }
+    }
 }
 
 class MySQLSearch extends SearchEngine {
     function query($q) {
-        return $this->profile->whereAdd('MATCH(nickname, fullname, location, bio, homepage) ' .
+        if ('identica_people' === $this->table)
+            return $this->target->whereAdd('MATCH(nickname, fullname, location, bio, homepage) ' .
+                                                  'against (\''.addslashes($q).'\')');
+        if ('identica_notices' === $this->table)
+            return $this->target->whereAdd('MATCH(content) ' .
                                                   'against (\''.addslashes($q).'\')');
     }
 }
 
 class PGSearch extends SearchEngine {
     function query($q) {
-        $this->profile->whereAdd('textsearch @@ plainto_tsquery(\''.addslashes($q).'\')');
+        if ('identica_people' === $this->table)
+            return $this->target->whereAdd('textsearch @@ plainto_tsquery(\''.addslashes($q).'\')');
+        if ('identica_notices' === $this->table)
+            return $this->target->whereAdd('to_tsvector(\'english\', content) @@ plainto_tsquery(\''.addslashes($q).'\')');
     }
 }
 

diff --git a/lib/common.php b/lib/common.php
index acea1252cce50b4f9b4911c913561b167d19fa93..9b4751c9ff14b13bda460fec786dfdec815d6ced 100644 (file)
--- a/lib/common.php
+++ b/lib/common.php
@@ -27,6 +27,7 @@ define('AVATAR_MINI_SIZE', 24);
 define('MAX_AVATAR_SIZE', 256 * 1024);
 
 define('NOTICES_PER_PAGE', 20);
+define('PROFILES_PER_PAGE', 10);
 
 define('FOREIGN_NOTICE_SEND', 1);
 define('FOREIGN_NOTICE_RECV', 2);