Add time-based cutoffs for public tag cloud, favorited lists to speed up those queries.

Defaulting to only looking at last 90 days of activity, can be adjusted up or down.

$config['tag']['cutoff'] = 86400 * 90;
$config['popular']['cutoff'] = 86400 * 90;

Per-user and per-group tag clouds do not use the cutoff (and it doesn't help with indexing on them).

4 files changed, 11 insertions, 4 deletions

diff --git a/lib/default.php b/lib/default.php
index 2bedc4bf0..485a08ba4 100644
--- a/lib/default.php
+++ b/lib/default.php
@@ -144,9 +144,11 @@ $default =
         'invite' =>
         array('enabled' => true),
         'tag' =>
-        array('dropoff' => 864000.0),
+        array('dropoff' => 864000.0,   # controls weighting based on age
+              'cutoff' => 86400 * 90), # only look at notices posted in last 90 days
         'popular' =>
-        array('dropoff' => 864000.0),
+        array('dropoff' => 864000.0,   # controls weighting based on age
+              'cutoff' => 86400 * 90), # only look at notices favorited in last 90 days
         'daemon' =>
         array('piddir' => '/var/run',
               'user' => false,
diff --git a/lib/grouptagcloudsection.php b/lib/grouptagcloudsection.php
index 14ceda085..f1106cc7b 100644
--- a/lib/grouptagcloudsection.php
+++ b/lib/grouptagcloudsection.php
@@ -59,6 +59,7 @@ class GroupTagCloudSection extends TagCloudSection
     function getTags()
     {
         $weightexpr = common_sql_weight('notice_tag.created', common_config('tag', 'dropoff'));
+        // @fixme should we use the cutoff too? Doesn't help with indexing per-group.
 
         $names = $this->group->getAliases();
 
diff --git a/lib/personaltagcloudsection.php b/lib/personaltagcloudsection.php
index 091425f92..5ea3f188d 100644
--- a/lib/personaltagcloudsection.php
+++ b/lib/personaltagcloudsection.php
@@ -59,6 +59,7 @@ class PersonalTagCloudSection extends TagCloudSection
     function getTags()
     {
         $weightexpr = common_sql_weight('notice_tag.created', common_config('tag', 'dropoff'));
+        // @fixme should we use the cutoff too? Doesn't help with indexing per-user.
 
         $qry = 'SELECT notice_tag.tag, '.
           $weightexpr . ' as weight ' .
diff --git a/lib/popularnoticesection.php b/lib/popularnoticesection.php
index fbf9a60ab..296ddbbb5 100644
--- a/lib/popularnoticesection.php
+++ b/lib/popularnoticesection.php
@@ -59,12 +59,15 @@ class PopularNoticeSection extends NoticeSection
             }
         }
         $weightexpr = common_sql_weight('fave.modified', common_config('popular', 'dropoff'));
+        $cutoff = sprintf("fave.modified > '%s'",
+                          common_sql_date(time() - common_config('popular', 'cutoff')));
         $qry = "SELECT notice.*, $weightexpr as weight ";
         if(isset($tag)) {
             $qry .= 'FROM notice_tag, notice JOIN fave ON notice.id = fave.notice_id ' .
-                    "WHERE notice.id = notice_tag.notice_id and '$tag' = notice_tag.tag";
+                    "WHERE $cutoff and notice.id = notice_tag.notice_id and '$tag' = notice_tag.tag";
         } else {
-            $qry .= 'FROM notice JOIN fave ON notice.id = fave.notice_id';
+            $qry .= 'FROM notice JOIN fave ON notice.id = fave.notice_id ' .
+                    "WHERE $cutoff";
         }
         $qry .= ' GROUP BY notice.id,notice.profile_id,notice.content,notice.uri,' .
                 'notice.rendered,notice.url,notice.created,notice.modified,' .