Ticket 2141: bugs with weighted popularity lists across year boundary.

Consolidated several separate implementations of the same weighting algorithm into common_sql_weight() and fixed some bugs...
For MySQL, now using timestampdiff() instead of subtraction for the comparison, so we get sane results when the year doesn't match, and utc_timestamp() rather than now() so we don't get negative ages for recent items with local server timezone.
Unknown whether the same problems affect PostgreSQL, but note that it lacks the timestampdiff() SQL function.

4 files changed, 27 insertions, 17 deletions

diff --git a/lib/grouptagcloudsection.php b/lib/grouptagcloudsection.php
index 091cf4845..14ceda085 100644
--- a/lib/grouptagcloudsection.php
+++ b/lib/grouptagcloudsection.php
@@ -58,11 +58,7 @@ class GroupTagCloudSection extends TagCloudSection
 
     function getTags()
     {
-        if (common_config('db', 'type') == 'pgsql') {
-            $weightexpr='sum(exp(-extract(epoch from (now() - notice_tag.created)) / %s))';
-        } else {
-            $weightexpr='sum(exp(-(now() - notice_tag.created) / %s))';
-        }
+        $weightexpr = common_sql_weight('notice_tag.created', common_config('tag', 'dropoff'));
 
         $names = $this->group->getAliases();
 
@@ -99,7 +95,6 @@ class GroupTagCloudSection extends TagCloudSection
 
         $tag = Memcached_DataObject::cachedQuery('Notice_tag',
                                                  sprintf($qry,
-                                                         common_config('tag', 'dropoff'),
                                                          $this->group->id,
                                                          $namestring),
                                                  3600);
diff --git a/lib/personaltagcloudsection.php b/lib/personaltagcloudsection.php
index 0b29d58ca..091425f92 100644
--- a/lib/personaltagcloudsection.php
+++ b/lib/personaltagcloudsection.php
@@ -58,13 +58,9 @@ class PersonalTagCloudSection extends TagCloudSection
 
     function getTags()
     {
-        if (common_config('db', 'type') == 'pgsql') {
-            $weightexpr='sum(exp(-extract(epoch from (now() - notice_tag.created)) / %s))';
-        } else {
-            $weightexpr='sum(exp(-(now() - notice_tag.created) / %s))';
-        }
- 
-       $qry = 'SELECT notice_tag.tag, '.
+        $weightexpr = common_sql_weight('notice_tag.created', common_config('tag', 'dropoff'));
+
+        $qry = 'SELECT notice_tag.tag, '.
           $weightexpr . ' as weight ' .
           'FROM notice_tag JOIN notice ' .
           'ON notice_tag.notice_id = notice.id ' .
@@ -83,7 +79,6 @@ class PersonalTagCloudSection extends TagCloudSection
 
         $tag = Memcached_DataObject::cachedQuery('Notice_tag',
                                                  sprintf($qry,
-                                                         common_config('tag', 'dropoff'),
                                                          $this->user->id),
                                                  3600);
         return $tag;
diff --git a/lib/popularnoticesection.php b/lib/popularnoticesection.php
index 9fbc9d2dd..fbf9a60ab 100644
--- a/lib/popularnoticesection.php
+++ b/lib/popularnoticesection.php
@@ -48,17 +48,17 @@ class PopularNoticeSection extends NoticeSection
 {
     function getNotices()
     {
+        // @fixme there should be a common func for this
         if (common_config('db', 'type') == 'pgsql') {
-            $weightexpr='sum(exp(-extract(epoch from (now() - fave.modified)) / %s))';
             if (!empty($this->out->tag)) {
                 $tag = pg_escape_string($this->out->tag);
             }
         } else {
-            $weightexpr='sum(exp(-(now() - fave.modified) / %s))';
             if (!empty($this->out->tag)) {
                  $tag = mysql_escape_string($this->out->tag);
             }
         }
+        $weightexpr = common_sql_weight('fave.modified', common_config('popular', 'dropoff'));
         $qry = "SELECT notice.*, $weightexpr as weight ";
         if(isset($tag)) {
             $qry .= 'FROM notice_tag, notice JOIN fave ON notice.id = fave.notice_id ' .
@@ -78,7 +78,7 @@ class PopularNoticeSection extends NoticeSection
         $qry .= ' LIMIT ' . $limit . ' OFFSET ' . $offset;
 
         $notice = Memcached_DataObject::cachedQuery('Notice',
-                                                    sprintf($qry, common_config('popular', 'dropoff')),
+                                                    $qry,
                                                     1200);
         return $notice;
     }
diff --git a/lib/util.php b/lib/util.php
index 63656b604..50bd0e2ac 100644
--- a/lib/util.php
+++ b/lib/util.php
@@ -908,6 +908,26 @@ function common_sql_date($datetime)
     return strftime('%Y-%m-%d %H:%M:%S', $datetime);
 }
 
+/**
+ * Return an SQL fragment to calculate an age-based weight from a given
+ * timestamp or datetime column.
+ *
+ * @param string $column name of field we're comparing against current time
+ * @param integer $dropoff divisor for age in seconds before exponentiation
+ * @return string SQL fragment
+ */
+function common_sql_weight($column, $dropoff)
+{
+    if (common_config('db', 'type') == 'pgsql') {
+        // PostgreSQL doesn't support timestampdiff function.
+        // @fixme will this use the right time zone?
+        // @fixme does this handle cross-year subtraction correctly?
+        return "sum(exp(-extract(epoch from (now() - $column)) / $dropoff))";
+    } else {
+        return "sum(exp(timestampdiff(second, utc_timestamp(), $column) / $dropoff))";
+    }
+}
+
 function common_redirect($url, $code=307)
 {
     static $status = array(301 => "Moved Permanently",