Premature optimization indeed is the root of all evil

src/pretix/base/models.py

@@ -1215,7 +1215,6 @@ class Quota(Versionable):
         # TODO: Test for interference with old versions of Item-Quota-relations, etc.
         # TODO: Prevent corner-cases like people having ordered an item before it got
         #       its first variationsadded
-        cache = self.event.get_cache()
         quotalookup = (
             (  # Orders for items which do not have any variations
                 Q(variation__isnull=True)
@@ -1225,13 +1224,10 @@ class Quota(Versionable):
             )
         )
 
-        paid_orders = cache.get('quota_paid_%s' % self.identity)
-        if paid_orders is None:
-            paid_orders = OrderPosition.objects.current.filter(
-                Q(order__status=Order.STATUS_PAID)
-                & quotalookup
-            ).count()
-            cache.set('quota_paid_%s' % self.identity, paid_orders)
+        paid_orders = OrderPosition.objects.current.filter(
+            Q(order__status=Order.STATUS_PAID)
+            & quotalookup
+        ).count()
 
         if paid_orders >= self.size:
             return Quota.AVAILABILITY_GONE, 0
@@ -1296,7 +1292,6 @@ class Quota(Versionable):
         )
         self.locked_here = None
         self.locked = None
-        self.event.get_cache().delete('quota_paid_%s' % self.identity)
         return updated