Optimised use of 'in' operator on QuerySet using an explicit __contains__ method.

Without this change, use of 'in' on a QuerySet resulted in ._result_cache
being fully populated, which sometimes is unnecessary work.



git-svn-id: http://code.djangoproject.com/svn/django/trunk@11803 bcc190cf-cafb-0310-a4f2-bffc1f526a37

tests/modeltests/basic/models.py

@@ -211,6 +211,14 @@ True
 >>> Article.objects.get(id__exact=8) == Article.objects.get(id__exact=7)
 False
 
+# You can use 'in' to test for membership...
+>>> a8 in Article.objects.all()
+True
+
+# ... but there will often be more efficient ways if that is all you need:
+>>> Article.objects.filter(id=a8.id).exists()
+True
+
 # dates() returns a list of available dates of the given scope for the given field.
 >>> Article.objects.dates('pub_date', 'year')
 [datetime.datetime(2005, 1, 1, 0, 0)]