Hi all, I'm trying to write a utility method to retrieve generically related content (the particular model I'm using has the standard ContentType FK, object id and GenericFK) without ending up with one database hit per object:
http://dpaste.com/11633/ ------------------------ from django.contrib.contenttypes.models import ContentType def fetch_content_objects(item_queryset): """ Populates the given items with their ContentTypes and content objects, grouping the retrieval of content objects by model type to reduce the number of queries executed. """ items = list(item_queryset) # Group object ids by their content type ids objects_by_ctype = {} content_type_ids = set() for item in items: if not objects_by_ctype.has_key(item.content_type_id): objects_by_ctype[item.content_type_id] = [] objects_by_ctype[item.content_type_id].append(item.object_id) content_type_ids.add(item.content_type_id) # Retrieve content types and content objects in bulk content_types = ContentType._default_manager.in_bulk(list(content_type_ids)) for ctype_id, object_ids in objects_by_ctype.items(): model = content_types[ctype_id].model_class() objects_by_ctype[ctype_id] = model._default_manager.in_bulk(object_ids) # Place content types and content objects in the appropriate items for item in items: item.content_type = content_types[item.content_type_id] item.object = objects_by_ctype[item.content_type_id][item.object_id] return items So, I try to load and access the generic content for all the Items in my database at the moment (1744 of them at the moment). I can access the objects I've populated just fine without incurring any extra queries (12 queries for 1744 objects. Woo! Yay!), but as soon as I try to access their content_type attributes... http://dpaste.com/11634/ ------------------------ >>> from jellyroll.managers import fetch_content_objects >>> from jellyroll.models import Item >>> from django.db import connection >>> items = fetch_content_objects(Item.objects.all()) >>> len(connection.queries) 12 >>> len(items) 1744 >>> [item.object for item in items] ... >>> len(connection.queries) 12 >>> [item.content_type for item in items] ... >>> len(connection.queries) 1756 ...I'm back to one-query-per-instanceville. I've poked about a bit in django.db.models.fields.related.ReverseSingleRelatedObjectDescriptor, and even tried setting the '_content_type_cache' attribute it will eventually check for first in this particular case, but to no avail. Has anyone done much work on "manually" populating related data? Any ideas how I can prevent that database hit per instance happening when I access the content_type attribute? Thanks, Jonathan --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Django users" group. To post to this group, send email to django-users@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/django-users?hl=en -~----------~----~----~----~------~----~------~--~---
