Added ticket http://code.djangoproject.com/ticket/6155 with patch. Still need to write tests.
Michael On Dec 7, 2007 3:15 PM, Karen Tracey <[EMAIL PROTECTED]> wrote: > > On Dec 7, 2007 2:31 PM, Empty <[EMAIL PROTECTED]> wrote: > > > > > > > > > Seems dumpdata is coded to require a Manager named 'objects' for the > Models > > > it dumps, even though (as described here: > > > http://www.djangoproject.com/documentation/model-api/#manager-names), > Django > > > does not require that a Manager named objects exists for all Models. > > > Whether this is just an oversight in dumpdata or a conscious decision I > > > can't say, though it looks like an oversight to me. > > > > You're correct Karen. This bit of code is the culprit: > > > > objects = [] > > for app in app_list: > > for model in get_models(app): > > > > objects.extend(model.objects.all()) > > > > How would you propose that it be done differently? I know that > > through object._meta.admin.manager will give you the first manager > > like the Admin works off of, but that is only populated if the Admin > > inner class is defined. I looked for the logic that does that but I > > couldn't find it. > > > > I believe it should be using _default_manager, which looks to be set to > track the first Manager defined for a Model. I think a (legitimate) Model > will always have _default_manager, while it may not always have objects. > > Karen > > > > > --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
