On Wed, Dec 9, 2009 at 8:28 AM, Phlip <phlip2...@gmail.com> wrote: > Djangoids: > > Does the data fixture support automatically generated primary keys? > > We need to add a tiny stretch of new data to an existing website. A > fixture, such as new_data.json, would work great for both the new data > and for their new feature's unit tests. Then we would just run > loaddata on the server. > > However, the fixture format seems to require hard-coded PKs. So we > should not risk overwriting (or skipping?) any preexisting live data. > > I could code around this, either by looking up the current PKs and > then moving the new ones to miss them, or by writing a README.txt file > explaining how to add the same data thru the Admin. But this would > seem to be a systemic problem with fixtures, so I was curious if they > had a systemic solution.
At present, no. There is no systemic solution to this. Fixtures require that you define a primary key. In the past, it has been proposed that we allow fixtures to omit the primary key, which would result in the same behavior that Django has in code - a missing primary key results in an SQL INSERT and a new object instance. However, this introduces a new problem - how do you reference an object that is contained in a fixture (e.g., a foreign key reference in the fixture to an object that is also in the fixture)? As always, suggestions are welcome. Yours, Russ Magee %-) -- You received this message because you are subscribed to the Google Groups "Django users" group. To post to this group, send email to django-us...@googlegroups.com. To unsubscribe from this group, send email to django-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/django-users?hl=en.
