Usually, when I started working on the django + south project, which was developed for some time, I did not bother to apply all hundreds of migrations just to get to the current state of the database. I went straight to syncdb --allto create a database from the current state of the model, and then ran migrate --faketo mark all migrations as completed.
After a few seconds, I had a database that was synchronized with the migrations. Syncdbloaded the initial lights. In 99% of cases, this was all I needed. I was very happy.
How can I do it now? Syncdbis outdated, and the team migratecannot ignore migrations and simply create a database from the current state of the model classes.
The Migrate team makes me go through many migrations from many months of project development. These migrations have some strange errors that don't even bother me. It is also extremely slow, as if something had messed up the implementation of the migration mechanism.
I just want to be able to say: "Hey, Django, this is a model, create me a database from the model, then simulate all the migrations and let's work from there."
I know that I can remove all migrations locally, create initials for all applications and apply them, but this looks like a very strange workaround for the equivalent syncdb --all.
Did I miss something?