如何处理旧数据迁移中缺失的列？

Question

I want to add a new column to a django model. However, I have a past data migration that creates an object of this model, which now doesn't work because the column didn't exist at the time of that migration.

ie my model is:

class MyModel(models.Model):
    name = models.CharField()
    foo = models.IntegerField()  # <-- I want to add this column

however my old migration is this:

def add_my_model(apps, schema_editor):
    MyModel.objects.create(name="blah")

class Migration(migrations.Migration):
    # ...
    operations = [
        migrations.RunPython(add_my_model),
    ]

And when I try to run migrations on a fresh db, this migration fails with an error similar to this:

query = 'INSERT INTO "my_model" ("name", "foo") VALUES (?, ?)'
params = ['blah', None]

...

E       sqlite3.OperationalError: table my_model has no column named foo

What's the best way to deal with this situation?

Answer 1

This is explained in Django documentation :

def combine_names(apps, schema_editor):
    # We can't import the Person model directly as it may be a newer
    # version than this migration expects. We use the historical version.
    Person = apps.get_model('yourappname', 'Person')

basically instead of directly importing MyModel you should be doing something like

def add_my_model(apps, schema_editor):
    MyModel = apps.get_model('yourappname', 'MyModel')
    MyModel.objects.create(name="blah")