Django overriding filter() without change existing code logic

I have a table in production that is integrated everywhere in the system, now I need to make add a new column in the table with a default value, but don't want to change all the existing logic, what is the best to do that?

class People(models.Model):
      name = models.CharField(max_length=20)
      gender = models.CharField(max_length=20)
      class = models.CharField(max_length=20)

in the system, we have this kind of queries everywhere

People.objects.filter(gender='male')

People.objects.filter(gender='female', class="3rd")
...

Now we need to add a new field:

class People(models.Model):
      name = models.CharField(max_length=20)
      gender = models.CharField(max_length=20)
      class = models.CharField(max_length=20)
      graduated = models.BooleanField(default=False)

Assume that all the existing data should have graduated is False, so all the existing should work if we can add graduated=False, but is there any way we can do so we don't need to change any of the existing code but they will assume graduated=False?

Yes, you can make a manager such that .objects will only retain People with graduated=False:

class PeopleManager(models.Manager):
    
    def get_queryset(self):
        return super().get_queryset().filter(graduated=False)

class People(models.Model):
    name = models.CharField(max_length=20)
    gender = models.CharField(max_length=20)
    class = models.CharField(max_length=20)
    graduated = models.BooleanField(default=False)
    
    objects = PeopleManager()
    all = models.Manager()

You can use People.all.all() to retrieve all People, and People.objects.all() to retrieve all People that did not graduate.

That being said, I do not recommend this: often People.objects.all() gives the impression that one will retrieve all People. As the Zen of Python says: "explicit over implicit": it is better that the code explains and hints what it is doing, not move filtering somewhere hidden in a manager.

Back to Top