Let's say I have two Django models Person and Company as follows: -
class Company(models.Model):
name = models.CharField()
class Person(models.Model):
last_name = models.CharField(blank=True)
first_name = models.CharField()
company = models.ForeignKey(Company, null=True, blank=True)
A Person may or may not belong to a Company.
I am using MySQL. I want all Persons that do not belong to any Company, that is, Persons where company is null.
If I do Person.objects.filter(company__isnull=True)
I get an SQL which is essentially: -
SELECT * FROM PersonTable LEFT OUTER JOIN AgencyTable ON (PersonTable.company_id = AgencyTable.id) WHERE AgencyTable.id IS NULL
How do I go about achieving the following SQL: -
SELECT * FROM PersonTable INNER JOIN AgencyTable ON (PersonTable.company_id = AgencyTable.id) WHERE AgencyTable.id IS NULL
From what I gather from reading up the Django Users mailing list, this used to be the behavior before QuerySet Refactor.
EDIT -- Now I see the blasphemy of my question!
What I want to say is I simply want to do
SELECT * FROM PersonTable WHERE PersonTable.company_id IS NULL
(INNER) JOIN : Returns records that have matching values in both tables. LEFT (OUTER) JOIN : Returns all records from the left table, and the matched records from the right table.
There really is no difference between a LEFT JOIN and a LEFT OUTER JOIN. Both versions of the syntax will produce the exact same result in PL/SQL. Some people do recommend including outer in a LEFT JOIN clause so it's clear that you're creating an outer join, but that's entirely optional.
Difference between JOIN and INNER JOINJOIN returns all rows from tables where the key record of one table is equal to the key records of another table. The INNER JOIN selects all rows from both participating tables as long as there is a match between the columns.
A left outer join is a method of combining tables. The result includes unmatched rows from only the table that is specified before the LEFT OUTER JOIN clause. If you are joining two tables and want the result set to include unmatched rows from only one table, use a LEFT OUTER JOIN clause or a RIGHT OUTER JOIN clause.
Well, this question is old, and soon the patch will be in Django. But for the brief meantime, the answer is in http://code.djangoproject.com/ticket/10790:
Workaround: Instead of
Person.objects.filter(company=None)
use
Person.objects.exclude(company__isnull=False)
It should be as simple as:
Person.objects.filter(company_id__isnull=True)
Note the use of company_id
which is the default integer field created by the ForeignKey
Edit
Sorry, I haven't actively used django since 0.9.5. Either I'm thinking of pre-1.0 behavior, or I'm muddling up sqlalchemy and Django ORM. In either case, as the comments stated, the above appears to not work.
It looks like the only way to get the query you want in current django is to use the .extra
query parameter, which comes with a whole list of caveats.
Person.objects.extra(where=['company_id IS NULL'])
Note that this may not be portable to all DB's, and it may not work combined with filter(), and any number of possible issues. I would recommend not using this throughout your code, and instead moving it to a classmethod on Person like:
@classmethod
def list_unaffiliated_people(cls):
return cls.objects.extra(where=['company_id IS NULL'])
Alternately, just use the proper ORM query syntax and suck up the possible performance hit (have you actually benchmarked the more complicated query to see that it's any slower?)
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With