Note that although the
size parameter is passed to PostgreSQL, PostgreSQL will not enforce it.
ArrayFields one should keep in mind this word of warning from the Postgresql arrays documentation.
Tip: Arrays are not sets; searching for specific array elements can be a sign of database misdesign. Consider using a separate table with a row for each item that would be an array element. This will be easier to search, and is likely to scale better for a large number of elements.
To create a PostgreSQL ArrayField, we should give ArrayField the type of data we want it to store as a field as its first argument. Since we'll be storing book ratings, we will use
from django.db import models, FloatField from django.contrib.postgres.fields import ArrayField class Book(models.Model): ratings = ArrayField(FloatField())
from django.db import models, IntegerField from django.contrib.postgres.fields import ArrayField class IceCream(models.Model): scoops = ArrayField(IntegerField() # we'll use numbers to ID the scoops , size=6) # our parlor only lets you have 6 scoops
When you use the size parameter, it's passed through to postgresql, which accepts it and then ignores it! Thus it's quite possible to add 7 integers to the
scoops field above using the postgresql console.
This query returns all cones with a chocolate scoop and a vanilla scoop.
VANILLA, CHOCOLATE, MINT, STRAWBERRY = 1, 2, 3, 4 # constants for flavors choco_vanilla_cones = IceCream.objects.filter(scoops__contains=[CHOCOLATE, VANILLA])
Don't forget to import the
IceCream model from your
Also bear in mind that django will not create an index for
ArrayFields. If you are going to search them, you are going to need an index and it will need to be manually created with a call to RunSQL in your migrations file.
You can nest
ArrayFields by passing another
ArrayField as it's
from django.db import models, IntegerField from django.contrib.postgres.fields import ArrayField class SudokuBoard(models.Model): numbers = ArrayField( ArrayField( models.IntegerField(), size=9, ), size=9, )
This query returns all cones with either a mint scoop or a vanilla scoop.
minty_vanilla_cones = IceCream.objects.filter(scoops__contained_by=[MINT, VANILLA])