All of these fields are available from the django.contrib.postgres.fields\nmodule.
A field for storing lists of data. Most field types can be used, you simply\npass another field instance as the base_field. You may also specify a size. ArrayField can be nested to store multi-dimensional\narrays.
If you give the field a default, ensure\nit’s a callable such as list (for an empty default) or a callable that\nreturns a list (such as a function). Incorrectly using default=[]\ncreates a mutable default that is shared between all instances of\nArrayField.
This is a required argument.
\nSpecifies the underlying data type and behavior for the array. It\nshould be an instance of a subclass of\nField. For example, it could be an\nIntegerField or a\nCharField. Most field types are permitted,\nwith the exception of those handling relational data\n(ForeignKey,\nOneToOneField and\nManyToManyField).
It is possible to nest array fields - you can specify an instance of\nArrayField as the base_field. For example:
from django.db import models\nfrom django.contrib.postgres.fields import ArrayField\n\nclass ChessBoard(models.Model):\n board = ArrayField(\n ArrayField(\n models.CharField(max_length=10, blank=True),\n size=8,\n ),\n size=8,\n )\nTransformation of values between the database and the model, validation\nof data and configuration, and serialization are all delegated to the\nunderlying base field.
\nThis is an optional argument.
\nIf passed, the array will have a maximum size as specified. This will\nbe passed to the database, although PostgreSQL at present does not\nenforce the restriction.
\nThere are a number of custom lookups and transforms for ArrayField.\nWe will use the following example model:
from django.db import models\nfrom django.contrib.postgres.fields import ArrayField\n\nclass Post(models.Model):\n name = models.CharField(max_length=200)\n tags = ArrayField(models.CharField(max_length=200), blank=True)\n\n def __str__(self): # __unicode__ on Python 2\n return self.name\nThe contains lookup is overridden on ArrayField. The\nreturned objects will be those where the values passed are a subset of the\ndata. It uses the SQL operator @>. For example:
>>> Post.objects.create(name='First post', tags=['thoughts', 'django'])\n>>> Post.objects.create(name='Second post', tags=['thoughts'])\n>>> Post.objects.create(name='Third post', tags=['tutorial', 'django'])\n\n>>> Post.objects.filter(tags__contains=['thoughts'])\n[<Post: First post>, <Post: Second post>]\n\n>>> Post.objects.filter(tags__contains=['django'])\n[<Post: First post>, <Post: Third post>]\n\n>>> Post.objects.filter(tags__contains=['django', 'thoughts'])\n[<Post: First post>]\nThis is the inverse of the contains lookup -\nthe objects returned will be those where the data is a subset of the values\npassed. It uses the SQL operator <@. For example:
>>> Post.objects.create(name='First post', tags=['thoughts', 'django'])\n>>> Post.objects.create(name='Second post', tags=['thoughts'])\n>>> Post.objects.create(name='Third post', tags=['tutorial', 'django'])\n\n>>> Post.objects.filter(tags__contained_by=['thoughts', 'django'])\n[<Post: First post>, <Post: Second post>]\n\n>>> Post.objects.filter(tags__contained_by=['thoughts', 'django', 'tutorial'])\n[<Post: First post>, <Post: Second post>, <Post: Third post>]\nReturns objects where the data shares any results with the values passed. Uses\nthe SQL operator &&. For example:
>>> Post.objects.create(name='First post', tags=['thoughts', 'django'])\n>>> Post.objects.create(name='Second post', tags=['thoughts'])\n>>> Post.objects.create(name='Third post', tags=['tutorial', 'django'])\n\n>>> Post.objects.filter(tags__overlap=['thoughts'])\n[<Post: First post>, <Post: Second post>]\n\n>>> Post.objects.filter(tags__overlap=['thoughts', 'tutorial'])\n[<Post: First post>, <Post: Second post>, <Post: Third post>]\nReturns the length of the array. The lookups available afterwards are those\navailable for IntegerField. For example:
>>> Post.objects.create(name='First post', tags=['thoughts', 'django'])\n>>> Post.objects.create(name='Second post', tags=['thoughts'])\n\n>>> Post.objects.filter(tags__len=1)\n[<Post: Second post>]\nThis class of transforms allows you to index into the array in queries. Any\nnon-negative integer can be used. There are no errors if it exceeds the\nsize of the array. The lookups available after the\ntransform are those from the base_field. For\nexample:
>>> Post.objects.create(name='First post', tags=['thoughts', 'django'])\n>>> Post.objects.create(name='Second post', tags=['thoughts'])\n\n>>> Post.objects.filter(tags__0='thoughts')\n[<Post: First post>, <Post: Second post>]\n\n>>> Post.objects.filter(tags__1__iexact='Django')\n[<Post: First post>]\n\n>>> Post.objects.filter(tags__276='javascript')\n[]\nThis class of transforms allow you to take a slice of the array. Any two\nnon-negative integers can be used, separated by a single underscore. The\nlookups available after the transform do not change. For example:
\n>>> Post.objects.create(name='First post', tags=['thoughts', 'django'])\n>>> Post.objects.create(name='Second post', tags=['thoughts'])\n>>> Post.objects.create(name='Third post', tags=['django', 'python', 'thoughts'])\n\n>>> Post.objects.filter(tags__0_1=['thoughts'])\n[<Post: First post>, <Post: Second post>]\n\n>>> Post.objects.filter(tags__0_2__contains=['thoughts'])\n[<Post: First post>, <Post: Second post>]\nAt present using db_index will create a\nbtree index. This does not offer particularly significant help to querying.\nA more useful index is a GIN index, which you should create using a\nRunSQL operation.
A field for storing mappings of strings to strings. The Python data type\nused is a dict.
To use this field, you’ll need to:
\nAdd 'django.contrib.postgres' in your INSTALLED_APPS.
Setup the hstore extension in PostgreSQL before the first CreateModel\nor AddField operation by adding a migration with the\nHStoreExtension operation.\nFor example:
from django.contrib.postgres.operations import HStoreExtension\n\nclass Migration(migrations.Migration):\n ...\n\n operations = [\n HStoreExtension(),\n ...\n ]\nCreating the extension requires a database user with superuser\nprivileges. If the Django database user doesn’t have superuser\nprivileges, you’ll have to create the extension outside of Django\nmigrations with a user that has the appropriate privileges. In that\ncase, connect to your Django database and run the query\nCREATE EXTENSION IF NOT EXISTS hstore;
You’ll see an error like can't adapt type 'dict' if you skip the first\nstep, or type "hstore" does not exist if you skip the second.
In addition to the ability to query by key, there are a number of custom\nlookups available for HStoreField.
We will use the following example model:
\nfrom django.contrib.postgres.fields import HStoreField\nfrom django.db import models\n\nclass Dog(models.Model):\n name = models.CharField(max_length=200)\n data = HStoreField()\n\n def __str__(self): # __unicode__ on Python 2\n return self.name\nTo query based on a given key, you simply use that key as the lookup name:
\n>>> Dog.objects.create(name='Rufus', data={'breed': 'labrador'})\n>>> Dog.objects.create(name='Meg', data={'breed': 'collie'})\n\n>>> Dog.objects.filter(data__breed='collie')\n[<Dog: Meg>]\nYou can chain other lookups after key lookups:
\n>>> Dog.objects.filter(data__breed__contains='l')\n[<Dog: Rufus>, <Dog: Meg>]\nIf the key you wish to query by clashes with the name of another lookup, you\nneed to use the hstorefield.contains lookup instead.
The contains lookup is overridden on\nHStoreField. The returned objects are\nthose where the given dict of key-value pairs are all contained in the\nfield. It uses the SQL operator @>. For example:
>>> Dog.objects.create(name='Rufus', data={'breed': 'labrador', 'owner': 'Bob'})\n>>> Dog.objects.create(name='Meg', data={'breed': 'collie', 'owner': 'Bob'})\n>>> Dog.objects.create(name='Fred', data={})\n\n>>> Dog.objects.filter(data__contains={'owner': 'Bob'})\n[<Dog: Rufus>, <Dog: Meg>]\n\n>>> Dog.objects.filter(data__contains={'breed': 'collie'})\n[<Dog: Meg>]\nThis is the inverse of the contains lookup -\nthe objects returned will be those where the key-value pairs on the object are\na subset of those in the value passed. It uses the SQL operator <@. For\nexample:
>>> Dog.objects.create(name='Rufus', data={'breed': 'labrador', 'owner': 'Bob'})\n>>> Dog.objects.create(name='Meg', data={'breed': 'collie', 'owner': 'Bob'})\n>>> Dog.objects.create(name='Fred', data={})\n\n>>> Dog.objects.filter(data__contained_by={'breed': 'collie', 'owner': 'Bob'})\n[<Dog: Meg>, <Dog: Fred>]\n\n>>> Dog.objects.filter(data__contained_by={'breed': 'collie'})\n[<Dog: Fred>]\nReturns objects where the given key is in the data. Uses the SQL operator\n?. For example:
>>> Dog.objects.create(name='Rufus', data={'breed': 'labrador'})\n>>> Dog.objects.create(name='Meg', data={'breed': 'collie', 'owner': 'Bob'})\n\n>>> Dog.objects.filter(data__has_key='owner')\n[<Dog: Meg>]\nReturns objects where all of the given keys are in the data. Uses the SQL operator\n?&. For example:
>>> Dog.objects.create(name='Rufus', data={})\n>>> Dog.objects.create(name='Meg', data={'breed': 'collie', 'owner': 'Bob'})\n\n>>> Dog.objects.filter(data__has_keys=['breed', 'owner'])\n[<Dog: Meg>]\nReturns objects where the array of keys is the given value. Note that the order\nis not guaranteed to be reliable, so this transform is mainly useful for using\nin conjunction with lookups on\nArrayField. Uses the SQL function\nakeys(). For example:
>>> Dog.objects.create(name='Rufus', data={'toy': 'bone'})\n>>> Dog.objects.create(name='Meg', data={'breed': 'collie', 'owner': 'Bob'})\n\n>>> Dog.objects.filter(data__keys__overlap=['breed', 'toy'])\n[<Dog: Rufus>, <Dog: Meg>]\nReturns objects where the array of values is the given value. Note that the\norder is not guaranteed to be reliable, so this transform is mainly useful for\nusing in conjunction with lookups on\nArrayField. Uses the SQL function\navalues(). For example:
>>> Dog.objects.create(name='Rufus', data={'breed': 'labrador'})\n>>> Dog.objects.create(name='Meg', data={'breed': 'collie', 'owner': 'Bob'})\n\n>>> Dog.objects.filter(data__values__contains=['collie'])\n[<Dog: Meg>]\nThere are five range field types, corresponding to the built-in range types in\nPostgreSQL. These fields are used to store a range of values; for example the\nstart and end timestamps of an event, or the range of ages an activity is\nsuitable for.
\nAll of the range fields translate to psycopg2 Range objects in python, but also accept tuples as input if no bounds\ninformation is necessary. The default is lower bound included, upper bound\nexcluded.
\nStores a range of integers. Based on an\nIntegerField. Represented by an int4range in\nthe database and a NumericRange in\nPython.
Stores a range of large integers. Based on a\nBigIntegerField. Represented by an int8range\nin the database and a NumericRange in\nPython.
Stores a range of floating point values. Based on a\nFloatField. Represented by a numrange in the\ndatabase and a NumericRange in Python.
Stores a range of timestamps. Based on a\nDateTimeField. Represented by a tztsrange in\nthe database and a DateTimeTZRange in\nPython.
Stores a range of dates. Based on a\nDateField. Represented by a daterange in the\ndatabase and a DateRange in Python.
There are a number of custom lookups and transforms for range fields. They are\navailable on all the above fields, but we will use the following example\nmodel:
\nfrom django.contrib.postgres.fields import IntegerRangeField\nfrom django.db import models\n\nclass Event(models.Model):\n name = models.CharField(max_length=200)\n ages = IntegerRangeField()\n\n def __str__(self): # __unicode__ on Python 2\n return self.name\nWe will also use the following example objects:
\n>>> Event.objects.create(name='Soft play', ages=(0, 10))\n>>> Event.objects.create(name='Pub trip', ages=(21, None))\nand NumericRange:
>>> from psycopg2.extras import NumericRange\nAs with other PostgreSQL fields, there are three standard containment\noperators: contains, contained_by and overlap, using the SQL\noperators @>, <@, and && respectively.
>>> Event.objects.filter(ages__contains=NumericRange(4, 5))\n[<Event: Soft play>]\n>>> Event.objects.filter(ages__contained_by=NumericRange(0, 15))\n[<Event: Soft play>]\n>>> Event.objects.filter(ages__overlap=NumericRange(8, 12))\n[<Event: Soft play>]\nRange fields support the standard lookups: lt, gt,\nlte and gte. These are not particularly helpful - they\ncompare the lower bounds first and then the upper bounds only if necessary.\nThis is also the strategy used to order by a range field. It is better to use\nthe specific range comparison operators.
The returned ranges are strictly less than the passed range. In other words,\nall the points in the returned range are less than all those in the passed\nrange.
\n>>> Event.objects.filter(ages__fully_lt=NumericRange(11, 15))\n[<Event: Soft play>]\nThe returned ranges are strictly greater than the passed range. In other words,\nthe all the points in the returned range are greater than all those in the\npassed range.
\n>>> Event.objects.filter(ages__fully_gt=NumericRange(11, 15))\n[<Event: Pub trip>]\nThe returned ranges do not contain any points less than the passed range, that\nis the lower bound of the returned range is at least the lower bound of the\npassed range.
\n>>> Event.objects.filter(ages__not_lt=NumericRange(0, 15))\n[<Event: Soft play>, <Event: Pub trip>]\nThe returned ranges do not contain any points greater than the passed range, that\nis the upper bound of the returned range is at most the upper bound of the\npassed range.
\n>>> Event.objects.filter(ages__not_gt=NumericRange(3, 10))\n[<Event: Soft play>]\nThe returned ranges share a bound with the passed range.
\n>>> Event.objects.filter(ages__adjacent_to=NumericRange(10, 21))\n[<Event: Soft play>, <Event: Pub trip>]\nThere are three transforms available for use in queries. You can extract the\nlower or upper bound, or query based on emptiness.
\nReturned objects have the given lower bound. Can be chained to valid lookups\nfor the base field.
\n>>> Event.objects.filter(ages__startswith=21)\n[<Event: Pub trip>]\nReturned objects have the given upper bound. Can be chained to valid lookups\nfor the base field.
\n>>> Event.objects.filter(ages__endswith=10)\n[<Event: Soft play>]\nReturned objects are empty ranges. Can be chained to valid lookups for a\nBooleanField.
>>> Event.objects.filter(ages__isempty=True)\n[]\nPostgreSQL allows the definition of custom range types. Django’s model and form\nfield implementations use base classes below, and psycopg2 provides a\nregister_range() to allow use of custom range\ntypes.
Base class for model range fields.
\nThe model field to use.
\nThe psycopg2 range type to use.
\nThe form field class to use. Should be a subclass of\ndjango.contrib.postgres.forms.BaseRangeField.
Base class for form range fields.
\nThe form field to use.
\nThe psycopg2 range type to use.
\n