A model is the single, definitive source of information about your data. It\ncontains the essential fields and behaviors of the data you’re storing.\nGenerally, each model maps to a single database table.
\nThe basics:
\nEach model is a Python class that subclasses\ndjango.db.models.Model.
Each attribute of the model represents a database field.
With all of this, Django gives you an automatically-generated\ndatabase-access API; see Making queries.
This example model defines a Person, which has a first_name and\nlast_name:
from django.db import models\n\n\nclass Person(models.Model):\n first_name = models.CharField(max_length=30)\n last_name = models.CharField(max_length=30)\nfirst_name and last_name are fields of the model. Each field is\nspecified as a class attribute, and each attribute maps to a database column.
The above Person model would create a database table like this:
CREATE TABLE myapp_person (\n "id" bigint NOT NULL PRIMARY KEY GENERATED BY DEFAULT AS IDENTITY,\n "first_name" varchar(30) NOT NULL,\n "last_name" varchar(30) NOT NULL\n);\nSome technical notes:
\nThe name of the table, myapp_person, is automatically derived from\nsome model metadata but can be overridden. See Table names for more\ndetails.
An id field is added automatically, but this behavior can be\noverridden. See Automatic primary key fields.
The CREATE TABLE SQL in this example is formatted using PostgreSQL\nsyntax, but it’s worth noting Django uses SQL tailored to the database\nbackend specified in your settings file.
Once you have defined your models, you need to tell Django you’re going to use\nthose models. Do this by editing your settings file and changing the\nINSTALLED_APPS setting to add the name of the module that contains\nyour models.py.
For example, if the models for your application live in the module\nmyapp.models (the package structure that is created for an\napplication by the manage.py startapp script),\nINSTALLED_APPS should read, in part:
INSTALLED_APPS = [\n # ...\n "myapp",\n # ...\n]\nWhen you add new apps to INSTALLED_APPS, be sure to run\nmanage.py migrate, optionally making migrations\nfor them first with manage.py makemigrations.
The most important part of a model – and the only required part of a model –\nis the list of database fields it defines. Fields are specified by class\nattributes. Be careful not to choose field names that conflict with the\nmodels API like clean, save, or\ndelete.
Example:
\nfrom django.db import models\n\n\nclass Musician(models.Model):\n first_name = models.CharField(max_length=50)\n last_name = models.CharField(max_length=50)\n instrument = models.CharField(max_length=100)\n\n\nclass Album(models.Model):\n artist = models.ForeignKey(Musician, on_delete=models.CASCADE)\n name = models.CharField(max_length=100)\n release_date = models.DateField()\n num_stars = models.IntegerField()\nEach field in your model should be an instance of the appropriate\nField class. Django uses the field class types to\ndetermine a few things:
The column type, which tells the database what kind of data to store (e.g.\nINTEGER, VARCHAR, TEXT).
The default HTML widget to use when rendering a form\nfield (e.g. <input type="text">, <select>).
The minimal validation requirements, used in Django’s admin and in\nautomatically-generated forms.
Django ships with dozens of built-in field types; you can find the complete list\nin the model field reference. You can easily write\nyour own fields if Django’s built-in ones don’t do the trick; see\nHow to create custom model fields.
\nEach field takes a certain set of field-specific arguments (documented in the\nmodel field reference). For example,\nCharField (and its subclasses) require a\nmax_length argument which specifies the size\nof the VARCHAR database field used to store the data.
There’s also a set of common arguments available to all field types. All are\noptional. They’re fully explained in the reference, but here’s a quick summary of the most often-used\nones:
\nnullIf True, Django will store empty values as NULL in the database.\nDefault is False.
blankIf True, the field is allowed to be blank. Default is False.
Note that this is different than null.\nnull is purely database-related, whereas\nblank is validation-related. If a field has\nblank=True, form validation will\nallow entry of an empty value. If a field has blank=False, the field will be required.
choicesA sequence of 2-value tuples, a mapping, an\nenumeration type, or a callable (that\nexpects no arguments and returns any of the previous formats), to use as\nchoices for this field. If this is given, the default form widget will be a\nselect box instead of the standard text field and will limit choices to the\nchoices given.
\nA choices list looks like this:
\nYEAR_IN_SCHOOL_CHOICES = [\n ("FR", "Freshman"),\n ("SO", "Sophomore"),\n ("JR", "Junior"),\n ("SR", "Senior"),\n ("GR", "Graduate"),\n]\nThe first element in each tuple is the value that will be stored in the\ndatabase. The second element is displayed by the field’s form widget.
\nGiven a model instance, the display value for a field with choices can\nbe accessed using the get_FOO_display()\nmethod. For example:
from django.db import models\n\n\nclass Person(models.Model):\n SHIRT_SIZES = {\n "S": "Small",\n "M": "Medium",\n "L": "Large",\n }\n name = models.CharField(max_length=60)\n shirt_size = models.CharField(max_length=1, choices=SHIRT_SIZES)\n>>> p = Person(name="Fred Flintstone", shirt_size="L")\n>>> p.save()\n>>> p.shirt_size\n'L'\n>>> p.get_shirt_size_display()\n'Large'\nYou can also use enumeration classes to define choices in a concise\nway:
from django.db import models\n\n\nclass Runner(models.Model):\n MedalType = models.TextChoices("MedalType", "GOLD SILVER BRONZE")\n name = models.CharField(max_length=60)\n medal = models.CharField(blank=True, choices=MedalType, max_length=10)\nFurther examples are available in the model field reference.
\n\ndefaultThe default value for the field. This can be a value or a callable\nobject. If callable it will be called every time a new object is\ncreated.
\nhelp_textExtra “help” text to be displayed with the form widget. It’s useful for\ndocumentation even if your field isn’t used on a form.
\nprimary_keyIf True, this field is the primary key for the model.
If you don’t specify primary_key=True for\nany fields in your model, Django will automatically add an\nIntegerField to hold the primary key, so you don’t need to set\nprimary_key=True on any of your fields\nunless you want to override the default primary-key behavior. For more,\nsee Automatic primary key fields.
The primary key field is read-only. If you change the value of the primary\nkey on an existing object and then save it, a new object will be created\nalongside the old one. For example:
\nfrom django.db import models\n\n\nclass Fruit(models.Model):\n name = models.CharField(max_length=100, primary_key=True)\n>>> fruit = Fruit.objects.create(name="Apple")\n>>> fruit.name = "Pear"\n>>> fruit.save()\n>>> Fruit.objects.values_list("name", flat=True)\n<QuerySet ['Apple', 'Pear']>\nuniqueIf True, this field must be unique throughout the table.
Again, these are just short descriptions of the most common field options. Full\ndetails can be found in the common model field option reference.
\nBy default, Django gives each model an auto-incrementing primary key with the\ntype specified per app in AppConfig.default_auto_field or globally in the\nDEFAULT_AUTO_FIELD setting. For example:
id = models.BigAutoField(primary_key=True)\nIf you’d like to specify a custom primary key, specify\nprimary_key=True on one of your fields. If Django\nsees you’ve explicitly set Field.primary_key, it won’t add the automatic\nid column.
Each model requires exactly one field to have primary_key=True (either explicitly declared or automatically added).
Each field type, except for ForeignKey,\nManyToManyField and\nOneToOneField, takes an optional first positional\nargument – a verbose name. If the verbose name isn’t given, Django will\nautomatically create it using the field’s attribute name, converting underscores\nto spaces.
In this example, the verbose name is "person's first name":
first_name = models.CharField("person's first name", max_length=30)\nIn this example, the verbose name is "first name":
first_name = models.CharField(max_length=30)\nForeignKey,\nManyToManyField and\nOneToOneField require the first argument to be a\nmodel class, so use the verbose_name keyword argument:
poll = models.ForeignKey(\n Poll,\n on_delete=models.CASCADE,\n verbose_name="the related poll",\n)\nsites = models.ManyToManyField(Site, verbose_name="list of sites")\nplace = models.OneToOneField(\n Place,\n on_delete=models.CASCADE,\n verbose_name="related place",\n)\nThe convention is not to capitalize the first letter of the\nverbose_name. Django will automatically capitalize the first\nletter where it needs to.
Clearly, the power of relational databases lies in relating tables to each\nother. Django offers ways to define the three most common types of database\nrelationships: many-to-one, many-to-many and one-to-one.
\nTo define a many-to-one relationship, use django.db.models.ForeignKey.\nYou use it just like any other Field type: by\nincluding it as a class attribute of your model.
ForeignKey requires a positional argument: the class\nto which the model is related.
For example, if a Car model has a Manufacturer – that is, a\nManufacturer makes multiple cars but each Car only has one\nManufacturer – use the following definitions:
from django.db import models\n\n\nclass Manufacturer(models.Model):\n # ...\n pass\n\n\nclass Car(models.Model):\n manufacturer = models.ForeignKey(Manufacturer, on_delete=models.CASCADE)\n # ...\nYou can also create recursive relationships (an\nobject with a many-to-one relationship to itself) and relationships to\nmodels not yet defined; see the model field\nreference for details.
\nIt’s suggested, but not required, that the name of a\nForeignKey field (manufacturer in the example\nabove) be the name of the model, lowercase. You can call the field whatever you\nwant. For example:
class Car(models.Model):\n company_that_makes_it = models.ForeignKey(\n Manufacturer,\n on_delete=models.CASCADE,\n )\n # ...\nTo define a many-to-many relationship, use\nManyToManyField. You use it just like any other\nField type: by including it as a class attribute of\nyour model.
ManyToManyField requires a positional argument: the\nclass to which the model is related.
For example, if a Pizza has multiple Topping objects – that is, a\nTopping can be on multiple pizzas and each Pizza has multiple toppings\n– here’s how you’d represent that:
from django.db import models\n\n\nclass Topping(models.Model):\n # ...\n pass\n\n\nclass Pizza(models.Model):\n # ...\n toppings = models.ManyToManyField(Topping)\nAs with ForeignKey, you can also create\nrecursive relationships (an object with a\nmany-to-many relationship to itself) and relationships to models not yet\ndefined.
It’s suggested, but not required, that the name of a\nManyToManyField (toppings in the example above)\nbe a plural describing the set of related model objects.
It doesn’t matter which model has the\nManyToManyField, but you should only put it in one\nof the models – not both.
Generally, ManyToManyField instances should go in\nthe object that’s going to be edited on a form. In the above example,\ntoppings is in Pizza (rather than Topping having a pizzas\nManyToManyField ) because it’s more natural to think\nabout a pizza having toppings than a topping being on multiple pizzas. The way\nit’s set up above, the Pizza form would let users select the toppings.
ManyToManyField fields also accept a number of\nextra arguments which are explained in the model field reference. These options help define how the relationship\nshould work; all are optional.
When you’re only dealing with many-to-many relationships such as mixing and\nmatching pizzas and toppings, a standard\nManyToManyField is all you need. However, sometimes\nyou may need to associate data with the relationship between two models.
For example, consider the case of an application tracking the musical groups\nwhich musicians belong to. There is a many-to-many relationship between a person\nand the groups of which they are a member, so you could use a\nManyToManyField to represent this relationship.\nHowever, there is a lot of detail about the membership that you might want to\ncollect, such as the date at which the person joined the group.
For these situations, Django allows you to specify the model that will be used\nto govern the many-to-many relationship. You can then put extra fields on the\nintermediate model. The intermediate model is associated with the\nManyToManyField using the\nthrough argument to point to the model\nthat will act as an intermediary. For our musician example, the code would look\nsomething like this:
from django.db import models\n\n\nclass Person(models.Model):\n name = models.CharField(max_length=128)\n\n def __str__(self):\n return self.name\n\n\nclass Group(models.Model):\n name = models.CharField(max_length=128)\n members = models.ManyToManyField(Person, through="Membership")\n\n def __str__(self):\n return self.name\n\n\nclass Membership(models.Model):\n person = models.ForeignKey(Person, on_delete=models.CASCADE)\n group = models.ForeignKey(Group, on_delete=models.CASCADE)\n date_joined = models.DateField()\n invite_reason = models.CharField(max_length=64)\nWhen you set up the intermediary model, you explicitly specify foreign\nkeys to the models that are involved in the many-to-many relationship. This\nexplicit declaration defines how the two models are related.
\nThere are a few restrictions on the intermediate model:
\nYour intermediate model must contain one - and only one - foreign key\nto the source model (this would be Group in our example), or you must\nexplicitly specify the foreign keys Django should use for the relationship\nusing ManyToManyField.through_fields.\nIf you have more than one foreign key and through_fields is not\nspecified, a validation error will be raised. A similar restriction applies\nto the foreign key to the target model (this would be Person in our\nexample).
For a model which has a many-to-many relationship to itself through an\nintermediary model, two foreign keys to the same model are permitted, but\nthey will be treated as the two (different) sides of the many-to-many\nrelationship. If there are more than two foreign keys though, you\nmust also specify through_fields as above, or a validation error\nwill be raised.
Now that you have set up your ManyToManyField to use\nyour intermediary model (Membership, in this case), you’re ready to start\ncreating some many-to-many relationships. You do this by creating instances of\nthe intermediate model:
>>> ringo = Person.objects.create(name="Ringo Starr")\n>>> paul = Person.objects.create(name="Paul McCartney")\n>>> beatles = Group.objects.create(name="The Beatles")\n>>> m1 = Membership(\n... person=ringo,\n... group=beatles,\n... date_joined=date(1962, 8, 16),\n... invite_reason="Needed a new drummer.",\n... )\n>>> m1.save()\n>>> beatles.members.all()\n<QuerySet [<Person: Ringo Starr>]>\n>>> ringo.group_set.all()\n<QuerySet [<Group: The Beatles>]>\n>>> m2 = Membership.objects.create(\n... person=paul,\n... group=beatles,\n... date_joined=date(1960, 8, 1),\n... invite_reason="Wanted to form a band.",\n... )\n>>> beatles.members.all()\n<QuerySet [<Person: Ringo Starr>, <Person: Paul McCartney>]>\nYou can also use add(),\ncreate(), or\nset() to create\nrelationships, as long as you specify through_defaults for any required\nfields:
>>> beatles.members.add(john, through_defaults={"date_joined": date(1960, 8, 1)})\n>>> beatles.members.create(\n... name="George Harrison", through_defaults={"date_joined": date(1960, 8, 1)}\n... )\n>>> beatles.members.set(\n... [john, paul, ringo, george], through_defaults={"date_joined": date(1960, 8, 1)}\n... )\nYou may prefer to create instances of the intermediate model directly.
\nIf the custom through table defined by the intermediate model does not enforce\nuniqueness on the (model1, model2) pair, allowing multiple values, the\nremove() call will\nremove all intermediate model instances:
>>> Membership.objects.create(\n... person=ringo,\n... group=beatles,\n... date_joined=date(1968, 9, 4),\n... invite_reason="You've been gone for a month and we miss you.",\n... )\n>>> beatles.members.all()\n<QuerySet [<Person: Ringo Starr>, <Person: Paul McCartney>, <Person: Ringo Starr>]>\n>>> # This deletes both of the intermediate model instances for Ringo Starr\n>>> beatles.members.remove(ringo)\n>>> beatles.members.all()\n<QuerySet [<Person: Paul McCartney>]>\nThe clear()\nmethod can be used to remove all many-to-many relationships for an instance:
>>> # Beatles have broken up\n>>> beatles.members.clear()\n>>> # Note that this deletes the intermediate model instances\n>>> Membership.objects.all()\n<QuerySet []>\nOnce you have established the many-to-many relationships, you can issue\nqueries. Just as with normal many-to-many relationships, you can query using\nthe attributes of the many-to-many-related model:
\n# Find all the groups with a member whose name starts with 'Paul'\n>>> Group.objects.filter(members__name__startswith="Paul")\n<QuerySet [<Group: The Beatles>]>\nAs you are using an intermediate model, you can also query on its attributes:
\n# Find all the members of the Beatles that joined after 1 Jan 1961\n>>> Person.objects.filter(\n... group__name="The Beatles", membership__date_joined__gt=date(1961, 1, 1)\n... )\n<QuerySet [<Person: Ringo Starr]>\nIf you need to access a membership’s information you may do so by directly\nquerying the Membership model:
>>> ringos_membership = Membership.objects.get(group=beatles, person=ringo)\n>>> ringos_membership.date_joined\ndatetime.date(1962, 8, 16)\n>>> ringos_membership.invite_reason\n'Needed a new drummer.'\nAnother way to access the same information is by querying the\nmany-to-many reverse relationship from a\nPerson object:
>>> ringos_membership = ringo.membership_set.get(group=beatles)\n>>> ringos_membership.date_joined\ndatetime.date(1962, 8, 16)\n>>> ringos_membership.invite_reason\n'Needed a new drummer.'\nTo define a one-to-one relationship, use\nOneToOneField. You use it just like any other\nField type: by including it as a class attribute of your model.
This is most useful on the primary key of an object when that object “extends”\nanother object in some way.
\nOneToOneField requires a positional argument: the\nclass to which the model is related.
For example, if you were building a database of “places”, you would\nbuild pretty standard stuff such as address, phone number, etc. in the\ndatabase. Then, if you wanted to build a database of restaurants on\ntop of the places, instead of repeating yourself and replicating those\nfields in the Restaurant model, you could make Restaurant have\na OneToOneField to Place (because a\nrestaurant “is a” place; in fact, to handle this you’d typically use\ninheritance, which involves an implicit\none-to-one relation).
As with ForeignKey, a recursive relationship can be defined and references to as-yet\nundefined models can be made.
OneToOneField fields also accept an optional\nparent_link argument.
OneToOneField classes used to automatically become\nthe primary key on a model. This is no longer true (although you can manually\npass in the primary_key argument if you like).\nThus, it’s now possible to have multiple fields of type\nOneToOneField on a single model.
It’s perfectly OK to relate a model to one from another app. To do this, import\nthe related model at the top of the file where your model is defined. Then,\nrefer to the other model class wherever needed. For example:
\nfrom django.db import models\nfrom geography.models import ZipCode\n\n\nclass Restaurant(models.Model):\n # ...\n zip_code = models.ForeignKey(\n ZipCode,\n on_delete=models.SET_NULL,\n blank=True,\n null=True,\n )\nDjango places some restrictions on model field names:
\nA field name cannot be a Python reserved word, because that would result\nin a Python syntax error. For example:
\nclass Example(models.Model):\n pass = models.IntegerField() # 'pass' is a reserved word!\nA field name cannot contain more than one underscore in a row, due to\nthe way Django’s query lookup syntax works. For example:
\nclass Example(models.Model):\n foo__bar = models.IntegerField() # 'foo__bar' has two underscores!\nA field name cannot end with an underscore, for similar reasons.
These limitations can be worked around, though, because your field name doesn’t\nnecessarily have to match your database column name. See the\ndb_column option.
SQL reserved words, such as join, where or select, are allowed as\nmodel field names, because Django escapes all database table names and column\nnames in every underlying SQL query. It uses the quoting syntax of your\nparticular database engine.
If one of the existing model fields cannot be used to fit your purposes, or if\nyou wish to take advantage of some less common database column types, you can\ncreate your own field class. Full coverage of creating your own fields is\nprovided in How to create custom model fields.
\nMeta optionsLink to this heading#Give your model metadata by using an inner class Meta, like so:
from django.db import models\n\n\nclass Ox(models.Model):\n horn_length = models.IntegerField()\n\n class Meta:\n ordering = ["horn_length"]\n verbose_name_plural = "oxen"\nModel metadata is “anything that’s not a field”, such as ordering options\n(ordering), database table name (db_table), or\nhuman-readable singular and plural names (verbose_name and\nverbose_name_plural). None are required, and adding class\nMeta to a model is completely optional.
A complete list of all possible Meta options can be found in the model\noption reference.
objectsThe most important attribute of a model is the\nManager. It’s the interface through which\ndatabase query operations are provided to Django models and is used to\nretrieve the instances from the database. If no\ncustom Manager is defined, the default name is\nobjects. Managers are only accessible via\nmodel classes, not the model instances.
Define custom methods on a model to add custom “row-level” functionality to your\nobjects. Whereas Manager methods are intended to do\n“table-wide” things, model methods should act on a particular model instance.
This is a valuable technique for keeping business logic in one place – the\nmodel.
\nFor example, this model has a few custom methods:
\nfrom django.db import models\n\n\nclass Person(models.Model):\n first_name = models.CharField(max_length=50)\n last_name = models.CharField(max_length=50)\n birth_date = models.DateField()\n\n def baby_boomer_status(self):\n "Returns the person's baby-boomer status."\n import datetime\n\n if self.birth_date < datetime.date(1945, 8, 1):\n return "Pre-boomer"\n elif self.birth_date < datetime.date(1965, 1, 1):\n return "Baby boomer"\n else:\n return "Post-boomer"\n\n @property\n def full_name(self):\n "Returns the person's full name."\n return f"{self.first_name} {self.last_name}"\nThe last method in this example is a property.
\nThe model instance reference has a complete list\nof methods automatically given to each model.\nYou can override most of these – see overriding predefined model methods,\nbelow – but there are a couple that you’ll almost always want to define:
\n__str__()A Python “magic method” that returns a string representation of any\nobject. This is what Python and Django will use whenever a model\ninstance needs to be coerced and displayed as a plain string. Most\nnotably, this happens when you display an object in an interactive\nconsole or in the admin.
\nYou’ll always want to define this method; the default isn’t very helpful\nat all.
\nget_absolute_url()This tells Django how to calculate the URL for an object. Django uses\nthis in its admin interface, and any time it needs to figure out a URL\nfor an object.
\nAny object that has a URL that uniquely identifies it should define this\nmethod.
\nThere’s another set of model methods that\nencapsulate a bunch of database behavior that you’ll want to customize. In\nparticular you’ll often want to change the way save() and\ndelete() work.
You’re free to override these methods (and any other model method) to alter\nbehavior.
\nA classic use-case for overriding the built-in methods is if you want something\nto happen whenever you save an object. For example (see\nsave() for documentation of the parameters it accepts):
from django.db import models\n\n\nclass Blog(models.Model):\n name = models.CharField(max_length=100)\n tagline = models.TextField()\n\n def save(self, *args, **kwargs):\n do_something()\n super().save(*args, **kwargs) # Call the "real" save() method.\n do_something_else()\nYou can also prevent saving:
\nfrom django.db import models\n\n\nclass Blog(models.Model):\n name = models.CharField(max_length=100)\n tagline = models.TextField()\n\n def save(self, *args, **kwargs):\n if self.name == "Yoko Ono's blog":\n return # Yoko shall never have her own blog!\n else:\n super().save(*args, **kwargs) # Call the "real" save() method.\nIt’s important to remember to call the superclass method – that’s\nthat super().save(*args, **kwargs) business – to ensure\nthat the object still gets saved into the database. If you forget to\ncall the superclass method, the default behavior won’t happen and the\ndatabase won’t get touched.
It’s also important that you pass through the arguments that can be\npassed to the model method – that’s what the *args, **kwargs bit\ndoes. Django will, from time to time, extend the capabilities of\nbuilt-in model methods, adding new arguments. If you use *args,\n**kwargs in your method definitions, you are guaranteed that your\ncode will automatically support those arguments when they are added.
If you wish to update a field value in the save() method, you may\nalso want to have this field added to the update_fields keyword argument.\nThis will ensure the field is saved when update_fields is specified. For\nexample:
from django.db import models\nfrom django.utils.text import slugify\n\n\nclass Blog(models.Model):\n name = models.CharField(max_length=100)\n slug = models.TextField()\n\n def save(\n self, force_insert=False, force_update=False, using=None, update_fields=None\n ):\n self.slug = slugify(self.name)\n if update_fields is not None and "name" in update_fields:\n update_fields = {"slug"}.union(update_fields)\n super().save(\n force_insert=force_insert,\n force_update=force_update,\n using=using,\n update_fields=update_fields,\n )\nSee Specifying which fields to save for more details.
\n\nAnother common pattern is writing custom SQL statements in model methods and\nmodule-level methods. For more details on using raw SQL, see the documentation\non using raw SQL.
\nModel inheritance in Django works almost identically to the way normal\nclass inheritance works in Python, but the basics at the beginning of the page\nshould still be followed. That means the base class should subclass\ndjango.db.models.Model.
The only decision you have to make is whether you want the parent models to be\nmodels in their own right (with their own database tables), or if the parents\nare just holders of common information that will only be visible through the\nchild models.
\nThere are three styles of inheritance that are possible in Django.
\nOften, you will just want to use the parent class to hold information that\nyou don’t want to have to type out for each child model. This class isn’t\ngoing to ever be used in isolation, so Abstract base classes are\nwhat you’re after.
If you’re subclassing an existing model (perhaps something from another\napplication entirely) and want each model to have its own database table,\nMulti-table inheritance is the way to go.
Finally, if you only want to modify the Python-level behavior of a model,\nwithout changing the models fields in any way, you can use\nProxy models.
Abstract base classes are useful when you want to put some common\ninformation into a number of other models. You write your base class\nand put abstract=True in the Meta\nclass. This model will then not be used to create any database\ntable. Instead, when it is used as a base class for other models, its\nfields will be added to those of the child class.
An example:
\nfrom django.db import models\n\n\nclass CommonInfo(models.Model):\n name = models.CharField(max_length=100)\n age = models.PositiveIntegerField()\n\n class Meta:\n abstract = True\n\n\nclass Student(CommonInfo):\n home_group = models.CharField(max_length=5)\nThe Student model will have three fields: name, age and\nhome_group. The CommonInfo model cannot be used as a normal Django\nmodel, since it is an abstract base class. It does not generate a database\ntable or have a manager, and cannot be instantiated or saved directly.
Fields inherited from abstract base classes can be overridden with another\nfield or value, or be removed with None.
For many uses, this type of model inheritance will be exactly what you want.\nIt provides a way to factor out common information at the Python level, while\nstill only creating one database table per child model at the database level.
\nMeta inheritanceLink to this heading#When an abstract base class is created, Django makes any Meta\ninner class you declared in the base class available as an\nattribute. If a child class does not declare its own Meta\nclass, it will inherit the parent’s Meta. If the child wants to\nextend the parent’s Meta class, it can subclass it. For example:
\nfrom django.db import models\n\n\nclass CommonInfo(models.Model):\n # ...\n class Meta:\n abstract = True\n ordering = ["name"]\n\n\nclass Student(CommonInfo):\n # ...\n class Meta(CommonInfo.Meta):\n db_table = "student_info"\nDjango does make one adjustment to the Meta class of an\nabstract base class: before installing the Meta\nattribute, it sets abstract=False. This means that children of abstract\nbase classes don’t automatically become abstract classes themselves. To make\nan abstract base class that inherits from another abstract base class, you need\nto explicitly set abstract=True on the child.
Some attributes won’t make sense to include in the Meta class of an\nabstract base class. For example, including db_table would mean that all\nthe child classes (the ones that don’t specify their own Meta) would use\nthe same database table, which is almost certainly not what you want.
Due to the way Python inheritance works, if a child class inherits from\nmultiple abstract base classes, only the Meta options\nfrom the first listed class will be inherited by default. To inherit Meta options from multiple abstract base classes, you must\nexplicitly declare the Meta inheritance. For example:
\nfrom django.db import models\n\n\nclass CommonInfo(models.Model):\n name = models.CharField(max_length=100)\n age = models.PositiveIntegerField()\n\n class Meta:\n abstract = True\n ordering = ["name"]\n\n\nclass Unmanaged(models.Model):\n class Meta:\n abstract = True\n managed = False\n\n\nclass Student(CommonInfo, Unmanaged):\n home_group = models.CharField(max_length=5)\n\n class Meta(CommonInfo.Meta, Unmanaged.Meta):\n pass\nThe second type of model inheritance supported by Django is when each model in\nthe hierarchy is a model all by itself. Each model corresponds to its own\ndatabase table and can be queried and created individually. The inheritance\nrelationship introduces links between the child model and each of its parents\n(via an automatically-created OneToOneField).\nFor example:
from django.db import models\n\n\nclass Place(models.Model):\n name = models.CharField(max_length=50)\n address = models.CharField(max_length=80)\n\n\nclass Restaurant(Place):\n serves_hot_dogs = models.BooleanField(default=False)\n serves_pizza = models.BooleanField(default=False)\nAll of the fields of Place will also be available in Restaurant,\nalthough the data will reside in a different database table. So these are both\npossible:
>>> Place.objects.filter(name="Bob's Cafe")\n>>> Restaurant.objects.filter(name="Bob's Cafe")\nIf you have a Place that is also a Restaurant, you can get from the\nPlace object to the Restaurant object by using the lowercase version of\nthe model name:
>>> p = Place.objects.get(id=12)\n# If p is a Restaurant object, this will give the child class:\n>>> p.restaurant\n<Restaurant: ...>\nHowever, if p in the above example was not a Restaurant (it had been\ncreated directly as a Place object or was the parent of some other class),\nreferring to p.restaurant would raise a Restaurant.DoesNotExist\nexception.
The automatically-created OneToOneField on\nRestaurant that links it to Place looks like this:
place_ptr = models.OneToOneField(\n Place,\n on_delete=models.CASCADE,\n parent_link=True,\n primary_key=True,\n)\nYou can override that field by declaring your own\nOneToOneField with parent_link=True on Restaurant.
Meta and multi-table inheritanceLink to this heading#In the multi-table inheritance situation, it doesn’t make sense for a child\nclass to inherit from its parent’s Meta class. All the Meta options\nhave already been applied to the parent class and applying them again would\nnormally only lead to contradictory behavior (this is in contrast with the\nabstract base class case, where the base class doesn’t exist in its own\nright).
\nSo a child model does not have access to its parent’s Meta class. However, there are a few limited cases where the child\ninherits behavior from the parent: if the child does not specify an\nordering attribute or a\nget_latest_by attribute, it will inherit\nthese from its parent.
If the parent has an ordering and you don’t want the child to have any natural\nordering, you can explicitly disable it:
\nclass ChildModel(ParentModel):\n # ...\n class Meta:\n # Remove parent's ordering effect\n ordering = []\nBecause multi-table inheritance uses an implicit\nOneToOneField to link the child and\nthe parent, it’s possible to move from the parent down to the child,\nas in the above example. However, this uses up the name that is the\ndefault related_name value for\nForeignKey and\nManyToManyField relations. If you\nare putting those types of relations on a subclass of the parent model, you\nmust specify the related_name\nattribute on each such field. If you forget, Django will raise a validation\nerror.
For example, using the above Place class again, let’s create another\nsubclass with a ManyToManyField:
class Supplier(Place):\n customers = models.ManyToManyField(Place)\nThis results in the error:
\nReverse query name for 'Supplier.customers' clashes with reverse query\nname for 'Supplier.place_ptr'.\n\nHINT: Add or change a related_name argument to the definition for\n'Supplier.customers' or 'Supplier.place_ptr'.\nAdding related_name to the customers field as follows would resolve the\nerror: models.ManyToManyField(Place, related_name='provider').
As mentioned, Django will automatically create a\nOneToOneField linking your child\nclass back to any non-abstract parent models. If you want to control the\nname of the attribute linking back to the parent, you can create your\nown OneToOneField and set\nparent_link=True\nto indicate that your field is the link back to the parent class.
When using multi-table inheritance, a new\ndatabase table is created for each subclass of a model. This is usually the\ndesired behavior, since the subclass needs a place to store any additional\ndata fields that are not present on the base class. Sometimes, however, you\nonly want to change the Python behavior of a model – perhaps to change the\ndefault manager, or add a new method.
\nThis is what proxy model inheritance is for: creating a proxy for the\noriginal model. You can create, delete and update instances of the proxy model\nand all the data will be saved as if you were using the original (non-proxied)\nmodel. The difference is that you can change things like the default model\nordering or the default manager in the proxy, without having to alter the\noriginal.
\nProxy models are declared like normal models. You tell Django that it’s a\nproxy model by setting the proxy attribute of\nthe Meta class to True.
For example, suppose you want to add a method to the Person model. You can do it like this:
from django.db import models\n\n\nclass Person(models.Model):\n first_name = models.CharField(max_length=30)\n last_name = models.CharField(max_length=30)\n\n\nclass MyPerson(Person):\n class Meta:\n proxy = True\n\n def do_something(self):\n # ...\n pass\nThe MyPerson class operates on the same database table as its parent\nPerson class. In particular, any new instances of Person will also be\naccessible through MyPerson, and vice-versa:
>>> p = Person.objects.create(first_name="foobar")\n>>> MyPerson.objects.get(first_name="foobar")\n<MyPerson: foobar>\nYou could also use a proxy model to define a different default ordering on\na model. You might not always want to order the Person model, but regularly\norder by the last_name attribute when you use the proxy:
class OrderedPerson(Person):\n class Meta:\n ordering = ["last_name"]\n proxy = True\nNow normal Person queries will be unordered\nand OrderedPerson queries will be ordered by last_name.
Proxy models inherit Meta attributes in the same way as regular\nmodels.
QuerySets still return the model that was requestedLink to this heading#There is no way to have Django return, say, a MyPerson object whenever you\nquery for Person objects. A queryset for Person objects will return\nthose types of objects. The whole point of proxy objects is that code relying\non the original Person will use those and your own code can use the\nextensions you included (that no other code is relying on anyway). It is not\na way to replace the Person (or any other) model everywhere with something\nof your own creation.
A proxy model must inherit from exactly one non-abstract model class. You\ncan’t inherit from multiple non-abstract models as the proxy model doesn’t\nprovide any connection between the rows in the different database tables. A\nproxy model can inherit from any number of abstract model classes, providing\nthey do not define any model fields. A proxy model may also inherit from any\nnumber of proxy models that share a common non-abstract parent class.
\nIf you don’t specify any model managers on a proxy model, it inherits the\nmanagers from its model parents. If you define a manager on the proxy model,\nit will become the default, although any managers defined on the parent\nclasses will still be available.
\nContinuing our example from above, you could change the default manager used\nwhen you query the Person model like this:
from django.db import models\n\n\nclass NewManager(models.Manager):\n # ...\n pass\n\n\nclass MyPerson(Person):\n objects = NewManager()\n\n class Meta:\n proxy = True\nIf you wanted to add a new manager to the Proxy, without replacing the\nexisting default, you can use the techniques described in the custom\nmanager documentation: create a base class\ncontaining the new managers and inherit that after the primary base class:
\n# Create an abstract class for the new manager.\nclass ExtraManagers(models.Model):\n secondary = NewManager()\n\n class Meta:\n abstract = True\n\n\nclass MyPerson(Person, ExtraManagers):\n class Meta:\n proxy = True\nYou probably won’t need to do this very often, but, when you do, it’s\npossible.
\nProxy model inheritance might look fairly similar to creating an unmanaged\nmodel, using the managed attribute on a\nmodel’s Meta class.
With careful setting of Meta.db_table you could create an unmanaged model that\nshadows an existing model and adds Python methods to it. However, that would be\nvery repetitive and fragile as you need to keep both copies synchronized if you\nmake any changes.
On the other hand, proxy models are intended to behave exactly like the model\nthey are proxying for. They are always in sync with the parent model since they\ndirectly inherit its fields and managers.
\nThe general rules are:
\nIf you are mirroring an existing model or database table and don’t want\nall the original database table columns, use Meta.managed=False.\nThat option is normally useful for modeling database views and tables\nnot under the control of Django.
If you are wanting to change the Python-only behavior of a model, but\nkeep all the same fields as in the original, use Meta.proxy=True.\nThis sets things up so that the proxy model is an exact copy of the\nstorage structure of the original model when data is saved.
Just as with Python’s subclassing, it’s possible for a Django model to inherit\nfrom multiple parent models. Keep in mind that normal Python name resolution\nrules apply. The first base class that a particular name (e.g. Meta) appears in will be the one that is used; for example, this\nmeans that if multiple parents contain a Meta class,\nonly the first one is going to be used, and all others will be ignored.
\nGenerally, you won’t need to inherit from multiple parents. The main use-case\nwhere this is useful is for “mix-in” classes: adding a particular extra\nfield or method to every class that inherits the mix-in. Try to keep your\ninheritance hierarchies as simple and straightforward as possible so that you\nwon’t have to struggle to work out where a particular piece of information is\ncoming from.
\nNote that inheriting from multiple models that have a common id primary\nkey field will raise an error. To properly use multiple inheritance, you can\nuse an explicit AutoField in the base models:
class Article(models.Model):\n article_id = models.AutoField(primary_key=True)\n ...\n\n\nclass Book(models.Model):\n book_id = models.AutoField(primary_key=True)\n ...\n\n\nclass BookReview(Book, Article):\n pass\nOr use a common ancestor to hold the AutoField. This\nrequires using an explicit OneToOneField from each\nparent model to the common ancestor to avoid a clash between the fields that\nare automatically generated and inherited by the child:
class Piece(models.Model):\n pass\n\n\nclass Article(Piece):\n article_piece = models.OneToOneField(\n Piece, on_delete=models.CASCADE, parent_link=True\n )\n ...\n\n\nclass Book(Piece):\n book_piece = models.OneToOneField(Piece, on_delete=models.CASCADE, parent_link=True)\n ...\n\n\nclass BookReview(Book, Article):\n pass\nIn normal Python class inheritance, it is permissible for a child class to\noverride any attribute from the parent class. In Django, this isn’t usually\npermitted for model fields. If a non-abstract model base class has a field\ncalled author, you can’t create another model field or define\nan attribute called author in any class that inherits from that base class.
This restriction doesn’t apply to model fields inherited from an abstract\nmodel. Such fields may be overridden with another field or value, or be removed\nby setting field_name = None.
Overriding fields in a parent model leads to difficulties in areas such as\ninitializing new instances (specifying which field is being initialized in\nModel.__init__) and serialization. These are features which normal Python\nclass inheritance doesn’t have to deal with in quite the same way, so the\ndifference between Django model inheritance and Python class inheritance isn’t\narbitrary.
This restriction only applies to attributes which are\nField instances. Normal Python attributes\ncan be overridden if you wish. It also only applies to the name of the\nattribute as Python sees it: if you are manually specifying the database\ncolumn name, you can have the same column name appearing in both a child and\nan ancestor model for multi-table inheritance (they are columns in two\ndifferent database tables).
Django will raise a FieldError if you override\nany model field in any ancestor model.
Note that because of the way fields are resolved during class definition, model\nfields inherited from multiple abstract parent models are resolved in a strict\ndepth-first order. This contrasts with standard Python MRO, which is resolved\nbreadth-first in cases of diamond shaped inheritance. This difference only\naffects complex model hierarchies, which (as per the advice above) you should\ntry to avoid.
\nThe manage.py startapp command creates an application\nstructure that includes a models.py file. If you have many models,\norganizing them in separate files may be useful.
To do so, create a models package. Remove models.py and create a\nmyapp/models/ directory with an __init__.py file and the files to\nstore your models. You must import the models in the __init__.py file.
For example, if you had organic.py and synthetic.py in the models\ndirectory:
myapp/models/__init__.pyfrom .organic import Person\nfrom .synthetic import Robot\nExplicitly importing each model rather than using from .models import *\nhas the advantages of not cluttering the namespace, making code more readable,\nand keeping code analysis tools useful.