Saving ForeignKey and ManyToManyField fields¶

Updating a ForeignKey field works exactly the same way as saving a normal field – assign an object of the right type to the field in question. This example updates the blog attribute of an Entry instance entry, assuming appropriate instances of Entry and Blog are already saved to the database (so we can retrieve them below):

>>> from blog.models import Blog, Entry
>>> entry = Entry.objects.get(pk=1)
>>> cheese_blog = Blog.objects.get(name="Cheddar Talk")
>>> entry.blog = cheese_blog
>>> entry.save()

Updating a ManyToManyField works a little differently – use the add() method on the field to add a record to the relation. This example adds the Author instance joe to the entry object:

>>> from blog.models import Author
>>> joe = Author.objects.create(name="Joe")
>>> entry.authors.add(joe)

To add multiple records to a ManyToManyField in one go, include multiple arguments in the call to add(), like this:

>>> john = Author.objects.create(name="John")
>>> paul = Author.objects.create(name="Paul")
>>> george = Author.objects.create(name="George")
>>> ringo = Author.objects.create(name="Ringo")
>>> entry.authors.add(john, paul, george, ringo)

Django will complain if you try to assign or add an object of the wrong type.

Retrieving all objects¶

The simplest way to retrieve objects from a table is to get all of them. To do this, use the all() method on a Manager:

>>> all_entries = Entry.objects.all()

The all() method returns a QuerySet of all the objects in the database.

Retrieving specific objects with filters¶

The QuerySet returned by all() describes all objects in the database table. Usually, though, you’ll need to select only a subset of the complete set of objects.

To create such a subset, you refine the initial QuerySet, adding filter conditions. The two most common ways to refine a QuerySet are:

filter(**kwargs)

Returns a new QuerySet containing objects that match the given lookup parameters.

exclude(**kwargs)

Returns a new QuerySet containing objects that do not match the given lookup parameters.

The lookup parameters (**kwargs in the above function definitions) should be in the format described in Field lookups below.

For example, to get a QuerySet of blog entries from the year 2006, use filter() like so:

Entry.objects.filter(pub_date__year=2006)

With the default manager class, it is the same as:

Entry.objects.all().filter(pub_date__year=2006)

>>> Entry.objects.filter(headline__startswith="What").exclude(
...     pub_date__gte=datetime.date.today()
... ).filter(pub_date__gte=datetime.date(2005, 1, 30))

>>> q1 = Entry.objects.filter(headline__startswith="What")
>>> q2 = q1.exclude(pub_date__gte=datetime.date.today())
>>> q3 = q1.filter(pub_date__gte=datetime.date.today())

>>> q = Entry.objects.filter(headline__startswith="What")
>>> q = q.filter(pub_date__lte=datetime.date.today())
>>> q = q.exclude(body_text__icontains="food")
>>> print(q)

Retrieving a single object with get()¶

filter() will always give you a QuerySet, even if only a single object matches the query – in this case, it will be a QuerySet containing a single element.

If you know there is only one object that matches your query, you can use the get() method on a Manager which returns the object directly:

>>> one_entry = Entry.objects.get(pk=1)

You can use any query expression with get(), just like with filter() – again, see Field lookups below.

Note that there is a difference between using get(), and using filter() with a slice of [0]. If there are no results that match the query, get() will raise a DoesNotExist exception. This exception is an attribute of the model class that the query is being performed on – so in the code above, if there is no Entry object with a primary key of 1, Django will raise Entry.DoesNotExist.

Similarly, Django will complain if more than one item matches the get() query. In this case, it will raise MultipleObjectsReturned, which again is an attribute of the model class itself.

Other QuerySet methods¶

Most of the time you’ll use all(), get(), filter() and exclude() when you need to look up objects from the database. However, that’s far from all there is; see the QuerySet API Reference for a complete list of all the various QuerySet methods.

Limiting QuerySets¶

Use a subset of Python’s array-slicing syntax to limit your QuerySet to a certain number of results. This is the equivalent of SQL’s LIMIT and OFFSET clauses.

For example, this returns the first 5 objects (LIMIT 5):

>>> Entry.objects.all()[:5]

This returns the sixth through tenth objects (OFFSET 5 LIMIT 5):

>>> Entry.objects.all()[5:10]

Negative indexing (i.e. Entry.objects.all()[-1]) is not supported.

Generally, slicing a QuerySet returns a new QuerySet – it doesn’t evaluate the query. An exception is if you use the “step” parameter of Python slice syntax. For example, this would actually execute the query in order to return a list of every second object of the first 10:

>>> Entry.objects.all()[:10:2]

Further filtering or ordering of a sliced queryset is prohibited due to the ambiguous nature of how that might work.

To retrieve a single object rather than a list (e.g. SELECT foo FROM bar LIMIT 1), use an index instead of a slice. For example, this returns the first Entry in the database, after ordering entries alphabetically by headline:

>>> Entry.objects.order_by("headline")[0]

This is roughly equivalent to:

>>> Entry.objects.order_by("headline")[0:1].get()

Note, however, that the first of these will raise IndexError while the second will raise DoesNotExist if no objects match the given criteria. See get() for more details.

Field lookups¶

Field lookups are how you specify the meat of an SQL WHERE clause. They’re specified as keyword arguments to the QuerySet methods filter(), exclude() and get().

Basic lookups keyword arguments take the form field__lookuptype=value. (That’s a double-underscore). For example:

>>> Entry.objects.filter(pub_date__lte="2006-01-01")

translates (roughly) into the following SQL:

SELECT * FROM blog_entry WHERE pub_date <= '2006-01-01';

The field specified in a lookup has to be the name of a model field. There’s one exception though, in case of a ForeignKey you can specify the field name suffixed with _id. In this case, the value parameter is expected to contain the raw value of the foreign model’s primary key. For example:

>>> Entry.objects.filter(blog_id=4)

If you pass an invalid keyword argument, a lookup function will raise TypeError.

The database API supports about two dozen lookup types; a complete reference can be found in the field lookup reference. To give you a taste of what’s available, here’s some of the more common lookups you’ll probably use:

exact

An “exact” match. For example:

>>> Entry.objects.get(headline__exact="Cat bites dog")

Would generate SQL along these lines:

SELECT ... WHERE headline = 'Cat bites dog';

If you don’t provide a lookup type – that is, if your keyword argument doesn’t contain a double underscore – the lookup type is assumed to be exact.

For example, the following two statements are equivalent:

>>> Blog.objects.get(id__exact=14)  # Explicit form
>>> Blog.objects.get(id=14)  # __exact is implied

This is for convenience, because exact lookups are the common case.

iexact

A case-insensitive match. So, the query:

>>> Blog.objects.get(name__iexact="beatles blog")

Would match a Blog titled "Beatles Blog", "beatles blog", or even "BeAtlES blOG".

contains

Case-sensitive containment test. For example:

Entry.objects.get(headline__contains="Lennon")

Roughly translates to this SQL:

SELECT ... WHERE headline LIKE '%Lennon%';

Note this will match the headline 'Today Lennon honored' but not 'today lennon honored'.

There’s also a case-insensitive version, icontains.

startswith, endswith

Starts-with and ends-with search, respectively. There are also case-insensitive versions called istartswith and iendswith.

Again, this only scratches the surface. A complete reference can be found in the field lookup reference.

Lookups that span relationships¶

Django offers a powerful and intuitive way to “follow” relationships in lookups, taking care of the SQL JOINs for you automatically, behind the scenes. To span a relationship, use the field name of related fields across models, separated by double underscores, until you get to the field you want.

This example retrieves all Entry objects with a Blog whose name is 'Beatles Blog':

>>> Entry.objects.filter(blog__name="Beatles Blog")

This spanning can be as deep as you’d like.

It works backwards, too. While it can be customized, by default you refer to a “reverse” relationship in a lookup using the lowercase name of the model.

This example retrieves all Blog objects which have at least one Entry whose headline contains 'Lennon':

>>> Blog.objects.filter(entry__headline__contains="Lennon")

If you are filtering across multiple relationships and one of the intermediate models doesn’t have a value that meets the filter condition, Django will treat it as if there is an empty (all values are NULL), but valid, object there. All this means is that no error will be raised. For example, in this filter:

Blog.objects.filter(entry__authors__name="Lennon")

(if there was a related Author model), if there was no author associated with an entry, it would be treated as if there was also no name attached, rather than raising an error because of the missing author. Usually this is exactly what you want to have happen. The only case where it might be confusing is if you are using isnull. Thus:

Blog.objects.filter(entry__authors__name__isnull=True)

will return Blog objects that have an empty name on the author and also those which have an empty author on the entry. If you don’t want those latter objects, you could write:

Blog.objects.filter(entry__authors__isnull=False, entry__authors__name__isnull=True)

Blog.objects.filter(entry__headline__contains="Lennon", entry__pub_date__year=2008)

Blog.objects.filter(entry__headline__contains="Lennon").filter(
    entry__pub_date__year=2008
)

>>> from datetime import date
>>> beatles = Blog.objects.create(name='Beatles Blog')
>>> pop = Blog.objects.create(name='Pop Music Blog')
>>> Entry.objects.create(
...     blog=beatles,
...     headline='New Lennon Biography',
...     pub_date=date(2008, 6, 1),
... )
<Entry: New Lennon Biography>
>>> Entry.objects.create(
...     blog=beatles,
...     headline='New Lennon Biography in Paperback',
...     pub_date=date(2009, 6, 1),
... )
<Entry: New Lennon Biography in Paperback>
>>> Entry.objects.create(
...     blog=pop,
...     headline='Best Albums of 2008',
...     pub_date=date(2008, 12, 15),
... )
<Entry: Best Albums of 2008>
>>> Entry.objects.create(
...     blog=pop,
...     headline='Lennon Would Have Loved Hip Hop',
...     pub_date=date(2020, 4, 1),
... )
<Entry: Lennon Would Have Loved Hip Hop>
>>> Blog.objects.filter(
...     entry__headline__contains='Lennon',
...     entry__pub_date__year=2008,
... )
<QuerySet [<Blog: Beatles Blog>]>
>>> Blog.objects.filter(
...     entry__headline__contains='Lennon',
... ).filter(
...     entry__pub_date__year=2008,
... )
<QuerySet [<Blog: Beatles Blog>, <Blog: Beatles Blog>, <Blog: Pop Music Blog]>

Blog.objects.exclude(
    entry__headline__contains="Lennon",
    entry__pub_date__year=2008,
)

Blog.objects.exclude(
    entry__in=Entry.objects.filter(
        headline__contains="Lennon",
        pub_date__year=2008,
    ),
)

Filters can reference fields on the model¶

In the examples given so far, we have constructed filters that compare the value of a model field with a constant. But what if you want to compare the value of a model field with another field on the same model?

Django provides F expressions to allow such comparisons. Instances of F() act as a reference to a model field within a query. These references can then be used in query filters to compare the values of two different fields on the same model instance.

For example, to find a list of all blog entries that have had more comments than pingbacks, we construct an F() object to reference the pingback count, and use that F() object in the query:

>>> from django.db.models import F
>>> Entry.objects.filter(number_of_comments__gt=F("number_of_pingbacks"))

Django supports the use of addition, subtraction, multiplication, division, modulo, and power arithmetic with F() objects, both with constants and with other F() objects. To find all the blog entries with more than twice as many comments as pingbacks, we modify the query:

>>> Entry.objects.filter(number_of_comments__gt=F("number_of_pingbacks") * 2)

To find all the entries where the rating of the entry is less than the sum of the pingback count and comment count, we would issue the query:

>>> Entry.objects.filter(rating__lt=F("number_of_comments") + F("number_of_pingbacks"))

You can also use the double underscore notation to span relationships in an F() object. An F() object with a double underscore will introduce any joins needed to access the related object. For example, to retrieve all the entries where the author’s name is the same as the blog name, we could issue the query:

>>> Entry.objects.filter(authors__name=F("blog__name"))

For date and date/time fields, you can add or subtract a timedelta object. The following would return all entries that were modified more than 3 days after they were published:

>>> from datetime import timedelta
>>> Entry.objects.filter(mod_date__gt=F("pub_date") + timedelta(days=3))

The F() objects support bitwise operations by .bitand(), .bitor(), .bitxor(), .bitrightshift(), and .bitleftshift(). For example:

>>> F("somefield").bitand(16)

Expressions can reference transforms¶

Django supports using transforms in expressions.

For example, to find all Entry objects published in the same year as they were last modified:

>>> from django.db.models import F
>>> Entry.objects.filter(pub_date__year=F("mod_date__year"))

To find the earliest year an entry was published, we can issue the query:

>>> from django.db.models import Min
>>> Entry.objects.aggregate(first_published_year=Min("pub_date__year"))

This example finds the value of the highest rated entry and the total number of comments on all entries for each year:

>>> from django.db.models import OuterRef, Subquery, Sum
>>> Entry.objects.values("pub_date__year").annotate(
...     top_rating=Subquery(
...         Entry.objects.filter(
...             pub_date__year=OuterRef("pub_date__year"),
...         )
...         .order_by("-rating")
...         .values("rating")[:1]
...     ),
...     total_comments=Sum("number_of_comments"),
... )

The pk lookup shortcut¶

For convenience, Django provides a pk lookup shortcut, which stands for “primary key”.

In the example Blog model, the primary key is the id field, so these three statements are equivalent:

>>> Blog.objects.get(id__exact=14)  # Explicit form
>>> Blog.objects.get(id=14)  # __exact is implied
>>> Blog.objects.get(pk=14)  # pk implies id__exact

The use of pk isn’t limited to __exact queries – any query term can be combined with pk to perform a query on the primary key of a model:

# Get blogs entries with id 1, 4 and 7
>>> Blog.objects.filter(pk__in=[1, 4, 7])

# Get all blog entries with id > 14
>>> Blog.objects.filter(pk__gt=14)

pk lookups also work across joins. For example, these three statements are equivalent:

>>> Entry.objects.filter(blog__id__exact=3)  # Explicit form
>>> Entry.objects.filter(blog__id=3)  # __exact is implied
>>> Entry.objects.filter(blog__pk=3)  # __pk implies __id__exact

Escaping percent signs and underscores in LIKE statements¶

The field lookups that equate to LIKE SQL statements (iexact, contains, icontains, startswith, istartswith, endswith and iendswith) will automatically escape the two special characters used in LIKE statements – the percent sign and the underscore. (In a LIKE statement, the percent sign signifies a multiple-character wildcard and the underscore signifies a single-character wildcard.)

This means things should work intuitively, so the abstraction doesn’t leak. For example, to retrieve all the entries that contain a percent sign, use the percent sign as any other character:

>>> Entry.objects.filter(headline__contains="%")

Django takes care of the quoting for you; the resulting SQL will look something like this:

SELECT ... WHERE headline LIKE '%\%%';

Same goes for underscores. Both percentage signs and underscores are handled for you transparently.

Caching and QuerySets¶

Each QuerySet contains a cache to minimize database access. Understanding how it works will allow you to write the most efficient code.

In a newly created QuerySet, the cache is empty. The first time a QuerySet is evaluated – and, hence, a database query happens – Django saves the query results in the QuerySet’s cache and returns the results that have been explicitly requested (e.g., the next element, if the QuerySet is being iterated over). Subsequent evaluations of the QuerySet reuse the cached results.

Keep this caching behavior in mind, because it may bite you if you don’t use your QuerySets correctly. For example, the following will create two QuerySets, evaluate them, and throw them away:

>>> print([e.headline for e in Entry.objects.all()])
>>> print([e.pub_date for e in Entry.objects.all()])

That means the same database query will be executed twice, effectively doubling your database load. Also, there’s a possibility the two lists may not include the same database records, because an Entry may have been added or deleted in the split second between the two requests.

To avoid this problem, save the QuerySet and reuse it:

>>> queryset = Entry.objects.all()
>>> print([p.headline for p in queryset])  # Evaluate the query set.
>>> print([p.pub_date for p in queryset])  # Reuse the cache from the evaluation.

>>> queryset = Entry.objects.all()
>>> print(queryset[5])  # Queries the database
>>> print(queryset[5])  # Queries the database again

>>> queryset = Entry.objects.all()
>>> [entry for entry in queryset]  # Queries the database
>>> print(queryset[5])  # Uses cache
>>> print(queryset[5])  # Uses cache

>>> [entry for entry in queryset]
>>> bool(queryset)
>>> entry in queryset
>>> list(queryset)

Query iteration¶

The default way of iterating over a query – with for – will result in a blocking database query behind the scenes as Django loads the results at iteration time. To fix this, you can swap to async for:

async for entry in Authors.objects.filter(name__startswith="A"):
    ...

Be aware that you also can’t do other things that might iterate over the queryset, such as wrapping list() around it to force its evaluation (you can use async for in a comprehension, if you want it).

Because QuerySet methods like filter() and exclude() do not actually run the query – they set up the queryset to run when it’s iterated over – you can use those freely in asynchronous code. For a guide to which methods can keep being used like this, and which have asynchronous versions, read the next section.

QuerySet and manager methods¶

Some methods on managers and querysets – like get() and first() – force execution of the queryset and are blocking. Some, like filter() and exclude(), don’t force execution and so are safe to run from asynchronous code. But how are you supposed to tell the difference?

While you could poke around and see if there is an a-prefixed version of the method (for example, we have aget() but not afilter()), there is a more logical way – look up what kind of method it is in the QuerySet reference.

In there, you’ll find the methods on QuerySets grouped into two sections:

• Methods that return new querysets: These are the non-blocking ones, and don’t have asynchronous versions. You’re free to use these in any situation, though read the notes on defer() and only() before you use them.

• Methods that do not return querysets: These are the blocking ones, and have asynchronous versions – the asynchronous name for each is noted in its documentation, though our standard pattern is to add an a prefix.

Using this distinction, you can work out when you need to use asynchronous versions, and when you don’t. For example, here’s a valid asynchronous query:

user = await User.objects.filter(username=my_input).afirst()

filter() returns a queryset, and so it’s fine to keep chaining it inside an asynchronous environment, whereas first() evaluates and returns a model instance – thus, we change to afirst(), and use await at the front of the whole expression in order to call it in an asynchronous-friendly way.

Transactions¶

Transactions are not currently supported with asynchronous queries and updates. You will find that trying to use one raises SynchronousOnlyOperation.

If you wish to use a transaction, we suggest you write your ORM code inside a separate, synchronous function and then call that using sync_to_async – see Asynchronous support for more.

Storing and querying for None¶

As with other fields, storing None as the field’s value will store it as SQL NULL. While not recommended, it is possible to store JSON scalar null instead of SQL NULL by using Value(None, JSONField()).

Whichever of the values is stored, when retrieved from the database, the Python representation of the JSON scalar null is the same as SQL NULL, i.e. None. Therefore, it can be hard to distinguish between them.

This only applies to None as the top-level value of the field. If None is inside a list or dict, it will always be interpreted as JSON null.

When querying, None value will always be interpreted as JSON null. To query for SQL NULL, use isnull:

>>> Dog.objects.create(name="Max", data=None)  # SQL NULL.
<Dog: Max>
>>> Dog.objects.create(name="Archie", data=Value(None, JSONField()))  # JSON null.
<Dog: Archie>
>>> Dog.objects.filter(data=None)
<QuerySet [<Dog: Archie>]>
>>> Dog.objects.filter(data=Value(None, JSONField()))
<QuerySet [<Dog: Archie>]>
>>> Dog.objects.filter(data__isnull=True)
<QuerySet [<Dog: Max>]>
>>> Dog.objects.filter(data__isnull=False)
<QuerySet [<Dog: Archie>]>

Unless you are sure you wish to work with SQL NULL values, consider setting null=False and providing a suitable default for empty values, such as default=dict.

Key, index, and path transforms¶

To query based on a given dictionary key, use that key as the lookup name:

>>> Dog.objects.create(
...     name="Rufus",
...     data={
...         "breed": "labrador",
...         "owner": {
...             "name": "Bob",
...             "other_pets": [
...                 {
...                     "name": "Fishy",
...                 }
...             ],
...         },
...     },
... )
<Dog: Rufus>
>>> Dog.objects.create(name="Meg", data={"breed": "collie", "owner": None})
<Dog: Meg>
>>> Dog.objects.filter(data__breed="collie")
<QuerySet [<Dog: Meg>]>

Multiple keys can be chained together to form a path lookup:

>>> Dog.objects.filter(data__owner__name="Bob")
<QuerySet [<Dog: Rufus>]>

If the key is an integer, it will be interpreted as an index transform in an array:

>>> Dog.objects.filter(data__owner__other_pets__0__name="Fishy")
<QuerySet [<Dog: Rufus>]>

If the key you wish to query by clashes with the name of another lookup, use the contains lookup instead.

To query for missing keys, use the isnull lookup:

>>> Dog.objects.create(name="Shep", data={"breed": "collie"})
<Dog: Shep>
>>> Dog.objects.filter(data__owner__isnull=True)
<QuerySet [<Dog: Shep>]>

Containment and key lookups¶

>>> Dog.objects.create(name="Rufus", data={"breed": "labrador", "owner": "Bob"})
<Dog: Rufus>
>>> Dog.objects.create(name="Meg", data={"breed": "collie", "owner": "Bob"})
<Dog: Meg>
>>> Dog.objects.create(name="Fred", data={})
<Dog: Fred>
>>> Dog.objects.filter(data__contains={"owner": "Bob"})
<QuerySet [<Dog: Rufus>, <Dog: Meg>]>
>>> Dog.objects.filter(data__contains={"breed": "collie"})
<QuerySet [<Dog: Meg>]>

>>> Dog.objects.create(name="Rufus", data={"breed": "labrador", "owner": "Bob"})
<Dog: Rufus>
>>> Dog.objects.create(name="Meg", data={"breed": "collie", "owner": "Bob"})
<Dog: Meg>
>>> Dog.objects.create(name="Fred", data={})
<Dog: Fred>
>>> Dog.objects.filter(data__contained_by={"breed": "collie", "owner": "Bob"})
<QuerySet [<Dog: Meg>, <Dog: Fred>]>
>>> Dog.objects.filter(data__contained_by={"breed": "collie"})
<QuerySet [<Dog: Fred>]>

>>> Dog.objects.create(name="Rufus", data={"breed": "labrador"})
<Dog: Rufus>
>>> Dog.objects.create(name="Meg", data={"breed": "collie", "owner": "Bob"})
<Dog: Meg>
>>> Dog.objects.filter(data__has_key="owner")
<QuerySet [<Dog: Meg>]>

>>> Dog.objects.create(name="Rufus", data={"breed": "labrador"})
<Dog: Rufus>
>>> Dog.objects.create(name="Meg", data={"breed": "collie", "owner": "Bob"})
<Dog: Meg>
>>> Dog.objects.filter(data__has_keys=["breed", "owner"])
<QuerySet [<Dog: Meg>]>

>>> Dog.objects.create(name="Rufus", data={"breed": "labrador"})
<Dog: Rufus>
>>> Dog.objects.create(name="Meg", data={"owner": "Bob"})
<Dog: Meg>
>>> Dog.objects.filter(data__has_any_keys=["owner", "breed"])
<QuerySet [<Dog: Rufus>, <Dog: Meg>]>

One-to-many relationships¶

>>> e = Entry.objects.get(id=2)
>>> e.blog  # Returns the related Blog object.

>>> e = Entry.objects.get(id=2)
>>> e.blog = some_blog
>>> e.save()

>>> e = Entry.objects.get(id=2)
>>> e.blog = None
>>> e.save()  # "UPDATE blog_entry SET blog_id = NULL ...;"

>>> e = Entry.objects.get(id=2)
>>> print(e.blog)  # Hits the database to retrieve the associated Blog.
>>> print(e.blog)  # Doesn't hit the database; uses cached version.

>>> e = Entry.objects.select_related().get(id=2)
>>> print(e.blog)  # Doesn't hit the database; uses cached version.
>>> print(e.blog)  # Doesn't hit the database; uses cached version.

>>> b = Blog.objects.get(id=1)
>>> b.entry_set.all()  # Returns all Entry objects related to Blog.

# b.entry_set is a Manager that returns QuerySets.
>>> b.entry_set.filter(headline__contains="Lennon")
>>> b.entry_set.count()

>>> b = Blog.objects.get(id=1)
>>> b.entries.all()  # Returns all Entry objects related to Blog.

# b.entries is a Manager that returns QuerySets.
>>> b.entries.filter(headline__contains="Lennon")
>>> b.entries.count()

from django.db import models


class Entry(models.Model):
    # ...
    objects = models.Manager()  # Default Manager
    entries = EntryManager()  # Custom Manager


b = Blog.objects.get(id=1)
b.entry_set(manager="entries").all()

b.entry_set(manager="entries").is_published()

from django.db.models import Prefetch

prefetch_manager = Prefetch("entry_set", queryset=Entry.entries.all())
Blog.objects.prefetch_related(prefetch_manager)

b = Blog.objects.get(id=1)
b.entry_set.set([e1, e2])

Many-to-many relationships¶

Both ends of a many-to-many relationship get automatic API access to the other end. The API works similar to a “backward” one-to-many relationship, above.

One difference is in the attribute naming: The model that defines the ManyToManyField uses the attribute name of that field itself, whereas the “reverse” model uses the lowercased model name of the original model, plus '_set' (just like reverse one-to-many relationships).

An example makes this easier to understand:

e = Entry.objects.get(id=3)
e.authors.all()  # Returns all Author objects for this Entry.
e.authors.count()
e.authors.filter(name__contains="John")

a = Author.objects.get(id=5)
a.entry_set.all()  # Returns all Entry objects for this Author.

Like ForeignKey, ManyToManyField can specify related_name. In the above example, if the ManyToManyField in Entry had specified related_name='entries', then each Author instance would have an entries attribute instead of entry_set.

Another difference from one-to-many relationships is that in addition to model instances, the add(), set(), and remove() methods on many-to-many relationships accept primary key values. For example, if e1 and e2 are Entry instances, then these set() calls work identically:

a = Author.objects.get(id=5)
a.entry_set.set([e1, e2])
a.entry_set.set([e1.pk, e2.pk])

One-to-one relationships¶

One-to-one relationships are very similar to many-to-one relationships. If you define a OneToOneField on your model, instances of that model will have access to the related object via an attribute of the model.

For example:

class EntryDetail(models.Model):
    entry = models.OneToOneField(Entry, on_delete=models.CASCADE)
    details = models.TextField()


ed = EntryDetail.objects.get(id=2)
ed.entry  # Returns the related Entry object.

The difference comes in “reverse” queries. The related model in a one-to-one relationship also has access to a Manager object, but that Manager represents a single object, rather than a collection of objects:

e = Entry.objects.get(id=2)
e.entrydetail  # returns the related EntryDetail object

If no object has been assigned to this relationship, Django will raise a DoesNotExist exception.

Instances can be assigned to the reverse relationship in the same way as you would assign the forward relationship:

e.entrydetail = ed

How are the backward relationships possible?¶

Other object-relational mappers require you to define relationships on both sides. The Django developers believe this is a violation of the DRY (Don’t Repeat Yourself) principle, so Django only requires you to define the relationship on one end.

But how is this possible, given that a model class doesn’t know which other model classes are related to it until those other model classes are loaded?

The answer lies in the app registry. When Django starts, it imports each application listed in INSTALLED_APPS, and then the models module inside each application. Whenever a new model class is created, Django adds backward-relationships to any related models. If the related models haven’t been imported yet, Django keeps tracks of the relationships and adds them when the related models eventually are imported.

For this reason, it’s particularly important that all the models you’re using be defined in applications listed in INSTALLED_APPS. Otherwise, backwards relations may not work properly.

Queries over related objects¶

Queries involving related objects follow the same rules as queries involving normal value fields. When specifying the value for a query to match, you may use either an object instance itself, or the primary key value for the object.

For example, if you have a Blog object b with id=5, the following three queries would be identical:

Entry.objects.filter(blog=b)  # Query using object instance
Entry.objects.filter(blog=b.id)  # Query using id from instance
Entry.objects.filter(blog=5)  # Query using id directly