Generic Relations for Django Rest Framework
Project description
Rest Framework Generic Relations
This library implements Django REST Framework serializers to handle generic foreign keys.
Maintenance status
Originally by Lily Foote. now maintained by Craig de Stigter.
I don't intend to make any major changes to this library, but I will attempt to fix reported bugs and keep it up to date with recent versions of Python / Django / DRF.
Requirements
- At least Python 3.8
- A supported version of Django REST Framework (at least 3.12+)
- A supported version of Django (at least 3.2+)
Installation
Install using pip
...
pip install rest-framework-generic-relations
Add 'generic_relations'
to your INSTALLED_APPS
setting.
INSTALLED_APPS = (
...
'generic_relations',
)
API Reference
GenericRelatedField
This field serializes generic foreign keys. For a primer on generic foreign keys, first see: https://docs.djangoproject.com/en/dev/ref/contrib/contenttypes/
Let's assume a TaggedItem
model which has a generic relationship with other arbitrary models:
class TaggedItem(models.Model):
tag_name = models.SlugField()
content_type = models.ForeignKey(ContentType, on_delete=models.CASCADE)
object_id = models.PositiveIntegerField()
tagged_object = GenericForeignKey('content_type', 'object_id')
And the following two models, which may have associated tags:
class Bookmark(models.Model):
"""
A bookmark consists of a URL, and 0 or more descriptive tags.
"""
url = models.URLField()
tags = GenericRelation(TaggedItem)
class Note(models.Model):
"""
A note consists of some text, and 0 or more descriptive tags.
"""
text = models.CharField(max_length=1000)
tags = GenericRelation(TaggedItem)
Now we define serializers for each model that may get associated with tags.
class BookmarkSerializer(serializers.ModelSerializer):
class Meta:
model = Bookmark
fields = ('url',)
class NoteSerializer(serializers.ModelSerializer):
class Meta:
model = Note
fields = ('text',)
The model serializer for the TaggedItem
model could look like this:
from generic_relations.relations import GenericRelatedField
class TagSerializer(serializers.ModelSerializer):
"""
A `TaggedItem` serializer with a `GenericRelatedField` mapping all possible
models to their respective serializers.
"""
tagged_object = GenericRelatedField({
Bookmark: BookmarkSerializer(),
Note: NoteSerializer()
})
class Meta:
model = TaggedItem
fields = ('tag_name', 'tagged_object')
The JSON representation of a TaggedItem
object with name='django'
and its generic foreign key pointing at a Bookmark
object with url='https://www.djangoproject.com/'
would look like this:
{
"tagged_object": {
"url": "https://www.djangoproject.com/"
},
"tag_name": "django"
}
If you want to have your generic foreign key represented as hyperlink, simply use HyperlinkedRelatedField
objects:
class TagSerializer(serializers.ModelSerializer):
"""
A `Tag` serializer with a `GenericRelatedField` mapping all possible
models to properly set up `HyperlinkedRelatedField`s.
"""
tagged_object = GenericRelatedField({
Bookmark: serializers.HyperlinkedRelatedField(
queryset = Bookmark.objects.all(),
view_name='bookmark-detail',
),
Note: serializers.HyperlinkedRelatedField(
queryset = Note.objects.all(),
view_name='note-detail',
),
})
class Meta:
model = TaggedItem
fields = ('tag_name', 'tagged_object')
The JSON representation of the same TaggedItem
example object could now look something like this:
{
"tagged_object": "/bookmark/1/",
"tag_name": "django"
}
Writing to generic foreign keys
The above TagSerializer
is also writable. By default, a GenericRelatedField
iterates over its nested serializers and returns the value of the first serializer that is actually able to perform to_internal_value()
without any errors.
Note, that (at the moment) only HyperlinkedRelatedField
is able to serialize model objects out of the box.
The following operations would create a TaggedItem
object with it's tagged_object
property pointing at the Bookmark
object found at the given detail end point.
tag_serializer = TagSerializer(data={
'tag_name': 'python',
'tagged_object': '/bookmark/1/'
})
tag_serializer.is_valid()
tag_serializer.save()
If you feel that this default behavior doesn't suit your needs, you can subclass GenericRelatedField
and override its get_serializer_for_instance
or get_deserializer_for_data
respectively to implement your own way of decision-making.
GenericModelSerializer
Sometimes you may want to serialize a single list of different top-level things. For instance, suppose I have an API view that returns what items are on my bookshelf. Let's define some models:
from django.core.validators import MaxValueValidator
class Book(models.Model):
title = models.CharField(max_length=255)
author = models.CharField(max_length=255)
class Bluray(models.Model):
title = models.CharField(max_length=255)
rating = models.PositiveSmallIntegerField(
validators=[MaxValueValidator(5)],
)
Then we could have a serializer for each type of object:
class BookSerializer(serializers.ModelSerializer):
class Meta:
model = Book
fields = ('title', 'author')
class BluraySerializer(serializers.ModelSerializer):
class Meta:
model = Bluray
fields = ('title', 'rating')
Now we can create a generic list serializer, which delegates to the above serializers based on the type of model it's serializing:
bookshelf_item_serializer = GenericModelSerializer(
{
Book: BookSerializer(),
Bluray: BluraySerializer(),
},
many=True,
)
Then we can serialize a mixed list of items:
>>> bookshelf_item_serializer.to_representation([
Book.objects.get(title='War and Peace'),
Bluray.objects.get(title='Die Hard'),
Bluray.objects.get(title='Shawshank Redemption'),
Book.objects.get(title='To Kill a Mockingbird'),
])
[
{'title': 'War and Peace', 'author': 'Leo Tolstoy'},
{'title': 'Die Hard', 'rating': 5},
{'title': 'Shawshank Redemption', 'rating': 5},
{'title': 'To Kill a Mockingbird', 'author': 'Harper Lee'}
]
A few things you should note:
- Although
GenericForeignKey
fields can be set to any model object, theGenericRelatedField
only handles models explicitly defined in its configuration dictionary. - Reverse generic keys, expressed using the
GenericRelation
field, can be serialized using the regular relational field types, since the type of the target in the relationship is always known. - The order in which you register serializers matters as far as write operations are concerned.
- Unless you provide a custom
get_deserializer_for_data()
method, onlyHyperlinkedRelatedField
provides write access to generic model relations.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for rest-framework-generic-relations-2.2.0.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 17a1f8a129a1c72c8ddc82e9454ea09012f60c332778578aa6a5a3a7135c0a35 |
|
MD5 | 24a867b229dfe0fe5f9b048eec423211 |
|
BLAKE2b-256 | d058d5fda1dfac96a85b1b138bfafd2438c17083110879f814cb3a6c9986297a |
Hashes for rest_framework_generic_relations-2.2.0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | fea13c42501cf6ee93c951d38d47d2ea727d29f4348b5d551d763a024cf28660 |
|
MD5 | 25d407d34428d7493066c82ab809ca9d |
|
BLAKE2b-256 | ab38ef41228094d2bdc1665f4599ff353ca298c6548968b8eb45e283ad2974bd |