Related
I'm trying to serializer two nested models linked by a foreing key:
class Category(models.Model):
sp = models.ForeignKey('species.Sp', on_delete=models.CASCADE, related_name='species_category')
category_name = models.CharField(max_length=50)
class Catch(models.Model):
weight = models.IntegerField()
category = models.ForeignKey('species.Category', on_delete=models.CASCADE,)
I know is possible to use depth option, but it serialize all the fields of the related model, for example:
class CatchesSerializer(serializers.ModelSerializer):
class Meta:
model = Catch
fields = ['id', 'weight', 'category', ]
depth = 1
returns
[
{
"id": 335,
"weight": 4710,
"category": {
"id": 1,
"category_name": "1",
"sp": 41
}
},
...
]
How is the way to serialize only certains fields of the related model? for example:
[
{
"id": 335,
"weight": 4710,
"category": {
"sp": 41,
}
},
...
]
Serializer can be nested, you can try:
class CategorySerializer(serializers.ModelSerializer):
class Meta:
model = Category
fields = ['sp']
class CatchesSerializer(serializers.ModelSerializer):
category = CategorySerializer()
class Meta:
model = Catch
fields = ['id', 'weight', 'category']
What I'm trying to do in Django REST Framework: Return only latest nested object in list for an object and return it as JSON, with the sub-object un-nested.
My models:
class BaseObject(models.Model):
name = models.TextField()
object_type = models.ForeignKey(ObjectType)
class ObjectStatus(models.Model):
baseobject_id = models.ForeignKey('objects.BaseObject', related_name='status')
object_status = models.IntegerField()
object_status_timestamp = models.DateTimeField()
My serializers:
class ObjectStatusSimplifiedSerializer(serializers.ModelSerializer): #helper serializer to simplify status objects
class Meta:
model = ObjectStatus
fields = ['object_status', 'object_status_timestamp']
class ObjectStatusListSerializer(serializers.ModelSerializer): #request for last status of several objects
status = ObjectStatusSimplifiedSerializer(many=True)
class Meta:
model = BaseObject
fields = ['id', 'name', 'object_type', 'status']
My current view:
class ObjectStatusListView(generics.ListCreateAPIView):
serializer_class = ObjectStatusListSerializer
def get_queryset(self):
queryset = BaseObject.objects.all()
id = self.request.query_params.getlist('id')
if id:
queryset = queryset.filter(id__in=id)
return queryset
Current URL:
url(r'^objectstatus/status/list$', views.ObjectStatusListView.as_view()),
So now, when going to, for example, [...]/objectstatus/status/list?id=9, the result I get looks like this:
[
{
"id": 9,
"name": "r5",
"object_type": "router",
"status": [
{
"object_status": 1,
"object_status_timestamp": "2019-10-24T09:40:15.605391Z"
},
{
"object_status": 2,
"object_status_timestamp": "2019-10-24T09:40:28.133296Z"
},
{
"object_status": 3,
"object_status_timestamp": "2019-10-24T09:40:40.829486Z"
},
{
"object_status": 1,
"object_status_timestamp": "2019-10-24T09:40:53.333332Z"
}
]
}
]
What I want is to display only the object status with the most recent timestamp.
Also, I can't figure out how to flatten the JSON object, like this:
[
{
"id": 9,
"name": "r5",
"object_type": "router",
"object_status": 1,
"object_status_timestamp": "2019-10-24T09:40:53.333332Z"
}
]
With the following serializer, you should get the desired output. We filter the status list and get only the latest one and then we flatten the structure as you need.
class ObjectStatusListSerializer(serializers.ModelSerializer): #request for last status of several objects
status = serializers.SerializerMethodField(read_only=True)
class Meta:
model = BaseObject
fields = ['id', 'name', 'object_type', 'status']
def get_status(self, obj):
return ObjectStatusSimplifiedSerializer(instance=obj.status.order_by('object_status_timestamp').first()).data
def to_representation(self, obj):
"""Move fields from status to main object representation."""
representation = super().to_representation(obj)
status_representation = representation.pop('status')
for key in status_representation:
representation[key] = status_representation[key]
return representation
you can try change serializer to like this. I assum your ObjectType have field is name for line code object_type.name
class ObjectStatusSimplifiedSerializer(serializers.ModelSerializer):
name = serializers.SerializerMethodField()
object_type = serializers.SerializerMethodField()
#staticmethod
def get_name(instance):
return instance.status.name
#staticmethod
def get_object_type(instance):
return instance.status.object_type.name
class Meta:
model = ObjectStatus
fields = ['id', 'name', 'object_type', 'object_status', 'object_status_timestamp']
class ObjectStatusListSerializer(serializers.ModelSerializer):
status = serializers.SerializerMethodField()
#staticmethod
def get_status(instance):
queryset = ObjectStatus.objects.filter(baseobject_id=instance).order_by('-object_status_timestamp')[:1]
if queryset.count():
return ObjectStatusSimplifiedSerializer(queryset, many=True).data
return []
class Meta:
model = BaseObject
fields = ['id', 'name', 'object_type', 'status']
Say, there are two models:
class Model1(models.Model):
r = models.CharField(max_length=200)
class Model2(models.Model):
p = models.CharField(max_length=200)
m = models.ForeignKey(Model1,on_delete=models.CASCADE)
The serializers are :
class Model1Serializer(serializers.Serializer):
class Meta:
model = Model1
fields = '__all__'
class Model2Serializer(serializers.Serializer):
class Meta:
model = Model2
fields = '__all__'
The given Model1 serializer returns the output as:
{
"id": 1,
"r": "r_value"
}
and model 2 serializer output is:
{
"id":1,
"p: "p_value",
"m": 1
}
The thing is that I also want the r value in the model2 serializer output. How to do that?
You need to specify new field with correct source - you can read more in docs.
Option 1: With class Serializer:
class Model2Serializer(serializers.Serializer):
id = serializers.IntegerField()
p = serializers.CharField()
r = serializers.CharField(source='m.r')
class Meta:
model = Model2
fields = '__all__'
Output: {'p': u'pppp', 'r': u'rrrrr', 'id': 1}
Option 2: With class ModelSerializer:
class Model2Serializer(serializers.ModelSerializer):
r = serializers.CharField(source='m.r')
class Meta:
model = Model2
fields = '__all__'
Output: {'p': u'pppp', 'r': u'rrrrr', 'm': 1L, u'id': 1}
Option 3: To include whole Model1:
class Model1Serializer(serializers.ModelSerializer):
class Meta:
model = Model1
fields = '__all__'
class Model2Serializer(serializers.Serializer):
m = Model1Serializer()
class Meta:
model = Model2
fields = '__all__'
Output: {'m': OrderedDict([(u'id', 1), ('r', u'rrrrr')])}
1. If you want r as attrs of m
define Serializer class of specific Relational Field
class Model2Serializer(serializers.ModelSerializer):
m = Model1Serializer()
class Meta:
model = Model2
fields = '__all__'
output:
[
{
"id": 1,
"m": {
"id": 1,
"r": "RED"
},
"p": "Light RED"
},
{
"id": 2,
"m": {
"id": 1,
"r": "RED"
},
"p": "Dark RED"
}
]
2. If you want just r Using ReadOnlyField
You can use ReadOnlyField
class Model2Serializer(serializers.ModelSerializer):
r = serializers.ReadOnlyField(source='m.r')
class Meta:
model = Model2
fields = '__all__'
output:
[
{
"id": 1,
"r": "RED",
"p": "Light RED",
"m": 1
},
{
"id": 2,
"r": "RED",
"p": "Dark RED",
"m": 1
}
]
3. If you want just r Using SerializerMethodField()
You can you SerializerMethodField and ist read only as well
class Model2Serializer(serializers.ModelSerializer):
r = serializers.SerializerMethodField()
class Meta:
model = Model2
fields = '__all__'
def get_r(self, instance):
return instance.m.r
output:
[
{
"id": 1,
"r": "RED",
"p": "Light RED",
"m": 1
},
{
"id": 2,
"r": "RED",
"p": "Dark RED",
"m": 1
}
]
Try this
class Model1Serializer(serializers.Serializer):
class Meta:
model = Model1
fields = '__all__'
class Model2Serializer(serializers.Serializer):
r = Model1Serializer(many=True, source="model1_set")
class Meta:
model = Model2
fields = '__all__'
DRF makes it quite easy if you already have a Model1Serializer for Model1:
class Model2Serializer(serializers.Serializer):
m = Model1Serializer()
class Meta:
model = Model2
fields = '__all__'
That should be it.
Simply use depth=1 in your serializer Meta. Like this:
class Model2Serializer(serializers.Serializer):
class Meta:
model = Model2
fields = '__all__'
depth = 1
This will go for a level of 1 into the Foreign keys that do exist in that model. (By the way, I don't recommend it for big models. But it's perfectly suitable for your situation)
I'm struggling to write a Django GET that returns the following looking response:
{
"lists": [
{
"id": "123",
"list_order": [
{
"id": "123_1",
"order": 1,
"list_id": "123",
"item_id": 9876,
"item": {
"id": 9876,
"name": "item1",
"location": "California"
}
},
{
"id": "123_2",
"order": 2,
"list_id": "123",
"item_id": 2484,
"item": {
"id": 2484,
"name": "item2",
"location": "California"
}
}
],
"updated_date": "2018-03-15T00:00:00Z"
}
]
}
Given a list_id, the response returns the basic information on the list ("id", "updated_date"), as well as the order of items in the list. Inside each item in the list order, it also grabs the related item details (nested in "item"). I'm able to get this response without the "item" details ("id", "name", "location" fields) and with no error:
{
"lists": [
{
"id": "123",
"list_order": [
{
"id": "123_1",
"order": 1,
"list_id": "123",
"item_id": 9876
},
{
"id": "123_2",
"order": 2,
"list_id": "123",
"item_id": 2484
}
],
"updated_date": "2018-03-15T00:00:00Z"
}
]
}
Again there is no error, and I can retrieve the first nested level without any issue. The problem is getting the "item" information to show within each "list_order". Below are my models, serializers, and views.
models.py
class Lists(models.Model):
id = models.CharField(null=False, primary_key=True, max_length=900)
updated_date = models.DateTimeField(blank=True, null=True)
class Meta:
managed = False
db_table = 'tbl_lists'
class Items(models.Model):
id = models.BigIntegerField(primary_key=True)
name = models.TextField(blank=True, null=True)
location = models.TextField(blank=True, null=True)
class Meta:
managed = False
db_table = 'tbl_items'
class ListOrder(models.Model):
id = models.CharField(null=False, primary_key=True, max_length=900)
list_id = models.ForeignKey(Lists, db_column='list_id', related_name='list_order')
item_id = models.ForeignKey(Items, db_column='item_id', related_name='items')
order = models.BigIntegerField(blank=True, null=True)
class Meta:
managed = False
db_table = 'tbl_list_order'
serializers.py
class ItemsSerializer(serializers.ModelSerializer):
class Meta:
model = Items
fields = '__all__'
class ListOrderSerializer(serializers.ModelSerializer):
item = ItemsSerializer(many=False, read_only=True)
class Meta:
model = ListOrder
fields = '__all__'
class ListsSerializer(serializers.ModelSerializer):
list_order = ListOrderSerializer(many=True, read_only=True)
class Meta:
model = Lists
fields = '__all__'
views.py
class ListsViewSet(generics.ListCreateAPIView):
"""
API endpoint that returns a list with its meta-information
"""
queryset = Lists.objects.all()
serializer_class = ListsSerializer
def get_queryset(self):
list_id = self.kwargs['list_id']
filters = [Q(id=list_id)]
return Lists.objects.filter(*filters)
def list(self, request, list_id):
queryset = self.get_queryset()
list_serializer = ListsSerializer(queryset, many=True)
return Response({ 'lists': list_serializer.data })
I'm pretty new to Django and like what it offers so far, though maybe I'm thinking of doing this in too much of a "SQL" way. I've read about select_related() and prefetch_related(), but not sure how I would apply it to this case. Any assistance is greatly appreciated and let me know if there's any other information I can provide.
In your ListOrderSerializer you are trying to serialize item. while in ListOrder model you used the field name item_id
Solution:
In ListOrderSerializer:
class ListOrderSerializer(serializers.ModelSerializer):
item_id = ItemsSerializer(many=False, read_only=True)
...
Given the following code I was wondering how to populate RecordsResource with each real record data:
models.py
class Record(models.Model):
content_type = models.ForeignKey(ContentType, editable=False, null=True)
user = models.ForeignKey(User, related_name='records')
issued = models.DateTimeField(auto_now_add=True)
date = models.DateField()
def save(self, *args, **kwargs):
if not self.content_type:
self.content_type = ContentType.objects.get_for_model(self.__class__)
super(Record, self).save(*args, **kwargs)
def as_leaf_class(self):
model = self.content_type.model_class()
if model == self.__class__:
return self
return model.objects.get(pk=self.id)
class Record1(Record):
# some fields
# ...
class RecordN(Record):
# some fields
api.py
class BaseModelResource(ModelResource):
class Meta(object):
authentication = ApiKeyPlusWebAuthentication()
authorization= Authorization()
cache = SimpleCache()
throttle = CacheDBThrottle(
throttle_at=350,
# 1 day
expiration=86400
)
if settings.DEBUG:
serializer = PrettyJSONSerializer()
def obj_create(self, bundle, request=None, **kwargs):
return super(BaseModelResource, self).obj_create(bundle, request, user=request.user)
def apply_authorization_limits(self, request, object_list):
return object_list.filter(user=request.user)
class BaseRecordResource(BaseModelResource):
class Meta(BaseModelResource.Meta):
filtering = {
'date': ALL
}
excludes = ['issued']
class RecordsResource(BaseRecordResource):
class Meta(BaseRecordResource.Meta):
resource_name = 'records'
queryset = Record.objects.all()
class Record1Resource(BaseRecordResource):
class Meta(BaseRecordResource.Meta):
resource_name = 'record1'
queryset = Record1.objects.all()
# ...
class RecordNResource(BaseRecordResource):
class Meta(BaseRecordResource.Meta):
resource_name = 'recordn'
queryset = RecordN.objects.all()
Ok, I just solved it. I've simplified the code.
Given the following code...
models.py
from django.db import models
from model_utils.managers import InheritanceManager
class Place(models.Model):
name = models.CharField(max_length=50)
address = models.CharField(max_length=80)
# https://github.com/carljm/django-model-utils#inheritancemanager
objects = InheritanceManager()
class Restaurant(Place):
custom_field = models.BooleanField()
class Bar(Place):
custom_field = models.BooleanField()
api.py
from core.models import Place, Restaurant, Bar
# http://django-tastypie.readthedocs.org/en/latest/cookbook.html#pretty-printed-json-serialization
from core.utils import PrettyJSONSerializer
from tastypie.resources import ModelResource
class PlaceResource(ModelResource):
class Meta:
queryset = Place.objects.select_subclasses()
resource_name = 'place'
serializer = PrettyJSONSerializer()
class RestaurantResource(ModelResource):
class Meta:
queryset = Restaurant.objects.all()
resource_name = 'restaurant'
serializer = PrettyJSONSerializer()
class BarResource(ModelResource):
class Meta:
queryset = Bar.objects.all()
resource_name = 'bar'
serializer = PrettyJSONSerializer()
Output
http://localhost:8000/api/v1/bar/?format=json
{
"meta": {
"limit": 20,
"next": null,
"offset": 0,
"previous": null,
"total_count": 1
},
"objects": [
{
"address": "dawdaw",
"custom_field": true,
"id": "1",
"name": "dwdwad",
"resource_uri": "/api/v1/bar/1/"
}
]
}
OK
http://localhost:8000/api/v1/restaurant/?format=json
{
"meta": {
"limit": 20,
"next": null,
"offset": 0,
"previous": null,
"total_count": 1
},
"objects": [
{
"address": "nhnhnh",
"custom_field": true,
"id": "2",
"name": "nhnhnh",
"resource_uri": "/api/v1/restaurant/2/"
}
]
}
OK
http://localhost:8000/api/v1/place/?format=json
{
"meta": {
"limit": 20,
"next": null,
"offset": 0,
"previous": null,
"total_count": 2
},
"objects": [
{
"address": "dawdaw",
"id": "1",
"name": "dwdwad",
"resource_uri": "/api/v1/place/1/"
},
{
"address": "nhnhnh",
"id": "2",
"name": "nhnhnh",
"resource_uri": "/api/v1/place/2/"
}
]
}
What I want to achieve
{
"meta": {
"limit": 20,
"next": null,
"offset": 0,
"previous": null,
"total_count": 2
},
"objects": [
{
"address": "dawdaw",
"custom_field": true,
"id": "1",
"name": "dwdwad",
"resource_uri": "/api/v1/bar/1/"
},
{
"address": "nhnhnh",
"custom_field": true,
"id": "2",
"name": "nhnhnh",
"resource_uri": "/api/v1/restaurant/2/"
}
]
}
Solution:
from core.models import Place, Restaurant, Bar
# http://django-tastypie.readthedocs.org/en/latest/cookbook.html#pretty-printed-json-serialization
from core.utils import PrettyJSONSerializer
from tastypie.resources import ModelResource
class RestaurantResource(ModelResource):
class Meta:
queryset = Restaurant.objects.all()
resource_name = 'restaurant'
serializer = PrettyJSONSerializer()
class BarResource(ModelResource):
class Meta:
queryset = Bar.objects.all()
resource_name = 'bar'
serializer = PrettyJSONSerializer()
class PlaceResource(ModelResource):
class Meta:
queryset = Place.objects.select_subclasses()
resource_name = 'place'
serializer = PrettyJSONSerializer()
def dehydrate(self, bundle):
# bundle.data['custom_field'] = "Whatever you want"
if isinstance(bundle.obj, Restaurant):
restaurant_res = RestaurantResource()
rr_bundle = restaurant_res.build_bundle(obj=bundle.obj, request=bundle.request)
bundle.data = restaurant_res.full_dehydrate(rr_bundle).data
elif isinstance(bundle.obj, Bar):
bar_res = BarResource()
br_bundle = bar_res.build_bundle(obj=bundle.obj, request=bundle.request)
bundle.data = bar_res.full_dehydrate(br_bundle).data
return bundle
In RecordsResource class, you need to add model field as well (see https://github.com/tomchristie/django-rest-framework/blob/master/djangorestframework/resources.py#L232-234)
class RecordsResource(BaseRecordResource):
model = Record
class Meta(BaseRecordResource.Meta):
resource_name = 'records'
queryset = Record.objects.all()
Explaining from the beginning:
There are three styles of inheritance that are possible in Django.
Often, you will just want to use the parent class to hold
information that you don't want to have to type out for each child
model. This class isn't going to ever be used in isolation, so
Abstract base classes are what you're after.
If you're subclassing an existing model (perhaps something from
another application entirely) and want each model to have its own
database table, Multi-table inheritance is the way to go.
Finally, if you only want to modify the Python-level behavior of a
model, without changing the models fields in any way, you can use
Proxy models.
The choice here is Multi-table inheritance
Multi-table inheritance
The second type of model inheritance supported by Django is when each model in the hierarchy is a model all by itself. Each model corresponds to its own database table and can be queried and created individually. The inheritance relationship introduces links between the child model and each of its parents (via an automatically-created OneToOneField) Ref
To go from Record to Recordx where 1 <= x <= n you do a_example_record = Record.objects,get(pk=3) and then check what type of Recordx it is by using something like below
if hasattr(a_example_record, 'record1'):
# ...
elif hasattr(a_example_record, 'record2'):
# ...
So now that we know how to get the children from the parent and we need to provide TastyPie with a queryset in its meta, you need to write a custom queryset backed by a custom manager on the Record model that takes all your records (More here Custom QuerySet and Manager without breaking DRY?), checks what type of child it is and appends it to a queryset or a list. You can read about appending here How to combine 2 or more querysets in a Django view?