使用DRF将嵌套对象保存到ManyToMany字段

Question

型号：

class Thread(models.Model):
"""Thread model for messages."""
participants = models.ManyToManyField(to=User, verbose_name=_("Participants"))
created_at = models.DateTimeField(auto_now_add=True, verbose_name=_("Created"))
updated_at = models.DateTimeField(auto_now=True, verbose_name=_("Updated"))

我有这个序列化程序：

class UserSerializer(serializers.ModelSerializer):
"""Serialize User model to get data from ManyToMany field to use in Thread participants."""
class Meta:
model = User
fields = ("id", "email")

class ThreadSerializer(serializers.ModelSerializer):
"""Serializer for create, get, edit, delete Thread."""
participants = UserSerializer(read_only=True, many=True)
class Meta:
model = Thread
fields = ("id", "participants", "created_at", "updated_at")

和视图：

class ThreadListCreate(generics.ListCreateAPIView):
"""
Get threads list with participants details.
Creates a new thread with two participants.
"""
permission_classes = (AllowAny,)
queryset = Thread.objects.all()
serializer_class = ThreadSerializer

我需要在"；参与者"；，我该怎么办？此外，为了将新用户分配给"用户"；参与者"；我只需要键入"；id"；，类似下面的代码：

{
"participants": [
{"id": 1},
{"id": 4}
]
}

现在作为上面POST代码的结果，我得到了这个：

{
"id": 28,
"participants": [],
"created_at": "2022-02-16T18:22:17.524911Z",
"updated_at": "2022-02-16T18:22:17.524911Z"
}

Answer 1

1. 也许您想将视图集从ThreadListCreate重命名为ThreadListCreateViewSet(更可读(，也许您可以直接使用ModelSerailizer，因为它具有您所需要的所有功能
2. 最好只是拆分序列化程序(一个仅用于创建目的(

我会做一些类似的事情

ViewSet

class ThreadListCreateViewSet(ModelViewset):
permission_classes = (AllowAny,)
queryset = Thread.objects.all()
serializer_class = ThreadSerializer
create_serializer_class = CreateThreadSerializer
def get_serializer_class(self):
if self.action == "create":
return self.create_serilaizer_class
return self.serializer_class

序列化程序

class CreateThreadSerializer(serializers.ModelSerializer):
participants = UserSerializer(many=True)
def create(self, validated_data):
instance = Thread.objects.create()
for user in validated_data["participants"]:
instance.participants.add(user)
instance.save()
return instance

这就是我要做的。。。希望它对你有用。