如何通过Serializer从SimpleJWT自动填充SQLAlchemy模型的owner_id？

方案一：在视图中手动注入owner_id #

直接在视图的create方法里将当前用户ID添加到序列化器输入数据中，逻辑简单直接：

def create(self, request):
    # 复制请求数据，避免修改原始对象
    request_data = request.data.copy()
    # 注入当前用户ID到owner_id字段
    request_data['owner_id'] = request.user.id
    
    serializer = PlaceInfoSerializer(data=request_data)
    if serializer.is_valid():
        Session = sessionmaker(bind=engine)
        session = Session()
        place = PlaceInfoModel(**serializer.validated_data)
        session.add(place)
        session.commit()
        return Response(serializer.data, status=status.HTTP_201_CREATED)
    return Response(serializer.errors, status=status.HTTP_400_BAD_REQUEST)

同时建议把序列化器里的owner_id设为只读，防止前端传入字段覆盖自动填充的值：

class PlaceInfoSerializer(serializers.Serializer):
    name = serializers.CharField()
    address = serializers.CharField()
    rating = serializers.FloatField()
    image = serializers.CharField(required=False)
    owner_id = serializers.IntegerField(read_only=True)  # 设置为只读

方案二：通过序列化器上下文传递用户并自动填充 #

如果希望逻辑封装在序列化器中，可以将请求对象传入序列化器上下文，再在序列化器内处理owner_id填充：

1. 修改视图，创建序列化器时传入上下文：

def create(self, request):
    # 将request传入序列化器上下文
    serializer = PlaceInfoSerializer(data=request.data, context={'request': request})
    if serializer.is_valid():
        Session = sessionmaker(bind=engine)
        session = Session()
        place = PlaceInfoModel(**serializer.validated_data)
        session.add(place)
        session.commit()
        return Response(serializer.data, status=status.HTTP_201_CREATED)
    return Response(serializer.errors, status=status.HTTP_400_BAD_REQUEST)

2. 修改序列化器，用validate方法填充owner_id：

class PlaceInfoSerializer(serializers.Serializer):
    name = serializers.CharField()
    address = serializers.CharField()
    rating = serializers.FloatField()
    image = serializers.CharField(required=False)
    owner_id = serializers.IntegerField(required=False)  # 设为非必填

    def validate(self, attrs):
        # 从上下文获取当前用户ID并填充
        user = self.context['request'].user
        attrs['owner_id'] = user.id
        return attrs

或者重写序列化器的create方法，把SQLAlchemy的入库逻辑也移到序列化器中：

class PlaceInfoSerializer(serializers.Serializer):
    name = serializers.CharField()
    address = serializers.CharField()
    rating = serializers.FloatField()
    image = serializers.CharField(required=False)
    owner_id = serializers.IntegerField(read_only=True)

    def create(self, validated_data):
        # 从上下文获取用户ID并添加到数据中
        user = self.context['request'].user
        validated_data['owner_id'] = user.id
        
        Session = sessionmaker(bind=engine)
        session = Session()
        place = PlaceInfoModel(**validated_data)
        session.add(place)
        session.commit()
        return place

这种情况下视图的create方法可以简化为：

def create(self, request):
    serializer = PlaceInfoSerializer(data=request.data, context={'request': request})
    if serializer.is_valid():
        serializer.save()
        return Response(serializer.data, status=status.HTTP_201_CREATED)
    return Response(serializer.errors, status=status.HTTP_400_BAD_REQUEST)

关于CurrentUserDefault()失败的原因 #

CurrentUserDefault()需要序列化器上下文包含request对象，且通常要和HiddenField配合。如果想用这个方法，需要做以下修改：

from rest_framework.fields import CurrentUserDefault

class PlaceInfoSerializer(serializers.Serializer):
    # ...其他字段
    owner_id = serializers.HiddenField(
        default=serializers.CreateOnlyDefault(
            lambda obj: obj['request'].user.id
        )
    )

同时必须在创建序列化器时传入context={'request': request}，否则无法获取用户。不过这种方式在SQLAlchemy场景下不如前面的方案直接。

内容的提问来源于stack exchange，提问作者Ryan