在 Django REST Framework 中使用 multipart/form-data 上传多个图像和嵌套 json

Question

我在解析视图集中的request.data时遇到问题。我有一个模型，可以根据产品添加多个图像。

我想从传入数据中分离图像，将产品数据发送到 ProductSerializer，然后将图像与产品数据一起发送到其序列化器并保存。

我有两个模型，就像这样：

def Product(models.Model):
    name = models.CharField(max_length=20)
    color = models.ForeignKey(Color, on_delete=models.CASCADE)

def Color(models.Model):
    name = models.CharField(max_length=15)

def ProductImage(models.Model):
    product = models.ForeignKey(Product, on_delete=models.CASCADE)
    image = models.ImageField(upload_to='product_pics/')

我想发送到产品（127.0.0.1:8000/products/）的请求就像：

{
    "name": "strawberry",
    "color": {
        "name": "red"
    },
    "productimage_set": [
        {"image": "<some_encode_image_data>"}
    ]
}

序列化器没有什么特别的，只是提取标签链接，所以我没有写它。如何发送多部分/表单数据以及如何在视图集中解析它？或者解决办法是什么？

Answer 1

我开发了一个解决方案。使用 Postman，我发送了包含多个图像、单个数据和嵌套数据的多部分/表单数据。

在我的模型文件中，我添加了 Tags 模型作为 ManyToManyField 作为示例，还添加了 django-taggit。表单数据将如图所示。

和models.py

class Product(models.Model):
    name = models.CharField(max_length=20, blank=True)
    tags = models.ManyToManyField(Tags)
    taggit = TaggableManager(blank=True)

class ProductImage(models.Model):
    product = models.ForeignKey(Product, on_delete=models.CASCADE)
    image = models.ImageField(upload_to='image_path/', null=True, blank=True)

class Tags(models.Model):
    name = models.CharField(max_length=15, blank=True)

要事第一；第一个数据未正确解析。作为解决方案，并在该答案的帮助下，我创建了这个自定义解析器：

class MultipartJsonParser(parsers.MultiPartParser):

    def parse(self, stream, media_type=None, parser_context=None):
        result = super().parse(
            stream,
            media_type=media_type,
            parser_context=parser_context
        )
        data = {}

        for key, value in result.data.items():
            if type(value) != str:
                data[key] = value
                continue
            if '{' in value or "[" in value:
                try:
                    data[key] = json.loads(value)
                except ValueError:
                    data[key] = value
            else:
                data[key] = value
        return parsers.DataAndFiles(data, result.files)

现在我们可以使用这个解析器和 Django REST 内置 JSONParser 来解析我们的数据。现在是时候构建我们​​的视图集了。

class ProductViewSet(ModelViewSet):
    queryset = Product.objects.all()
    serializer_class = ProductSerializer
    parser_classes = [MultipartJsonParser, JSONParser]

    def get_serializer_context(self):
        context = super(ProductViewSet, self).get_serializer_context()

        # appending extra data to context
        if len(self.request.FILES) > 0:
            context.update({
                'included_images': self.request.FILES
            })

        return context

    def create(self, request, *args, **kwargs):
        # Validating images with its own serializer, but not creating.
        # The adding process must be through Serializer.
        try:
            image_serializer = ProductImageSerializer(data=request.FILES)
            image_serializer.is_valid(raise_exception=True)
        except Exception:
            raise NotAcceptable(
                detail={
                    'message': 'Upload a valid image. The file you uploaded was either not '
                               'an image or a corrupted image.'}, code=406)

        # the rest of method is about the product serialization(with extra context), 
        # validation and creation.
        serializer = self.get_serializer(data=request.data)
        serializer.is_valid(raise_exception=True)
        self.perform_create(serializer)
        headers = self.get_success_headers(serializer.data)
        return Response(serializer.data, status=status.HTTP_201_CREATED, headers=headers)

class ProductImageViewSet(ModelViewSet):
    queryset = ProductImage.objects.all()
    serializer_class = ProductImageSerializer

class TagsViewSet(ModelViewSet):
    queryset = Tags.objects.all()
    serializer_class = TagsSerializer

让我们在这里检查一下。正如我在评论中提到的，图像文件将包含在 request.FILES 中。为此，我首先将数据发送到 ProductImageSerializer并对其进行验证。如果发生验证错误，该过程将停止，API 将发送一条错误消息作为响应。然后，我将数据发送到 ProductSerializer，并将图片信息附加到get_serializer_context方法中的上下文中。

我们已经完成了create方法，其他细节都写在代码中了。

最后，序列化器.py

from django.forms import ImageField as DjangoImageField

class TagsSerializer(HyperlinkedModelSerializer):
    class Meta:
    model = Tags
    fields = ['url', 'pk', 'name']

class ProductImageSerializer(HyperlinkedModelSerializer):
    class Meta:
        model = ProductImage
        fields = ['url', 'pk', 'product', 'image']
        # attention!!! if you not use this bottom line,
        # it will show error like "product required" and
        # indirectly our validation at ProductViewSet will raise error.
        extra_kwargs = {
            'product': {'required': False}
        }
    # we created Object-level custom validation because validation not working correctly.
    # when ProductImageSerializer get single image, everything just fine but
    # when it get multiple image, serializer is just passing all the files.
    def validate(self, attrs):
        default_error_messages = {
            'invalid_image':
                'Upload a valid image. The file you uploaded was either not an image or a corrupted image.',
        }
        # in here we're verifying image with using django.forms; Pillow not necessary !!
        for i in self.initial_data.getlist('image'):
            django_field = DjangoImageField()
            django_field.error_messages = default_error_messages
            django_field.clean(i)
        return attrs

class ProductSerializer(HyperlinkedModelSerializer, TaggitSerializer):
    tags = TagsSerializer(allow_null=True, many=True, required=False)
    # you can delete this line. If you delete it, it will appear as url in response.
    productimage_set = ProductImageSerializer(allow_null=True, many=True, required=False)
    taggit = TagListSerializerField(allow_null=True, required=False)

    class Meta:
        model = Product
        fields = ['url', 'pk', 'name', 'tags', 'taggit', 'productimage_set']

    def create(self, validated_data):
        # create product
        try:
            product_obj = Product.objects.create(
                name=validated_data['name']
            )
        except Exception:
            raise NotAcceptable(detail={'message': 'The request is not acceptable.'}, code=406)

        if 'included_images' in self.context:  # checking if key is in context
            images_data = self.context['included_images']
            for i in images_data.getlist('image'):
                ProductImage.objects.create(
                    product=product_obj,
                    image=i
                )

        # pop taggit and create
        if 'taggit' in validated_data:
            taggit_data = validated_data.pop('taggit')
            for taggit_data in taggit_data:
                taggit_obj, created = Tag.objects.get_or_create(name=taggit_data)
                product_obj.taggit.add(taggit_obj)

        # pop tags and create
        if 'tags' in validated_data:
            tags_data = validated_data.pop('tags')
            for tags_data in tags_data:
                for i in tags_data.items():
                    tags_obj, created = Tags.objects.get_or_create(name=i[1])
                    product_obj.tags.add(tags_obj)

        return product_obj

那么这里发生了什么？为什么我们要为图像创建额外的验证？虽然我不知道为什么，ImageSerializer 只对单个文件进行正确的验证。如果您尝试上传两个文件，甚至可以在图片旁边放一个电影，验证将不起作用。为了防止这种情况，我们使用django内置的形式按顺序验证图片；将.mp3格式更改为.jpg，尝试上传大文件，都不起作用。进行验证的东西是纯django。其他细节在代码中。

如果您按照我所说的进行所有操作，则响应将如下所示：

我想这会让大多数Postman用户感到高兴。我希望它有帮助。如果有什么引起您的注意，让我们在评论中见面。