This is a view for my project in which I am trying to upload images from my local system to s3 using boto.
class ImageList(generics.ListCreateAPIView):
queryset = Image.objects.all()
serializer_class = ImageSerializer
def post(self , request , format = None):
# import ipdb; ipdb.set_trace()
serializer = ImageSerializer(data = request.data)
if serializer.is_valid():
serializer.save()
print request.data
return Response({'received data' : request.data})
return Response(serializer.errors , status = status.HTTP_400_BAD_REQUEST)
class ImageDetail(generics.RetrieveUpdateDestroyAPIView):
queryset = Image.objects.all()
serializer_class = ImageSerializer
It works fine for django admin but generates error while I try to upload it using Django Rest Framework html form. I have overridden the post method in generic class based view. The error that gets generated is :
UnicodeDecodeError at /image/
'utf8' codec can't decode byte 0xff in position 15: invalid start byte
Request Method: POST
Request URL: http://127.0.0.1:8000/image/
Django Version: 1.9
Exception Type: UnicodeDecodeError
Exception Value:
'utf8' codec can't decode byte 0xff in position 15: invalid start byte
Exception Location: /System/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/json/encoder.py in encode, line 210
Python Executable: /usr/bin/python
Python Version: 2.7.10
Unicode error hint
The string that could not be encoded/decoded was: "����\u
I have tried almost everything available on stack overflow but nothing worked for me. I didn't copy the code from somewhere so there would be no strange characters.
For reference this is the models.py file:
class Image(models.Model):
image_meta = models.ForeignKey('Image_Meta',on_delete=models.CASCADE,)
image = models.ImageField(upload_to='images-data')
# image = models.URLField(max_length = 500)
order = models.IntegerField()
version = models.CharField(max_length=10)
def __unicode__(self):
return (self.image)
JSON strings are unicode strings, not binary strings. Your image contains binary data, and the JSON serializer is complaining about that.
One common approach to work around this problem is to use base64 or base85. Python has builtin support for those two (and other) encodings via the base64 module.