I have a model with a custom json serializer that performs some processing prior to dumping to json.
Now, when fetching a single obj i want to use the custom serializer from the model to fetch the entire object (with the processing mentioned above). When fetching a list i want to use the default serializer to fetch only the headers (render only the model fields).
I looked into three options
overriding obj_get
def obj_get(self, bundle, **kwargs):
obj = ComplexModel.objects.get(pk=kwargs['pk'])
return obj.to_serializable()
i got thrown with
{"error": "The object LONG JSON DUMP has an empty attribute 'description' and doesn't allow a default or null value."}
not sure why this is happening - the field description is nullable, Plus - why tastypie is checking validation for objects already in the database, and... while fetching ??
using dehydrate
def dehydrate(self, bundle):
return bundle.obj.to_serializable()
This is great but the cycle is executed before each object - so i cann't tell if I'm fetching a list or a single object. The result here is the full serizliazed objects whether it's a list or a single entry.
creating a custom serializer
class CustomComplexSerializer(Serializer):
def to_json(self, data, options=None):
if isinstance(data,ComplexModel):
data = data.to_serializable()
return super(CustomComplexSerializer,self).to_json(data)
Same problem here, when fetching one entry the serializer accepts the obj in data.obj, when it's fetching a list it accepts a dict (odd...). I can check if bundle is an instance of dict as well - but testing for the type of ComplexModel felt awkward enough.
So what is the best way to implement a custom serialization for fetching only a single entry ?
Just for future reference, I think i found the right way to do this and it's by using full_dehydrate.
def full_dehydrate(self, bundle, for_list=False):
if not for_list:
return bundle.obj.to_serializable()
return super(ReportResource,self).full_dehydrate(bundle,for_list)