2. Deserialization

Deserialization is similar to serialization. Deserialization allows native Python datatypes to be converted to Python objects.

2.1. Defining our deserailizer

Let’s consider a similar example that we used in serialization example. Let’s assume we have a comment object and the comment object has a user attached to it, Now let’s define a deserializer class:

from pyserializer.serializers import Serializer
from pyserializer import fields


class UserDeserializer(Serializer):
    email = fields.CharField()
    username = fields.CharField()

    class Meta:
        fields = (
            'email',
            'username'
        )

    def __repr__(self):
        return '<User(%r)>' % (self.username)


class CommentDeserializer(Serializer):
    user = UserDeserializer()
    content = fields.CharField()
    created_date = fields.DateField(format='%Y-%m-%d')
    created_time = fields.DateTimeField(format='%Y-%m-%dT%H:%M:%SZ')

    def __repr__(self):
        return '<Comment(%r)>' % (self.content)

2.2. Deserialize the object

Let’s use the deserializer class we defined above to deserialize a Python dict:

data_dict = {
    'user': {
        'email': 'foo@example.com',
        'username': 'JohnSmith'
    },
    'content': 'foo bar',
    'created_date': '2015-01-01',
    'created_time': '2012-01-01T16:00:00Z'
}
deserializer = CommentDeserializer(data_dict=data_dict)
deserializer.object.user.username
'JohnSmith'
deserializer.object.created_time
datetime.datetime(2012, 1, 1, 16, 0)

Note

If your fields in the deserializer class has validators defined, the validators will run before deserializing the objects. If any error is encountered during the validation process, the deserializer.object will return None. You can check the error object on the deserializer (deserializer.errors) to get detailed information on the errors.