Better way convert json to SQLAlchemy object

Question

These days I am learning SQLAlchemy. When I want to load an object from json and save it to MySQL, things get difficult because the fields in my model are more that 20 and I wonder whether there're better ways to do this.

My original code follows as an example:

class User(Base):
    __tablename__ = 'parent'
    id = Column(Integer, primary_key=True)
    name = Column(String)

json_str = """{"id": 1, "name": "yetship"}"""
obj = json.loads(json_str)
user = User(id=obj.get("id"), name=obj.get("name"))

It can work but as I add more fields, it gets awful.

Answer 1

If you have in your json file only fields that you can initialize your User from. Then you simply can do.

user = User(**obj)

**obj will unpack your dict object, so if have obj = {'id': 1, 'name': 'Awesome'}, User(**obj) will do like User(id=1, name='Awesome')

You can see the docs

NOTE: it's better to avoid using id as a variable|field, because 'id' in dir(__builtin__), so id in bultins

UPD

If you have in your json fields that don't belong to that model you can filter them out with dict comprehension

user = User(**{k:v for k, v in obj.items() if k in {'id', 'name'}})

UPD for UPD

It depends on the size of you obj, in most of the cases it's probably better (performance and clarity) to iterate over ('id', 'name') rather than obj.items():

user = User(**{k: obj[k] for k in ('id', 'name') if k in obj})