How to Convert YAML to Dataclasses in Python

In this tutorial, you’ll learn various methods to convert YAML data into Python dataclasses.

We’ll explore both manual methods and multiple popular libraries that simplify this conversion.

 

 

Manual parsing

To manually convert YAML data into dataclasses, start by defining your dataclass structure and then parse the YAML content accordingly.

import yaml
from dataclasses import dataclass
from typing import List
yaml_data = """
users:
  - name: Ahmed
    age: 30
    email: ahmed@example.com
  - name: Fatma
    age: 25
    email: fatma@example.com
"""

@dataclass
class User:
    name: str
    age: int
    email: str

@dataclass
class UsersData:
    users: List[User]

data = yaml.safe_load(yaml_data)
users = [User(**user) for user in data['users']]
users_data = UsersData(users=users)
print(users_data)

Output:

UsersData(users=[User(name='Ahmed', age=30, email='ahmed@example.com'), User(name='Fatma', age=25, email='fatma@example.com')])

 

Using dacite

You can use the dacite library to simplify the conversion from dictionaries to dataclasses.

import yaml
from dataclasses import dataclass
from typing import List
from dacite import from_dict
yaml_data = """
users:
  - name: Sara
    age: 27
    email: sara@example.com
  - name: Karim
    age: 35
    email: karim@example.com
"""

@dataclass
class User:
    name: str
    age: int
    email: str

@dataclass
class UsersData:
    users: List[User]

data = yaml.safe_load(yaml_data)
users_data = from_dict(data_class=UsersData, data=data)
print(users_data)

Output:

UsersData(users=[User(name='Sara', age=27, email='sara@example.com'), User(name='Karim', age=35, email='karim@example.com')])

 

Using marshmallow-dataclass

You can use the marshmallow-dataclass to integrate data validation with dataclass conversion.

import yaml
from dataclasses import dataclass
from typing import List
import marshmallow_dataclass
yaml_data = """
users:
  - name: Mona
    age: 32
    email: mona@example.com
  - name: Tarek
    age: 29
    email: tarek@example.com
"""

@dataclass
class User:
    name: str
    age: int
    email: str

@dataclass
class UsersData:
    users: List[User]

UsersSchema = marshmallow_dataclass.class_schema(UsersData)
data_dict = yaml.safe_load(yaml_data)
users_data = UsersSchema().load(data_dict)
print(users_data)

Output:

UsersData(users=[User(name='Mona', age=32, email='mona@example.com'), User(name='Tarek', age=29, email='tarek@example.com')])

 

Using dataclass-wizard

Implement dataclass-wizard for easy serialization and deserialization between YAML and dataclasses.

import yaml
from dataclass_wizard import YAMLWizard
from typing import List
from dataclasses import dataclass
yaml_data = """
users:
- name: Youssef
  age: 40
  email: youssef@example.com
- name: Dina
  age: 26
  email: dina@example.com
"""

@dataclass
class User(YAMLWizard):
  name: str
  age: int
  email: str

@dataclass
class UsersData(YAMLWizard):
  users: List[User]
users_data = UsersData.from_yaml(yaml_data)
print(users_data)

Output:

UsersData(users=[User(name='Youssef', age=40, email='youssef@example.com'), User(name='Dina', age=26, email='dina@example.com')])