matiz 0.1.4

Serves modern approach to data structures definition - with a minimal boilerplate.

matiz

Python's 3.7 dataclasses got maried with pyyaml.

Installation:

pip install matiz

Usage example:

import dataclasses
import matiz
from matiz import fields

@dataclasses.dataclass
class Some(matiz.ObsoleteAble):
    id: int = fields.Integer()
    mac: str = fields.MacAddress()
    ip: str = fields.IPv4()
    version: int = fields.Integer(default=411)

Now the Some object can be created as a pure python object (thanks to dataclass) it is equipped with __init__, __repr__, __eq__ and some other methods.

>>> obj = Some(123, 'aa:bb:cc:11:22:33', '192.168.11.6')
>>> obj
Some(id=123, mac='aa:bb:cc:11:22:33', ip='192.168.11.6', version=411)

>>> print(obj)
!iz/Some
id: 123
ip: 192.168.11.6
mac: aa:bb:cc:11:22:33
version: 411

It's attributes are regular python attributes:

>>> obj.id
123

>>> obj.mac
'aa:bb:cc:11:22:33'

>>> obj.ip
'192.168.11.6'

>>> obj.version = 0  # can be assigned
>>> print(repr(obj))
Ints(id=123, mac='aa:bb:cc:11:22:33', ip='192.168.11.6', version=0)

It can be also deserialized from yaml directly to the custom object:

>>> yaml_string = """
!iz/Some
id: 22001
ip: 12.13.14.15
mac: aa:bb:cc:11:22:33
version: 0
"""

>>> import yaml
>>> obj2 = yaml.load(yaml_string, Loader=yaml.Loader)
>>> repr(obj2)
Some(id=22001, mac='aa:bb:cc:11:22:33', ip='12.13.14.15', version=0)

To be further documented...