mappr - Easily convert between arbitrary types.

mappr’s goal is to make it as easy as possible to define custom converters between arbitrary types in python. It does not concern itself with validation or coercion. It only provides a simple way to define a mapping between two types + some extra functionality to automatically generate converters for simple cases (off by default).

Links

• Source Code

• Documentation

  • Contributing

  • Reference

Installation

$ pip install mappr

or if you’re using poetry.

$ poetry add mappr

Example

from dataclasses import dataclass

import mappr

# A few sample models to demonstrate mappr.
@dataclass
class User:
    username: str
    first_name: str
    last_name: str
    email: str


@dataclass
class Person:
    nick: str
    name: str
    email: str


@dataclass
class UserPublic:
    username: str
    first_name: str


# We can register a mapper from ``User`` to ``Person``. Fields not specified in
# mapping will be copied directly. The source type needs the have attributes
# that match the name, otherwise an exception is raised.
mappr.register(User, Person, mapping=dict(
    nick=lambda o: o.username,
    name=lambda o: f"{o.first_name} {o.last_name}",
))

# We can now create a an instance of ``User`` so we can test our converter.
user = User(
    username='john.doe',
    first_name='John',
    last_name='Doe',
    email='john.doe@example.com',
)

# This will use the converter registered above. To allow conversion in the
# reverse direction, you need to register the appropriate converter first.
# Each converter works only one way.
assert mappr.convert(Person, user) == Person(
    nick='john.doe',
    name='John Doe',
    email='john.doe@example.com',
)

# For simple conversions, where the target type attributes are a subset of
# source's attributes, we can just pass ``strict=False`` to let mappr create
# an ad-hoc converter. This will only work if the attribute names are
# exactly the same.
assert mappr.convert(UserPublic, user, strict=False) == UserPublic(
    username='john.doe',
    first_name='John',
)

More Documentation

• Contributing
• Reference Documentation