Skip to main content

Fast and flexible object mapping (serialization, marshalling)

Project description

Lion

Lion is a very flexible yet fast library for mapping objects to dictionaries. It uses a declarative API and supports a query language similar to GraphQL. Unlike other serialization libraries it also allows to skip entire fields instead of having a null value in the dictionary.

It is inspired by libraries like serpy, marshmallow, Django REST Framework and Kim.

Example

import lion

class GroupMapper(lion.Mapper):
    id = lion.UUIDField()
    name = lion.StrField()

class UserMapper(lion.Mapper):
    id = lion.UUIDField()
    email = lion.StrField(condition=lion.skip_empty)
    first_name = lion.StrField()
    last_name = lion.StrField()
    groups = lion.ListField(GroupMapper)

user = User(
    id=UUID('ad94d0e8-2526-4d9b-ad76-0fbffcf41033'),
    email='john.doe@example.com',
    first_name='John',
    last_name='Doe',
    groups=[
        Group(
            id=UUID('95a326fc-32e5-4d9b-a385-1ea1257d98da'),
            name='Awesome people'
        )
    ]
)

# Dump all fields to a dictionary
assert UserMapper().dump(user) == {
    'id': 'ad94d0e8-2526-4d9b-ad76-0fbffcf41033',
    'email': 'john.doe@example.com',
    'first_name': 'John',
    'last_name': 'Doe',
    'groups': [
        {
            'id': '95a326fc-32e5-4d9b-a385-1ea1257d98da',
            'name': 'Awesome people'
        }
    ]
}

# Load user object from a dictionary
assert user == UserMapper().load({
    'id': 'ad94d0e8-2526-4d9b-ad76-0fbffcf41033',
    'email': 'john.doe@example.com',
    'first_name': 'John',
    'last_name': 'Doe',
    'groups': [
        {
            'id': '95a326fc-32e5-4d9b-a385-1ea1257d98da',
            'name': 'Awesome people'
        }
    ]
})

Query language

By using the GraphQL-like query language it is possible to dump and load only parts of a given structure:

# Dump a subset of fields
assert UserMapper('{id,email}').dump(user) == {
    'id': 'ad94d0e8-2526-4d9b-ad76-0fbffcf41033',
    'email': 'john.doe@example.com'
}

# Dump subset of a nested mapper
assert UserMapper('{id,email,groups{id}}').dump(user) == {
    'id': 'ad94d0e8-2526-4d9b-ad76-0fbffcf41033',
    'email': 'john.doe@example.com',
    'groups': [
        'id': '95a326fc-32e5-4d9b-a385-1ea1257d98da'
    ]
}

Performance

The performance is somewhat slower than serpy but still far ahead of marshmallow and Django REST Framework.

Simple Benchmark

Complex Benchmark

Caveats

Lion also supports loading (serialization/marshalling) of data but currently does not perform any kind of validation. This is not a big deal if using Lion as part of a project which uses something like connexion which already performs validation using the provided OpenAPI specification file. Just be warned that loading an unvalidated data structure using Lion might result in somewhat weird looking data.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

lion-0.4.0.tar.gz (7.7 kB view hashes)

Uploaded Source

Built Distribution

lion-0.4.0-py3-none-any.whl (7.9 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page