pavlova: simplified deserialization using dataclasses

pavlova is a library that assists in mapping an unknown input into a dataclass.

from datetime import datetime
from dataclasses import dataclass

from pavlova import Pavlova


@dataclass
class Input:
    id: int
    name: str
    date: datetime


Pavlova().from_mapping({
    'id': 10,
    'name': 100
    'date': '2018-08-10',
}, Input)
# Input(id=10, name='100', date=datetime.datetime(2018, 8, 10, 0, 0))

Pavlova was born out of frustration with the lack of typing support for existing deserialization libraries. With the introduction of dataclasses in Python 3.7, they seemed like the perfect use for defining a deserialization schema.

Supported functionality

Parsing of booleans, datetimes, floats, ints, strings, decimals, dictionaries, enums, lists are currently supported.

There are more parsers to come, however to implement your own custom parser, simply implement PavlovaParser in pavlova.parsers, and register it with the Pavlova object with the register_parser method.

Installation

pip install pavlova

Usage with Flask

from dataclasses import dataclass, asdict

from flask import Flask, jsonify
from pavlova.flask import FlaskPavlova

pavlova = FlaskPavlova()
app = Flask(__name__)

@dataclass
class SampleInput:
    id: int
    name: str

@app.route('/post', methods=['POST'])
@pavlova.use(SampleInput)
def data(data: SampleInput):
    data.id = data.id * len(data.name)
    return jsonify(asdict(data))


app.run()

Adding Custom Types

There are a couple of different ways to implement new types for parsing in pavlova. In general, the process is to add a parser a specific type. For validation you should raise a TypeError or ValueError.

The first one, is creating a new type that extends an existing base type. Here is an example on how to implement an Email type, which is a string but performs validation.

from pavlova import Pavlova
from pavlova.parsers import GenericParser

class Email(str):
    def __new__(cls, input_value: typing.Any) -> str:
        if isinstance(input_value, str):
            if '@' in input_value:
                return str(input_value)
            raise ValueError()
        raise TypeError()

pavlova = Pavlova()
pavlova.register_parser(Email, GenericParser(pavlova, Email))

Another way, is to implement your own pavlova parser, rather than using your the built in GenericParser parser.

import datetime
from typing import Any, Tuple

import dateparser
from pavlova import Pavlova
from pavlova.parsers import PavlovaParser

class DatetimeParser(PavlovaParser[datetime.datetime]):
    "Parses a datetime"

    def parse_input(self,
                    input_value: Any,
                    field_type: Type,
                    path: Tuple[str, ...]) -> datetime.datetime:
        return dateparser.parse(input_value)

pavlova = Pavlova()
pavlova.register_parser(datetime.DateTime, DatetimeParser(pavlova))

Requirements

Pavlova is only supported on Python 3.6 and higher. With Python 3.6, it will install the dataclasses module. With Python 3.7 and higher, it will use the built-in dataclasses module.

License

GNU LGPLv3. Please see LICENSE and COPYING.LESSER.