pysimplemodel 1.1.3

Data handling made easy

Simple Model

SimpleModel offers a simple way to handle data using classes instead of a plenty of lists and dicts.

It has simple objectives:

• Define your fields easily (just a tuple, nor dicts or instances of type classes whatever)

• Support for field validation

• Conversion to dict

That’s it. If you want something more complex there are plenty of libraries and frameworks that does a lot of cool stuff.

Table of Contents

• Simple Model

  • How to install

  • How to use

  • Validation

  • Cleaning

  • Build many models

  • Conversion to Dict

• Changes

  • 1.1.3 / 2018-27-02

  • 1.1.2 / 2018-21-02

  • 1.1.1 / 2018-15-02

  • 1.1.0 / 2018-15-02

  • 1.0.2 / 2018-01-10

  • 1.0.1 / 2018-01-10

  • 1.0.0 / 2018-01-10

  • 0.15.0 / 2017-19-12

  • 0.14.0 / 2017-21-11

  • 0.13.0 / 2017-21-11

How to install

pip install pysimplemodel

How to use

from simple_model import Model
from simple_model.exceptions import ValidationError


class Person(Model):
    age: int
    height: float
    name: str
    weight: float

    class Meta:
        allow_empty = ('height', 'weight')

    def clean_name(self, name):
        return name.strip()

    def validate_age(self, age):
        if age < 0 or age > 150:
            raise ValidationError('Invalid value for age "{!r}"'.format(age))

    def validate_height(self, height):
        if height <= 0:
            raise ValidationError('Invalid value for height "{!r}"'.format(age))

>>> person = Person(age=18.0, name='John Doe ')
>>> person.name
'John Doe '
>>> person.clean()
>>> person.name
'John Doe'
>>> person.age
18
>>> person.validate(raise_exception=False)
True
>>> dict(person)
{
    'age': 18,
    'height': '',
    'name': 'John Doe',
    'weight': '',
}

Validation

Model values aren’t validated until the validated method is called:

>>> person = Person()  # no exception
>>> person.validate()
...
EmptyField: name field cannot be empty
>>> person = Person(name='Jane Doe', age=60)
>>> person.validate()  # now it's ok!

You may change the validate method to return a boolean instead of raising an exception:

>>> person = Person()
>>> person.validate(raise_exception=False)
False
>>> person = Person(name='Jane Doe', age=60)
>>> person.validate(raise_exception=False)
True

Cleaning

Sometimes it is necessary to clean some values of your models, this can be easily done using simple-model:

class CleanPerson(Model):
    age: int
    name: str

    def clean_name(self, name):
        return name.strip()


>>> person = CleanPerson(name='John Doe  \n', age='10')
>>> person.name, person.age
('John Doe  \n', '10')
>>> person.clean()
>>> person.name, person.age
('John Doe', 10)

Build many models

It’s possible to build many models in a single step, it can be done by passing an iterable to the build_many method.

>>> people = [
    {'name': 'John Doe'},
    {'name': 'John Doe II'},
]
>>> models = Person.build_many(people)

Conversion to Dict

To convert to dict is pretty straight-forward task:

>>> person = Person(name='Jane Doe', age=60)
>>> dict(person)
{
    'age': 60,
    'height': None,
    'name': 'Jane Doe',
    'weight': None,
}

Simple model also supports dict conversion of nested models:

class SocialPerson(Model):
    friend: Person
    name: str


>>> person = Person(name='Jane Doe', age=60)
>>> other_person = SocialPerson(name='John Doe', friend=person)
>>> dict(other_person)
{
    'friend': {
        'age': 60,
        'height': None,
        'name': 'Jane Doe',
        'weight': None,
    },
    'name': 'John Doe',
}

It also supports nested models as lists:

import typing


class MoreSocialPerson(Model):
    friends: typing.List[Friend]
    name: str


>>> person = Person(name='Jane Doe', age=60)
>>> other_person = Person(name='John Doe', age=15)
>>> social_person = MoreSocialPerson(name='Foo Bar', friends=[person, other_person])
>>> dict(social_person)
{
    'name': 'Foo Bar',
    'friends': [
        {
            'age': 60,
            'height': None,
            'name': 'Jane Doe',
            'weight': None,
        },
        {
            'age': 15,
            'height': None,
            'name': 'John Doe',
            'weight': None,
        }
    ]
}

Changes

1.1.3 / 2018-27-02

• Fix model_many_builder to stop raising errors when empty iterable is received as argument

1.1.2 / 2018-21-02

• Fix field conversion to only happen when value is not None

• Raise exception when trying to convert field with invalid model type

• Fix model fields to stop including some methods and properties

1.1.1 / 2018-15-02

• Fix attribute default value as function so when the model receives the field value the default value is ignored

1.1.0 / 2018-15-02

• Fix setup.py long_description

• Allow models fields be defined with class attributes without typing

• Fix type conversion on fields using typing.List[...]

• Bugfix: remove Meta attribute from model class meta fields

• Fields attributes may receive function as default values. The function is executed (without passing arguments to it) on model instantiation

1.0.2 / 2018-01-10

• Add missing function name to __all__ on simple_model.__init__

1.0.1 / 2018-01-10

• Fix setup.py

1.0.0 / 2018-01-10

• Move model field customization to Meta class inside model

• Support field definition using type hints (python 3.6 only)

• Drop support for python 3.4 and 3.5

• Remove DynamicModel

• Add Changes file and automate versioning from parsing it

• Move main docs to sphinx

• Improve documentation

0.15.0 / 2017-19-12

• Use pipenv

• Drop python 3.3 support

0.14.0 / 2017-21-11

• Add model_many_builder(). It builds lists of models from data lists

• Fix travis config

0.13.0 / 2017-21-11

• Transfrom BaseModel.is_empty from an instance method to a class method

• Don’t raise an exception when BaseModel.build_many receives empty iterable. Instead returns another empty iterable