Assignables

Helper package for Flask and Flask-API

Third Party Library Requirements

• inflection

Introduction

This package is helper package for assigning values to SqlAlchemy db model objects and to get their dict representation so they can be sent as response from Flask-API endpoints. This package can be used without using SqlAlchemy, Flask and Flask-API.

Installation

pip install assignables

Usage

from assignables import Assignable

class SomeModel(db.Model, Assignable):
    id = db.Column(db.Integer, primary_key=True)
    username = db.Column(db.String(80), unique=True, nullable=False)
    email = db.Column(db.String(120), unique=True, nullable=False)

    def __init__(self, username=None, email=None):
        Assignable.__init__()
        self.username = username
        self.email = email

data = {
    "username": 'user123',
    "email": 'user@email.com'
}

obj = SomeModel()
obj.assign(data) # username and email will be assigned.

obj_dict = obj.get_json_dict() # obj_dict won't contain _sa_instance_state and Assignable class atributes.

Assignable will give your model two new methods:

1. assign(data_dict) - this method will assign coresponding atributes from respective key value pairs from data_dict.
2. get_json_dict() - this method will return objects dictionary.

Using assign method by default will not assign objects id, but will other atributes if they exist in data_dict. Method get_json_dict will not have _sa_instance_state key inside by default and atriubets Assignable class contains.

If you want to specify custom atributes for not assigning or not serializing you can do that: from assignables import Assignables

class SomeModel(db.Model, Assignable):
    id = db.Column(db.Integer, primary_key=True)
    username = db.Column(db.String(80), unique=True, nullable=False)
    email = db.Column(db.String(120), unique=True, nullable=False)

    def __init__(self, username=None, email=None):
        unassignables = ['id', 'email']
        unserializables = ['_sa_instance_state', 'email']
        Assignable.__init__(unassignables=unassignables, unserializables=unserializables)
        self.username = username
        self.email = email

data = {
    "username": 'user123',
    "email": 'user@email.com'
}

obj = SomeModel()
obj.assign(data) # username will only be assigned.

obj_dict = obj.get_json_dict() # obj_dict will not contain _sa_instance_state and email atributes.

If used like this, assign method will not assign id and email atributes. Dictinary returned by calling get_json_dict will not have keys _sa_instance_state and email.

You can also add you custom data validator - Validation will be executed inside assign method. If validation failed ValidationError exception is raised.

from assignables import DataValidator, Assignables


class CustomValidator(DataValidator):
    def validate(self, obj):
        # return True if validated, false otherwise.


class SomeModel(db.Model, Assignable):
    id = db.Column(db.Integer, primary_key=True)
    username = db.Column(db.String(80), unique=True, nullable=False)
    email = db.Column(db.String(120), unique=True, nullable=False)

    def __init__(self, username=None, email=None):
        Assignable.__init__(validator=CustomValidator())
        self.username = username
        self.email = email

data = {
        "username": 'user123',
        "email": 'user@email.com'
    }

obj = SomeModel()
obj.assign(data) # Data will be validate using you custom data validator.

If there is a missmatch between your atribute naming or you wish to specify a naming convetion for resulting dictionary there is a way.

data = {
        "Username": 'user123',
        "Email": 'user@email.com'
    }

obj = SomeModel()
obj.assign(data, under_score_data=True) # This will handle create snake case keys from data dictionary keys.
obj_dict = obj.get_json_dict(naming_convetion='camel_case') # Output dict would have camel case keys.

Options for naming_convention are:

1. camel_case
2. upper_camel_case
3. snake_case