SQLAlchemy Model Faker

Generate SQLAlchemy models with fake data.

IMPORTANT: Documentation asumes previous knowledge on how to work with SQLAlchemy.

Install

pip install sqlalchemy_model_faker

Usage

The package expects a SQLAlchemy model that extends from declarative_base().

It reads the model columns and generates a fake value according with the column type.

Basic Usage

Let's create a Product model with a description and a price columns.

from sqlalchemy.ext.declarative import declarative_base

class Product(declarative_base()):
    __tablename__ = 'products'
    id = Column(Integer, primary_key=True, autoincrement=True)
    description = Column(Text)
    price = Column(Integer)

Use factory to create a fake Product model.

from sqlalchemy_model_faker import factory

product = factory(Product).make()

print(type(product.description)) # <class 'str'>
print(type(product.price)) # <class 'int'>

Use SQLAlchemy session to persist the product into the database.

Custom values

By passing a dict, you can force factory to use custom provided values.

Other column values will be set with fake data.

from sqlalchemy_model_faker import factory

product = factory(Product).make({'price': 288})

print(product.price) # 288

Specific fake types

Faker has methods to generate fake data in a specific format, like emails, addresses, IPs, etc.

The fake data types can be specified passing a dict with column names and fake data types.

from sqlalchemy_model_faker import factory

product = factory(Product).make(types={'description': 'email'})

# Emails have only 1 '@'
print(product.description.count('@')) # 1

# Emails have at least one '.'
print(product.description.count('.') # >= 1

Custom values and fake types can be passed together.

from sqlalchemy_model_faker import factory

product = factory(Product).make({'price': 288}, types={'description': 'email'})

print(product.price) # 288
print(product.description) # valid email string

Ignoring columns

Columns might be ignored. Their generated value will be None.

Other column values will be set with fake data.

from sqlalchemy_model_faker import factory

product = factory(Product).make(ignored_columns=['price'])

print(product.price) # None

Custom Faker instance

A custom faker instance can be passed to the factory constructor.

This is useful to extend Faker, or replace it with a Mock when running tests.

from faker import Faker
from sqlalchemy_model_faker import factory

faker = Faker() # Extend Faker as needed or replace it with a Mock
product = factory(Product, faker).make()

# etc

License

This project is open-sourced software licensed under the MIT license.