Getting Started

sillyORM is a frontend to the SQLAlchemy Core (https://docs.sqlalchemy.org/en/20/core/). It has been tested with SQLite and PostgreSQL, but is probably compatible with other DBMS supported by SQLAlchemy

Basic usage example

import tempfile  # to create a temporary SQLite database file
import sillyorm


# define a model, a model abstracts a table in the database
class Example(sillyorm.model.Model):
    _name = "example"  # database table name (sanitized) & name in environment

    # fields
    name = sillyorm.fields.String(length=255)
    date = sillyorm.fields.Date()


# Create the environment
tmpfile = tempfile.NamedTemporaryFile()
# PostgreSQL could be used here instead
registry = sillyorm.Registry(f"sqlite:///{tmpfile.name}")

# Register the model in the Registry
registry.register_model(Example)
# Resolve the inheritance in the registry
registry.resolve_tables()
# Create the model table & fields in the database
registry.init_db_tables()

# get an Environment object (imagine like a database cursor)
env = registry.get_environment(autocommit=True)

# start using the model
record = env["example"].create({"name": "hello world!"})
print(f"repr: {repr(record)}")
print(f"name: {record.name}")
print(f"date: {record.date}")

repr: example[1]
name: hello world!
date: None

How to integrate sillyORM into your application

First store the registry object somewhere where you can access it from whereever in your application you want to use sillyORM.

Now you can create your models and put them somewhere. To register them in the registry either call Registry.register_model for each model or to make it a little easier maybe use a decorator.

After all models have been registered call Registry.resolve_tables and then Registry.init_db_tables to initialize them in the database.

Then create Environment objects using Registry.get_environment as you wish to do operations on the database (they are sort of like database cursors and actually they hold one cursor each).

Further reading

Read Basic concepts. It explains far more about models, fields etc. than this guide.