Overview

This tool helps you to keep track and evolve your database schema (SQL).

It works like this. Whenever you make changes to the database schema you put them into a “migration” - simply an .sql file with a sequence of SQL statements. Sqlturk keeps track of which migrations have been applied and runs them in a guaranteed order.

Design

There are plenty of tools to approach the problem of database schema migration. Here is how sqlturk is different:

• one-way migrations. There is no way to “rollback” a migration. In my experience, “downgrade” scripts are almost never used yet writing them adds considerable burden onto the developer. Their presense may add a false sense of security – since they used so rarely there is a big chance they will not work properly anyway.

• Python-agnostic. Sqlturk operates in terms of SQL DDL. It does not work in terms of SQLAlchemy ORM models or Django’s and thus it can be used in a much wider context. Plain SQL also means that a competent DBA who knows nothing about Python can author these scripts. Or they could be generated automatically by another tool.

• no magic. I wanted a tool that works in an obvious, easy-to-understand way. Changing database schema can have big consequences so it is important to be able to see what’s going on.

Not convinced? A couple of more sophisticated tools: Django South and sqlalchemy-migrate.

API usage

Short example:

>>> from sqlturk.migration import MigrationTool
>>> tool = MigrationTool('sqlite:///:memory:', migration_dir='testmigrations')
>>> tool.install() # create a database table to track schema changes
>>> tool.find_migrations()
['1_foobar', '2_foobar_data', '10_foobar_delete']
>>> tool.run_migrations()
>>> tool.find_migrations() # check that all migrations have been applied
[]

Example migration scripts can be found in the sqlturk’ source. Sqlturk uses SQLAlchemy syntax to specify database connection string, see its documentation for details.