alembic-sequeled

Alembic but using raw SQL.

Template repository for using alembic as a database migration solution.

Usage

• Create a repository based on this template
• Install the required dependencies (pip install -r requirements.txt)
• optional: Adapt the script.py.mako template
• optional: Edit the alembic configuration
• Export the database URL to the DATABASE_URL environment variable
• Start writing your migrations

How this repository works

Alembic is used to handle SQL-based database migrations.

It aims to provide a git-like interface to handle migrations, using concepts like branches, head, ...

Each migration must implement both an upgrade and a downgrade SQL script. Using this repository, you may write migrations in raw SQL scripts, both for the upgrade and downgrade operation.

In addition, you will have to generate a sha1 hash for each migration to ensure its integrity during the migration execution.

To summarize it, each revision is made of:

• a main script (migration_slug.py)
• a script directory (migration_slug)
• an upgrade and downgrade SQL script (migration_slug/upgrade.sql)
• the corresponding hashes (migration_slug/upgrade.sha1)

Creating a migration

First of all, make sure you have properly exporter the DATABASE_URL variable.

Afterwards, you will have to create a new migration script:

alembic revision -m "my first revision"

You will then have to create the upgrade and downgrade SQL migration scripts. Create the upgrade.sql and downgrade.sql files and write your queries.

Once you finished writing your scripts, you will generate the corresponding sha1 files using the command below.

sha1 upgrade.sql > upgrade.sha1