NAME

evolve - evolve your schema to the latest version

DESCRIPTION

Schema evolution is a critical part of any developers work - the process of changing the structure of a database overtime. However, while there are tools to help with this task, the are not many that integrate well with version control systems.

SchemaEvolution is a very basic command line application that helps automate this process, while allowing you to also track changes in version control.

Each change to the database is tracked in a single SQL file, which are ordered numerically. The database itself has a special column (schema_version.version by default) that maintains track of the current version of the database, and the evolve command simply applies the set of SQL files after this version.

TUTORIAL

Let's briefly use evolve to track a small project - a little application that has users and messages, and messages can be sent between users. The first thing we need to do is create our evolution.ini file. This configuration file tells evolve how to connect to our database.

    dsn = dbi:Pg:dbname=messaging
    username = messaging
    password = hello

The dsn string is the same format as used by DBI, in this case we're using the PostgresSQL driver to connect to the messaging database, with the credentials "messaging" and "hello"

Next, we need to add the version column into our database. By default this is the version column in the schema_version table.

    CREATE TABLE schema_version (
         version INTEGER NOT NULL DEFAULT 0
    );

Great! Now we can start writing some schema evolutions. The first thing we need in our system is support for users. Let's go with something basic. In our application directory, create the folder "evolutions" and start with our first definition, 1_user.sql

    CREATE TABLE 'user' (
        id SERIAL PRIMARY KEY,
        name TEXT NOT NULL
    );

Now, with our first schema evolution, all we have to do is run "evolve"