rambler

A simple and language-independent SQL schema migration tool

Building

Automatic

Push a tag in the form vMAJOR.MINOR.PATCH and watch the actions take over.

Manual

Requires to be logged into the AWS account containing the ECR repository.

VERSION=x.y.z docker buildx build --push --platform linux/arm64,linux/amd64 -t wizehire/rambler:$VERSION -t wizehire/rambler:latest .

Usage

Migrations

In rambler, migrations are kept in the simplest form possible: a migration is a list of sections (up and down), each section being an SQL statement. Example:

-- rambler up

CREATE TABLE foo (
	id INTEGER UNSIGNED AUTO_INCREMENT,
	bar VARCHAR(60),
	PRIMARY KEY (id)
);

-- rambler down

DROP TABLE foo;

Sections are delimited by SQL comments suffixed by the rambler marker (white-spaces sensitive). While applying a migration, rambler will execute each up section in order, and while reversing it it will execute each down section in reverse order.

Migrations are executed in alphabetical order, thus a versioning scheme of the form version_description.sql is highly recommended, version being an integer value, and description an underscored string. Examples:

• 201409272258_Added_table_foo.sql
• 01_First_migration.sql

The migrations applied to the database are stored in a table named migration (can be changed with the table configuration option).

Configuration

Rambler configuration is lightweight: just dump the credentials of your database and the path to your migrations' directory into a JSON file, and you're done. Here is an example or JSON configuration file with the default values of rambler:

{
	"database": "",
	"directory": ".",
	"driver": "mysql",
	"host": "localhost",
	"password": "",
	"port": 3306,
	"protocol": "tcp",
	"schema": "public",
	"sslmode": "disable"
	"table": "migrations",
	"user": "root",
}

When running, rambler will try to find a configuration file in the working directory and use its values to connect to the managed database.

HJSON

Rambler supports HJSON configuration files, which is by the way retrocompatible with JSON.

Environment Variables

Alternatively, Rambler can read configuration from environment variables. The environment variables can override any of the confifuration file values and are prefixed with RAMBLER_.

Env Var	Config
RAMBLER_DATABASE	database
RAMBLER_DIRECTORY	directory
RAMBLER_DRIVER	driver
RAMBLER_HOST	host
RAMBLER_PASSWORD	password
RAMBLER_PORT	port
RAMBLER_PROTOCOL	protocol
RAMBLER_SCHEMA	schema
RAMBLER_SSLMODE	sslmode
RAMBLER_TABLE	table
RAMBLER_USER	user

Drivers

Rambler supports actually 3 drivers:

• mysql
• postgresql
• sqlite

Don't hesitate to get in touch if you want to see another one supported, provided a golang database/sql driver exist for your database vendor.

Note that some configuration options only apply to some drivers.

Applying a migration

To apply a migration, use the apply command.

rambler apply

Rambler will compare the migrations already applied and the available migrations in increasing order to find the next migration to apply, then execute all its up sections' statements in order.

Reversing a migration

To reverse a migration, use the reverse command.

rambler reverse

Rambler will compare the migrations already applied and the available migrations in decreasing order to find the last applied migrations, then execute all its down sections' statements in reverse order.

Options

• --all, -a repeat the command until there is no more migration to apply/reverse. This flag is exclusive with --migration.
• --no-save doesn't save the applied/reversed migration. This option is mainly destined to allow faster iteration when writing the migration.
• --migration bypass the migration automatic discovery and apply the migration whose path is given. This is exclusive with --all.

Errors

To ensure database schema consistency, rambler will complain and stop when encountering a new migration in the middle of the already existing ones or if it can't find a migration already applied.

Environments

An environment is an additional configuration that is given a name, and can be used to create multiple configurations for a single application (for example, to differentiate production, testing, etc).

Environments are defined in the configuration file, under the environments item. Each environment is defined as an attribute of this item, the key being the name and the value being the configuration options.

Environments configuration are derived from the default configuration of rambler (at the configuration file's root), so you only need to override the needed options:

{
    "driver": "mysql",
    "protocol": "tcp",
    "host": "localhost",
    "port": 3306,
    "user": "root",
    "password": "",
    "database": "rambler_default",
    "directory": "migrations",
    "table": "migrations",
    "environments": {
        "development": {
            "database": "rambler_development"
        },
        "testing": {
            "database": "rambler_testing"
        }
    }
}

Here we have three environments defined:

• default, will use the rambler_default database,
• development, will use the rambler_development database,
• testing, will use the rambler_testing database;

Logging

Rambler will log a few important informations for monitoring what is happening on stdout. If you suspect something of being wrong, you can also use the debug mode by adding --debug to your command line.

Dry-run

The --dry-run flag will print statements instead of executing them.

CONTRIBUTORS

• cjhubert
• shawndellysse

Feedback and contributions

Feel free to give feedback, make pull requests or simply open issues if you find a bug or have an idea.