seagull

Getting Started

The basic workflow is as follows:

1. seagull init

# seagull.toml

Commands

seagull init

creates a seagull.toml file for storing connection strings and other config values. This file is optional, you can use the other commands without it.

seagull poop

firstly creates a directory named migrations if one does not already exist. Secondly creates an empty .sql file in the format V{1}__{2}.sql where {1} is an auto-incremented version number and {2} is a description.

USAGE
# creates ./migrations/V1__initial.sql
$ seagull poop initial

# creates ./migrations/V1_create_users_table.sql 
$ seagull poop "create users table"

# creates ./migrations/V2_another_migration.sql assuming V1 exists
$ seagull poop another_migration

seagull migrate

firstly creates a database table named __migration_history if one does not already exist. Runs all migrations in the migrations directory in a single transaction. If one fails, they all fail and the database is rolled back.

USAGE

# reads config from seagull.toml
$ seagull migrate

# specify your PostgreSQL connection string
$ seagull migrate --database postgresql://postgres:mysecretpassword@localhost/postgres

# looks for migrations in src/migrations
$ seagull migrate --dir src/migrations

seagull remigrate

Same as seagull migrate except that it will firstly reset the whole database before running all migrations. Useful for development if you're using a Docker database and changing migrations often. Would NOT suggest running it on production! :skull_and_crossbones:

# reads config from seagull.toml
$ seagull remigrate

# specify your PostgreSQL connection string
$ seagull remigrate --database postgresql://postgres:mysecretpassword@localhost/postgres

# looks for migrations in src/migrations
$ seagull remigrate --dir src/migrations