README.md

poi

Syntax : poi migrate [prepare | up | down | redo | status | new <newMigration>] [--version <fileName> | --env ]

Commands

• prepare : Creates a schema_migrations table, a Migrations.hs file, and a Migrations directory in the current directory.
• new <newMigration> : Creates a new Migration file in the Migrations directory. By default it's a haskell file. Passing a --sql or an --yaml flag creates the respective migration files. The created haskell file is prefixed with the current timestamp.
• up : Runs the up code block present in the migration files that aren't run yet.
• down : Runs the down block present in the last run migration file.
• redo : Redoes the the last run migration.
• status : Lists all the migrations that aren't run yet and the last run migration.

Flags

• --hs : The default. Creates the following haskell migration file.

{-# LANGUAGE OverloadedString #-}

import Database.PostgreSQL.Simple

migrate :: (Query, Query)
migrate = (up, down)

up :: Query
up = undefined

down :: Query
down = undefined

-- The `Query` datatype is exported by postgresql-simple, which is basically a `String`.

• --yaml: Create a yaml migration file with up and down as keys

• --sql : Create a sql migration file, It doesn't contain an up and a down block, runs the same migration for both the commands.

• --version fuzzyFile : Fuzzy match a specific migration file to run migrations in it.

• --env environmentName: development is the default environment. If the flag is absent, it uses the APP_ENV environment variable.

Structure of schema_migrations table

The table contains one single column version, that stores the time stamp of the last run migration.

The database settings are read from a config.yml file that should be present in the same directory as the commands that are run. The config file should contain the following keys:

development:
  database:
    name: postgres
    username: postgres
    password: postgres
    host: localhost
    port: 5432

When run in the development environment, the migrate command creates a schema.sql file that dumps the entire database.