Skip to main content

Yet another Python migration tool

Project description

Simple, database-agnostic migration tool for Python applications. Inspired by node migrations.


The project is in beta now. Bugs and breaking changes may occur.


  • No specific database requirements, use it for anything you call database.

  • Pretty simple, just generate migration script and define self-explanatory up() and down() functions there.

  • Use imports in your migration scripts to load database bindings. Talk to the database the same way your application does.

  • Stores the sequence of already performed migrations. If the sequence does not match scripts in migrations directory (e.g. after merge), aborts and warns user.

  • TBD: Deeply configurable, including resources acquiring and releasing.


Only Python 3 is supported for now.


$ pip install migrations

Notice, this distribution provides package and executable script named migrate, so check if it does not mess with existing packages/scripts. Generally, you should neither install this tool globally, nor install several migration tools for one project.


usage: migrate [options] [action]

  up     [-h] [NAME|COUNT]   (default) perform COUNT migrations or till
                             given NAME (by default perform all available)
  down   [-h] [NAME|COUNT]   revert COUNT migrations or till
                             given NAME (by default revert one)
  create [-h]  NAME          create new migration file

  show   [-h]                print all migrations in chronological order

  -h, --help                 show this help message and exit
  -v, --version              show version and exit
  -d PATH, --migrations-dir PATH
                             directory where migrations are stored
  -s PATH, --state-file PATH
                             location of file which stores database state
  -t PATH, --template-file PATH
                             location of template file for new migrations

Each migration file must define functions up() and down() without required arguments.

Simple migration example:

import redis

db = redis.Redis(host='localhost', port=6379)

def up():
    db.rpush('used_libraries', 'migrations')

def down():
    db.rpop('used_libraries', 'migrations')

Current working directory is prepended to sys.path, so any import statement in migration file tries to find requested module in working directory at first. You can use this to manage database access for your both app and migrations with single piece of code. See an example. Let’s assume that in working directory we have module named db, which contains singleton object responsible for DB connection, for example PyMySQL Connection object.

from db import connection

def manage_cursor(action):
    def wrap():
        with connection.cursor() as cursor:
    return wrap

def up(cursor):
        "INSERT INTO used_libraries (`name`) VALUES ('migrations')"

def down(cursor):
        "DELETE FROM used_libraries WHERE `name`='migrations'"

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

migrations-0.1.1.tar.gz (6.2 kB view hashes)

Uploaded source

Built Distribution

migrations-0.1.1-py3-none-any.whl (9.3 kB view hashes)

Uploaded py3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page