Welcome to Docker-Map’s documentation!

Docker-Map enhances docker-py, the Docker Remote API client library for Python. It provides a set of utilities for building Docker images, create containers, connect dependent resources, and run them in development as well as production environments.

For connections through SSH tunnels, command-line tools, and additional deployment utilities, see Docker-Fabric.

The project is hosted on GitHub.


  • Configuration of container landscapes, including dependencies between containers, networks, and volumes.
  • Update check of containers, volumes, and networks against their running configuration. Consistency check of shared volume paths and re-creation of inconsistent or outdated containers.
  • Utility client functions. Some of these (e.g. cleanup_containers) have also been implemented on later versions of Docker directly; however this library can still provide more fine-grained control.
  • Creation of complex Dockerfile’s through Python code.
  • Simplified transfer of build resources (Context) to the Remote API.

Comparison to Compose

The functionality is quite similar to Docker Compose. When development of this library started, a predecessor of Compose was known as Fig. Docker-Map was mainly developed for the following reasons:

  • Docker-Map is intended to provide an API that could be embedded into other available systems such as configuration management. This implies that configuration can be implemented through code (although YAML input is also available) and functions can be invoked through Python directly.
  • Docker-Map is not a command-line utility, and it can operate on multiple clients. For example Docker-Fabric is one implementation that connect to other clients via SSH, but can also run on the same machine.
  • Fig and also its successor Compose target development and staging environments. Docker-Map aims to handle staging to production environments by allowing for embedding external variables, e.g. from configuration management. Although it can be combined with other tools to be suitable for development and CI, Compose is more common to use and might be the first choice there.


Docker-Map is being used for small-scale deployment scenarios in test and production. It should currently considered beta, due to pending new features, generalizations, and unit tests.

Indices and tables