DataJoint Development Python Docker Image with Dependencies and Tools

Tree

├── config -------------------------- Common configuration between all images
│   └── jupyter_notebook_config.py -- Jupyter Notebook server config
├── dist ---------------------------- Available image distributions
│   ├── alpine 
|   │   ├── .env -------------------- Uncommitted environment file
|   │   ├── docker-compose.yml ------ Reference compose for build/run
|   │   ├── Dockerfile -------------- Image build instructions
|   │   └── entrypoint.sh ----------- Startup script
│   └── ubuntu
|       ├── .dockerignore ----------- Files to disregard from build context
|       ├── .env -------------------- Uncommitted environment file
|       ├── docker-compose.yml ------ Reference compose for build/run
|       ├── Dockerfile -------------- Image build instructions
|       ├── entrypoint.sh ----------- Startup script
|       ├── requirements_post.txt --- Custom dependencies to be pre-installed
|       └── requirements.txt -------- Custom dependencies to be pre-installed
├── utility ------------------------- Common utilities between all images
│   └── startup.go ------------------ Utility that updates permissions as an underprivileged user
├── .gitignore ---------------------- Uncommited files
├── .travis.yml --------------------- Travis CI testing instructions
└── README.md ----------------------- Repo documentation

Use

To utilize the available containers, a .env and the appropriate docker-compose.yml is all that is needed. An an example, you may specify the .env as follows:

Alpine

ALPINE_VER=3.10
PY_VER=3.6
UID=1000
GID=1000

Ubuntu

UBUNTU_VER=16.04
PY_VER=3
UID=1000
GID=1000

You may start the container with the command docker-compose up.

Building

In order to build, the entire repo is required. See the appropriate docker-compose.yml and uncomment the build section. You may build with the command docker-compose build.

Features

• Pre-installed DataJoint dependencies
• Default application user dja aka DataJoint Admin
• Git
• Jupyter Notebook server
• GUI routing to Docker host e.g. matplotlib figures will display on host
• Alpine Image: Lightweight image ideal for Production, Scaling, Unit Testing, and Continuous Integation.
• Ubuntu Image: Complete suite of tools for on-going Development and ad hoc Analysis.

Note

To chain commands during startup, one may define the command section in docker-compose.yml as follows:

command: >
  /bin/sh -c
   "
    pip install --user datajoint;
    pip freeze | grep datajoint;
    jupyter notebook;
   "