Skip to content

Containerizing Python Applications with Docker

Ignore Python-specific artifacts

Often, when you build a Docker image, you will end up copying Python code directly of the local machine into a Docker image. If you have previously executed your program, you might end up copying your __pycache__ folders containing compiled .pyc files. PyTest is notorious for screaming at you, if you copy your code somewhere else and forget to clear the cache:

import file mismatch:
HINT: remove __pycache__ / .pyc files and/or use a unique basename for your test file modules

This is where a .dockerignore file comes handy. Similar to a .gitignore file, it will exclude directory/file patterns you put there from being copied onto the final Dcoker image. Here is a simple .dockerignore file you can add to the root of your Python project:1