This is the repository for the PyConDE 2024 talk "Unlock the Power of Dev Containers: Build a Consistent Python Development Environment in Seconds!". The slides of the talk can be found here.
If you want a reproducible development environment in Python that is easy to set up, then DevContainers are a good choice, if not the best. DevContainer allows you to define the operating system and system libraries next to the Python packages and the Python interpreter. All together this results in an always robust, reproducible development environment.
DevContainers are now supported by a large number of IDEs, which means next to VSCode also PyCharm and Nvim, for example. Furthermore, next to Github, more and more remote development platforms support DevContainers. The prerequisite is that docker is installed on the computer.
A wonderful source of inspiration can be found in the VSCode DevContainer documentation. The DevContainer specification can be viewed here.
In this repository you will find an example of how you can use DevContainer in your project. All you have to do is copy the file .devcontainer/devcontainer.json into your project and adapt the file .devcontainer/Dockerfile. You can then start the development environment by clicking on the green icon in the bottom left-hand corner of VSCode.
If you are interested in more complex setups, we recommend the following repositories:
There is an issue installing vscode extensions using MacOS (microsoft/vscode#173327). In case you use MacOS please be sure to uncomment the following line in the .devcontainer.json file:
"extensions.verifySignature": falseMacOS users can also use native x86_64 images very well using Rosetta. If you want to explicitly use an x86_64 image, you have to add the following line to the .devcontainer.json file:
"runArgs": ["--platform", "linux/amd64"],