RocketPy with docker#
RocketPy does not provide an official docker image, but you can build one yourself using the provided Dockerfile and docker-compose.yml files.
Docker allows you to run applications in containers. The main benefits of using docker are:
Isolation: run RocketPy in a fresh environment, without worrying about dependencies.
Portability: run RocketPy on any operational system that supports docker, including the 3 main operational systems (Windows, Linux and Mac).
Reproducibility: ensure that tour code is working regardless of the operational system.
Using docker will be specially important when you are not sure if the code additions will still run on different operational systems.
Although we have a set of GitHub actions to test the code on different operational systems every time a pull request is made, it is important to submit a PR only after you are sure that the code will run flawlessly, otherwise quota limits may be reached on GitHub.
Before you start, you need to install on your machine:
Build the image#
To build the image, run the following command on your terminal:
docker build -t rocketpy-image -f Dockerfile .
This will build the image and tag it as rocketpy-image (you can apply another name of your preference if you want).
An image is a read-only template with instructions for creating a Docker container (see Docker docs).
This process may take a while, since it will create an image that could easily be 1.5 GB in size. But don’t worry, you just need to build the image once.
Run the container#
Now that you have the image, you can run it as a container:
docker run -it --entrypoint /bin/bash rocketpy-image
This will run the container and open a bash terminal inside it. If you are using VSCode, you can even integrate the running container into your IDE, allowing you to code and test directly within the container environment. This is particularly useful if you want to maintain your usual development setup while ensuring consistency in the execution environment. For more details on how to do this, refer to the VSCode docs on developing inside a container.
Indeed, vscode offers a full support for docker, read the vscode docs for more details
Run the unit tests#
You might have noticed that the container is running in an isolated environment with no access to your machine’s files, but the Dockerfile already copied the RocketPy repository to the container. This means that you can run tests (and simulations!) as if you were running RocketPy on your machine.
As simple as that, you can run the unit tests:
To access a list of all available execution options, see the pytest docs.
Compose docker images#
We also made available a docker-compose.yml file that allows you to compose multiple docker images at once. Unfortunately, this file will not allow you to test the code on different operational systems at once, since docker images inherits from the host operational system. However, it is still useful to run the unit tests on different python versions.
Currently, the docker-compose.yml file is configured to run the unit tests on python 3.8 and 3.12.
To run the unit tests on both python versions, run the following command on your machine:
Also, you can check the logs of the containers by running:
This command is especially useful for debugging if any issues occur during the build process or when running the containers.
After you’re done testing, or if you wish to stop the containers and remove the services, use the command:
This will stop the running containers and remove the networks, volumes, and images created by up.
Changing to other operational systems#
The default image in the Dockerfile is based on a Linux distribution. However, you can alter the base image to use different operating systems, though the process may require additional steps depending on the OS’s compatibility with your project setup. For instance, certain dependencies or scripts may behave differently or require different installation procedures, so use it with caution.
To change the base image, you will need to modify the FROM statement in the Dockerfile. For example, to use a Windows-based image, you might change:
Please note, the above is just an example, and using a different OS may require further adjustments in the Dockerfile. We recommend you to see the official Python images on the Docker Hub for different OS options: Docker Hub Python Tags.
Be aware that switching to a non-Linux image can lead to larger image sizes and longer pull times.