container-tester 1.0.0rc1

Generate multiple docker cointainers to test projects

Container-tester

Overview

Container-Tester is a small Python utility that helps you test your project on top of many different Linux distributions without having to run and maintain lots of Docker tests yourself. It provides a command-line interface (CLI) that can:

• Build that Dockerfile into a image and report its size and metadata.
• Run a container from that image, execute a command inside it and capture the logs.
• Optionally clean up images and containers when you are done.
• Iterate over a curated list of popular Python, Debian/Ubuntu, Fedora and Alpine base images defined in docker-config.toml and perform the above steps for each of them.

Installation

You can use Container-Tester in two ways:

1. Install in your project:

   uv add container-tester
   uv run contest --help
   
   # or if you don't want to install:
   uvx --from container-tester contest --help
   

Using Container-tester

Usage: contest [OPTIONS] [OS_NAME]

Generate, build, and run Docker resources from a base image or config file.

+- Options -------------------------------------------------------------------+
| --json    --no-json          Show output in json format                     |
|                              (use --json to enable) [default: no-json]      |
| --pretty  --no-pretty        Show output in pretty format                   |
|                              (use --pretty to enable) [default: no-pretty]  |
| --clean   --no-clean         Clean Docker resources after run               |
|                              (use --clean to enable) [default: no-clean]    |
| --install-completion         Install completion for the current shell.      |
| --show-completion            Show completion for the current shell, to copy |
|                              it or customize the installation.              |
| --help                       Show this message and exit.                    |
+-----------------------------------------------------------------------------+

+- Commands ------------------------------------------------------------------+
| test-config     Generate, build, and run Docker resources from a docker     |
| test-container  Generate, build, and run Docker resources from a base image.|
+-----------------------------------------------------------------------------+

Example

Here is a quick example that runs a simple command across all default profiles:

uv run contest test-config

To test only on the Ubuntu image, you can specify its base image directly and use a custom command:

uv run contest test-container ubuntu:latest --command "echo Hello, world!"

Default profiles

By default, Container-Tester comes with a handful of profiles covering common Python, Debian/Ubuntu and Alpine base images. Each profile defines an image_tag (used as the resulting image name) and the corresponding os_name. A few of them are shown below- see src/container_tester/docker-config.toml for the full list:

image_tag	os_name
py312_trixie	python:3.12-slim-trixie
ubuntu_latest	ubuntu:latest
alpine_latest	alpine:latest

Custom Docker Configuration

You can define your own docker-config.toml file to run custom Docker images tailored to your needs. Use the following format to specify multiple profiles:

[py311_slim]
command = ""
os_name = "python:3.11-slim"
os_commands = []
pkg_manager = "apt"

[ubuntu_latest]
command = ""
os_name = "ubuntu:latest"
os_commands = ["apt-get update && apt-get install -y ca-certificates"]
pkg_manager = "apt"

# Add more profiles as needed...

Configuration Fields

• command: Shell command to execute inside the containers.
• image_tag: A unique identifier for the Docker image profile.
• os_name: The name and tag of the Docker image to use.
• os_commands: A list of shell commands to run after container startup (optional).
• pkg_manager: The package manager used by the OS (e.g., apk, dnf, apt).

Run with Your Custom Config

After creating your docker-config.toml file, launch your custom Docker setup by running:

uv run contest test-config

License

This project is licensed under the terms of the MIT license.