Blueprint/Boilerplate For Python Projects

If you find this useful, you can support me on Ko-Fi (Donations are always appreciated, but never required):

Blog Posts - More Information About This Repo

You can find more information about this project/repository and how to use it in following blog post:

• Ultimate Setup for Your Next Python Project
• Automating Every Aspect of Your Python Project
• Deploy Any Python Project to Kubernetes
• Implementing gRPC server using Python

Quick Start

To use this repository as starter for your project you can run configure_project.sh script, which sets up all variables and file names. This way you can avoid configuring and renaming things yourself:

./configure_project.sh MODULE="coolproject" REGISTRY="docker.pkg.github.com/martinheinz/repo-name"

Running

Using Python Interpreter

~ $ make run

Using Docker

Development image:

~ $ make build-dev
~ $ docker images --filter "label=name=blueprint"
REPOSITORY                                                             TAG                 IMAGE ID            CREATED             SIZE
docker.pkg.github.com/martinheinz/python-project-blueprint/blueprint   3492a40-dirty       acf8d09acce4        28 seconds ago      967MB
~ $ docker run acf8d09acce4
Hello World...

Production (Distroless) image:

~ $ make build-prod VERSION=0.0.5
~ $ docker images --filter "label=version=0.0.5"
REPOSITORY                                                             TAG                 IMAGE ID            CREATED             SIZE
docker.pkg.github.com/martinheinz/python-project-blueprint/blueprint   0.0.5               65e6690d9edd        5 seconds ago       86.1MB
~ $ docker run 65e6690d9edd
Hello World...

Testing

Test are ran every time you build dev or prod image. You can also run tests using:

~ $ make test

Pushing to GitHub Package Registry

~ $ docker login docker.pkg.github.com --username MartinHeinz
Password: ...
...
Login Succeeded
~ $ make push VERSION=0.0.5

Cleaning

Clean Pytest and coverage cache/files:

~ $ make clean

Clean Docker images:

~ $ make docker-clean

Kubernetes

Application can be easily deployed on k8s using KinD.

To create cluster and/or view status:

~ $ make cluster

To deploy application to local cluster:

~ $ make deploy-local

To get debugging information of running application:

~ $ make cluster-debug

To get remote shell into application pod:

~ $ make cluster-rsh

To apply/update Kubernetes manifest stored in k8s directory:

~ $ make manifest-update

Setting Up Sonar Cloud

• Navigate to https://sonarcloud.io/projects
• Click plus in top right corner -> analyze new project
• Setup with other CI tool -> other -> Linux
• Copy -Dsonar.projectKey= and -Dsonar.organization=
  • These 2 values go to sonar-project.properties file
• Click pencil at bottom of sonar-scanner command
• Generate token and save it
• Go to repo -> Settings tab -> Secrets -> Add a new secret
  • name: SONAR_TOKEN
  • value: Previously copied token

Creating Secret Tokens

Token is needed for example for GitHub Package Registry. To create one:

• Go to Settings tab
• Click Secrets
• Click Add a new secret
  • Name: name that will be accessible in GitHub Actions as secrets.NAME
  • Value: value

Resources

• https://realpython.com/python-application-layouts/
• https://dev.to/codemouse92/dead-simple-python-project-structure-and-imports-38c6
• https://github.com/navdeep-G/samplemod/blob/master/setup.py
• https://github.com/GoogleContainerTools/distroless/blob/master/examples/python3/Dockerfile