Welcome to my first devlog!

Here I would try to show what I have developed over the week! Even if the week is unproductive.

Python Backends⌗

Through a college project, I am getting to know the inner working of building a backend with Python. I have only built a backend with golang which resulted to nothing as I scrapped the entire project. Building a backend with Python was a great way of getting back to Python after a long time.

But I would say this project was one of the best project’s I have written in a while. Let me elaborate

Python ecosystem⌗

The first thing that impressed me was how cohesive the Python ecosystem has become. With the popularity of poetry and the usage of the pyproject.toml file, every tool is integrated and works with each other.

poetry is an project management tool for Python. It handles project creation, dependency management and even building the Python project without hassle. It’s one of the inspiration for groom

The second thing was the how easily I could use libraries like fastapi. When I imagined of working with Flask especially for building REST API’s I was a little disgusted.

Then I found out about fastapi. It looked similar to things I had been working with golang. After the initial introduction I was hooked. I Immediately started configuring a starter repo.

fastapi is a library that enables making REST API’s quite efficiently. If you are planning to make a REST API I would recommend fastapi for your next project instead of Flask or something heavy like Django.

AutoCode Bliss⌗

For the first time in my life, my entire repo was automated. Using tools like pdoc , poethepoet and pre-commit, I had setup a simple workflow which ensured that never any unformatted and untested code is pushed to the repository.

With pre-commit I can be confident that every commit is clean!

Why such precaution ?⌗

This project was not a personal project, it was a collaborative college project. Which meant a lot of people were about to contribute and commit to the repository. This workflow not only was enforced on myself but on everybody who wanted to contribute to the repository.

I knew this was needed to prevent problems in the future. I was mostly right!

CI/CD⌗

I also integrated a simple CI/CD pipeline within the repository which would check every PR. It would then build a docker image out of the repository and push it to DockerHub.

Hosting⌗

Through a friend’s help I had set up a lightweight x86 instance on the cloud.

I could run the docker image of my backend with the appropriate port forwarding.

But what happens when I update the image on DockerHub. Will I need to ssh into the cloud server and update the docker container manually ?

NO⌗

That’s where watchtower helped me. I found out about it through my self-hosting experiment and since then I have been using it regularly.

Basically whenever the DockerHub image updates, it automatically updates the local docker container with the updated image, preserving all the options like the port forwarding.

Frontend⌗

The frontend was handled by the aforementioned friend who is a wizard of React/React Native and gladly began work.

Thanks to him I am confident we can complete this college project and maybe learn a thing or two about collaborative development.

Writing this post has given me the idea of sharing my Python project setup. I believe it is quite good and maybe helpful to others.

BTW the code