I've been working up a fever on web-apps recently. Ones that I'm developing, ones that I'm deploying, and ones that I'm reviewing. That goes for both work and home. Gosh... I think I need a vacation. Maybe next lifetime.
Recently, at home, my focus has been on several "DJ Joe Services," things that I can utilize for my mobile DJ work and that will help me make those processes easier. Remember, I'm lazy! I want to find the easiest way to do things. Right now, I've got two apps deployed, and I'll be working on a third here pretty soon.
The first app was an availability calendar. I'll have to write about it soon, since it was a fun project tying APIs, Python, and React.js all into one solution. However, this is about my second app. What I call a "playlister," i.e., something that can slurp the playlist information out of another file/service/etc., and provide it in a more consumable manner. Right now, it's focus is on Spotify and Apple-Music, since those are the two prominent sources that are consistent enough for me to work with.
Inspiration
As a mobile DJ, I often am provided "playlists" in various forms: Word documents, text files, quickly-scribbled hand-written notes, Spotify playlists, and Apple Music playlists.
It quickly became apparent for me, that I spent way more time working through these Spotify playlists and Apple Music playlists to get them into a form that was actually helpful for me. In most cases, I could not simply copy/paste the Spotify list(s) out so that I could search for the songs of interest in my own library and then determine whether I'd need to aqcuire additional music. Thus... I came to the conclusion, I'd want a little assistance from my computer.
Stages of Development
I originally started with a simple Tkinter-app that used the spotipy
package to pull playlist information into a simple plain-text file. It was helpful, but
ended up incurring a few additional challenges of its own. The largest of which being the
fact I had to securly pass the API secrets around with the script itself. This became a
real burden, so I decided to enhance the system into a full-service mini web-app that
could be utilized for exactly this purpose. The web-app could run persistently on a server
that could hang on to those secrets and allow me to access the tool from anywhere.
Thus, the djjoeplaylister was born.
Technical Details
This app is built on the shoulders of giants, so let me give credit to those where it's due!
Technology Specs
- Language: Python 3
- Web Framework: FastAPI
- Web Listener/ASGI Server: Uvicorn
- Reverse Proxy: Nginx
- Hosting Provider: Linode virtual hosting
- Operating System: Ubuntu server
- App Deployment Enviromnent: Dockerized Container
Python Packages Leveraged
Additionally, I'd like to provide a special thanks and shout-out to this gist that helped me get up and running with consuming the Apple Music playlist without dealing with Apple's crummy developer program ($99 dolars a year, just to access an API? No thank you!) https://gist.github.com/aleclol/ef9e87d0964f00975f82d5373a814447
That's it! My little DJ Playlister! Want to go see it? Go Check it Out! It's not anything too terribly special, and it's got plenty of room to grow, but it's a helpful little tool, and I think it showcases the utility of the Python programming language.
I mean... just think about it.
I started with a simple little Tkinter script for which I had to lug secrets around all the time, and it was great! But it had some significant shortcomings. Python to the rescue though, a little refactoring, and throw in some HTML, CSS, and some more packages and I've got a full web-application. Still full Python, and it's fully-deployed! You can't do that with a lot of other tools. Imagine if I had started with some Excel macro, or some bash script. It would've been very difficult to scale those apps out to something that's actually useful in the context that I need.
Not with Python!
Preaching session over. Chat again soon, goodbye!