ti842py VS ImaginaryInfinity-Calculator

People have already responded saying yes, but I thought that I'd put my opinion in as well. Yes, you absolutely can distribute an application through PyPI even if it's not meant to mainly be a module, if fact I do in this project. There are other options if you want to go a different route, but PyPI may be the easiest.

ti842py - Some of you might remember making programs on the TI calculators in school, and I had an interesting idea a few months back. I found an old project that could decompile the programs into plain text, so I forked it and reverse-engineered a bunch of newer commands that TI-BASIC (the programming language used on some of the TI calculators) has added. I then set out to make a program that could transpile those programs into Python, and it works pretty well. Some examples of features that it's currently implemented are the drawing stuff and goto (ew goto). If you've got any old programs from those calculators lying around, and it seems interesting, give it a try.

TL;DR: I made a program to transpile programs made on the TI-84 into Python. link

On the other hand, if you're not trying to make colored logging and you just want to make a colored terminal-based application like this, then that's fine since you won't be doing anything with the stdout and stderr. If its logging, try to stay aqay from color.

As an example of other routes, I have this application. Instead of distributing through PyPI, I distribute it on a large variety of different platforms. I do include manual installers, but it's also distributed in the AUR, in a PPA, in the snap store, as an AppImage, through an exe installer (for windows), and as an rpm. It can also be run without installation to be even more accessible.

ImaginaryInfinity Calculator - I've been taking a break from developing this lately, but think of it as a super over the top calculator. It has a built-in package manager for downloading user-created plugins and themes, because the whole point of it is to be expandable. Can be installed on Linux, macOS, and Windows, and also ran without installation. Kind of overkill but really fun to work on.

If you need to install multiple files to different locations and also install Python, you could make an installer with a tool like Inno Setup. Though this does take some learning to get the syntax of both Inno Setup Script as well as Pascalscript (used in the dynamic parts of the installer), it is a valid option. Here's another example of me using Inno Setup in my project.

I mean, not the Python interpreter itself, but I use the calculator that I'm developing with my friend since it has things like plugin/theming support, and you can execute any arbitrary Python code in it as well.

If you're distributing to Windows users who could run a Powershell script, you could make the powershell script download the Python installer and launch it for them. Here's an example script that I use in one of my projects.

https://gitlab.com/TurboWafflz/ImaginaryInfinity-Calculator/-/tree/development - You could technically count this as mine since I develop it with my friend, but I'm putting it here anyways since I'm not the repo owner