Good Morning from my Robotics Lab! This is Shadow_8472 and today I am covering my programming work this past month. Let’s get started!
I am finding that even on the monthly cycle, I don’t always finish the projects I intend to. This month, I was intending to work on my phone, but that project hit some tough truth and the story isn’t ready yet.
HyperTrace
I might as well introduce my newest long-term project, HyperTrace. It started out in October as a spectator camera for a larger Redot community project. I studied up on using quaternions to rotate a camera node to avoid gimble lock, but just as I was starting to get it under control after I was bored stiff of them, I realized I could just attach a camera to a physics object.
But once I had a camera with 6 degrees of freedom (three positional and three rotational), I started thinking about how I could make a game out of it. I figured a space piloting game was in order. I got into studying up on using Perlin noise and similar methods of making structured noise to generate a shifting cave structure to represent a hyperspace mesh for the player to navigate – preferably at high speed.
The idea is that an unnamed pilot is tracing out a route through a shifting hyperspace (preferably at high speed), and if he collides with a wall, he is thrown out of hyperspace, where it isn’t safe to re-enter, so he must return to port under conventional physics. If he succeeds, his fleet can instantly jump across his route. This FTL system sets the stage for a future where humanity has colonized the solar system, but is hesitant to venture out to other stars. Think Star Wars meets Space Engineers meets Monkey Ball.
So far, my best tech demo is a blue fog you can navigate at slow, careful speeds. It takes an unacceptably long
Shell Scripting
My services were called upon for a “computer project.” My mother had some PDF files of recipes to separate. Over the course of an afternoon, I wrote a BASH script to separate out each page in each PDF. At its heart was the program qpdf. Along the way, I tried to keep good practices like leaving comments and having checks to prevent accidental overwriting.
Now, I only know a modest bit of BASH (popular Linux command line interface aimed at users), but I know enough about basic programming it was just a matter of syntax for me. But I figured I could do one better in around the same amount of time. I saved a copy of the script and its output, and I set about naming files after their contents, identifying+recombining second pages, and sorting out charts, and numbering repeats (some duplicate recipes had changes).
But for every hour I spent working out a feature, I figured I spent another debugging edge cases. Several recipes had BASH control characters in their titles like *, &, and /. I preferred to use special handling of strings, but in the end I sanitized / as ÷ and * was removed entirely. This additional work extended the project to a total of three days.
Closing thoughts
My goal with Hypertrace is to publish it to Itch.io or similar platform on a donation basis. And before I get past more than a couple tech demos, I’m going to want a version control server in my homelab just in case something happens to my laptop. While I really want to think ahead in terms of security, I know I don’t know enough to make an informed decision about the utility I need now.
Final Question
What Git setup do you recommend for a homelab?