Category: Programming

Good Morning from my Robotics Lab! This is Shadow_8472, and I am coming right off writing last week’s post because I have more to report, and it made a nice stopping point where I left off. Let’s get started!

I started work on my first prototype during my Third Workshop visit mentioned last time. It’s pretty chill there, so I have time to absorb what I’ve read as I read it slowly.

Philomena is on a Git repository, so I did some reading up on it the week before by going over some basic tutorials. I don’t actually remember much, but I will recognize stuff faster when reexposed to it.

As a side note, something seems to be failing on my laptop. I don’t know what, but it isn’t always booting on the first try. It took up several minutes of my visit to the workshop as it went through its battery of tests. Interestingly enough, the charger passed, even though I know for a fact that it won’t charge the battery due to a failure with the power port.

Git repositories have a special file that displays when you visit their web page. The first several times I looked through the one for Philomena, I had no clue what to do. This time, I recognized a number of commands to run, but require something called docker and docker-compose, neither of which come preinstalled on Raspian, apparently.

In the meantime, Git offers two ways to “clone” or download software contained in repositories. Recommended is over https, which almost all websites use by default now, and the other was over SSH.

Fun diversion, I had a really hard time with PuTTY when it kept complaining about Blinkie’s RSA host key whenever the IP changed. In the end, I hear most SSH clients don’t care about the IP as long as its host key is on a list of known hosts. PuTTY, however, saves host keys to each hostname unless unavailable, as in my case, then it uses the IP.

In light of the sideshow PuTTY put on for me to play through, I wanted to try downloading Philomena over SSH. When I first connected from Blinkie, I got a warning about an unknown host. The key was actually pretty easy to look up and verify. I put Blinkie’s SSH public key up there, but ran into a permission wall. I used wget instead. I later confirmed my heavy suspicion that wget operated on HTTPS and found it also supports a number of similar applications.

With the file downloaded from Git, I started building the tools I need to use it. I found a nice, command-by-command article on how to install both docker and docker-compose aimed at people who are not familiar with the command line. I followed the instructions, but not to the letter: I used Nano instead of vi when I needed a .txt editor, and I chose to use sudo instead of messing round with permissions and possibly introducing a security risk I don’t fully understand.

Docker went on fairly smoothly. They even have a program to verify that everything is working. Docker-compose gave me a little trouble though when I forgot to keep adding sudo.

Situation update: I got something to work. I’m operating blind right now. All I know is that I issued a command and was able to use a browser to navigate to Blinkie’s local IP with a port mentioned a few times during setup. I found what looked like a representation of Blinkie’s file system, but I’d need to double check. One thing I do know: whatever I was looking at went offline when I stopped it from the SSH terminal.

Update 2: I’m not sure where this style of updates came into happening, but I’ll go with it. I’ve done a little more research into Docker, and it looks like a sort of virtual machine –but not quite– and I think I logged into that VM earlier. I still have much to learn.

I’ve found several video series on YouTube documenting Docker and Docker-compose. With enough luck, going through some of those will let me understand what I’m at least looking at.

Update 3: Things have developed in the month or so since I went back to my other projects. I just need to call this a section and get back next week.

Final Question: How do you usually feel when returning to a project after a lengthy hiatus?

Good Morning from my Robotics Lab! This is Shadow 8472, and today, I am excited because I recently got a working prototype on my “preemptive” ban bot. Let’s get started!

First of all, I won’t claim to understand most of the stuff I’m working with, just that it isn’t as closely related to security as my last side project. If someone working all day every day could have done this project as easily as walking through a room, I would be crawling through a hole just barely my size. I’ve had a lot of help drilling that tunnel, and now widening it looks like it will be a whole lot faster.

A lot of what I did was trying to understand how Discord thinks, not quite fully getting it, getting help, and only understanding it until I needed another reminder. The long story is that I couldn’t repeat most of the details from memory. Guess what I’m basically doing right now.

I’ve been using PyCharm on my laptop as my IDE. On one of my previous trips to the workshop, Headcrash showed me how to use the debugger, but the neuances didn’t really stick. Before I went in this week, Damaged, the person who’s been helping me over Discord, helped me get through multiple hurdles, some of which should be obvious, others not so much. I was working up until right before sundown Friday, when I stopped for Sabbath.

The next night, I spent not two minutes before bed so I’d actually work on it in the morning, and got the early prototype working: I had someone on my a ban list who had never even been invited. Of course, I made a backup.

The next phase involved disarming the ban function’s “warhead” with a long comment and having it pull the target’s ID from the audit log. The function to loop over the audit log is even pretty user friendly at first glance. It was at this time I reached for a tool I wasn’t familiar with: the debugger.

A debugger is no doubt a valuable tool to know once you know how to use it. I came to a point where I had been introduced to it, but when I tried to use it, I tripped up on a small detail and took it in to the workshop. No sooner had I recreated my error and started explaining my problem than I poked around a little and figured out the answer on my own; the debugger has its own tab. I still learned a lot about how exactly this debugger works and how its interface is different than the ones I played with in the past.

At present, I have an untested/unproven candidate for my final bot before setting it loose on the real server. Right now, it can be told a bad mod’s ID and target logged unban events made by that account. When I’m done, I plan on having it tell me the names and numbers of all the people who were maliciously kicked.

Final Question: The filter for the audit log was not as user friendly as I had hopped., but not as bad as I feared (with the power of the debugger to view all active variable names and values at key points in the program). Have you ever come across a problem where the solution was on the path to that problem?

Good Morning from my Robotics Lab! This is Shadow_8472, and today I can taste victory on this mini-project, but as you can tell from the title, it shall be denied for this week. Let’s get started!

Last week, I talked about getting the PiCharm IDE running, and how it’s only as useful as the person maintaining it; as long as someone in the room knows what’s going on, things can get repaired when they break.

This week was another one where I went around building up my knowledge without feeling like I made any tangible headway. I keep telling people, “If I knew exactly what I was doing, I’d be done in two minutes,” and I still stand by that statement. Most of programming is figuring out how you want things to work overall, how your component parts were designed to work, and how the incomplete product actually works, then reconciling all three.

I spent most of my time studying how the async/await model of computing works. Before this week’s project chapter, I knew about multithreading and hyperthreading. All three are models for working on more than one thing at once in a CPU, and each has advantages and challenges. Imagine a computer thread as a guy at a workstation. Multithreading would be adding another guy at another workstation. Hyperthreading would be convincing two guys to share parts of the workstation they both aren’t using at the same time. “Coroutines” would be the first guy stepping aside from the workstation while he waits for more work to do and lets the next guy have a turn.

Other than that, I explored Redbot, some sort of bot to program for Discord. I ended up giving up on it for this project because it was a little expensive on the learning curve for me right now, and I’m not entirely sure it will suit my needs for now.

In the meantime, I’ve made the slightest of “outreach progress,” (yes, I just made that up to talk about when you aren’t ready for something, but try it anyway). I don’t know what’s going on, but hopefully I will know soon.

On the whole, a lot of my projects remind me of a few stories from doing trail work in Pathfinders, a youth club run by my church. One of our annual events involves working on trails at a local State park. While flattening the trail, everyone runs into rocks. Usually, they can just be relocated to somewhere nearby for drainage purposes, but every year, it seems, someone finds a rock in the way that just keeps going until it takes four or five teens to properly move it around once loose. Most of my projects feel like one of these boulders.

On the sad end of the spectrum, I was irresponsible with Blinky’s case. I ended up crushing it. Hopefully, I will get around to seeing to its repair.

Final Question: Windows recently updated. Among a bunch of stuff I didn’t bother reading, I spotted a Dark Mode feature I’m taking advantage of. Do you prefer to keep using the standard, bright theme, or will you be switching over to the dark theme?