More progress on my SSG app
Today I did more Python programming for SSG. I am slowly but surely chipping away at the development of it. And more importantly, I’ve finished all the concrete details for the design, so it’s very clear what I need to do. That is the hard part. Once you’ve figured out what you’re going to program, it’s not that hard to go and implement it. Programming is only more difficult when you’re not sure how to design something, or just coding without any sense of direction. But I like to plan things out before jumping right in and coding. That’s why this phase is a lot easier for me now. Project management, sprints, etc. all really matter.
Today, I worked on my initial setup module more.
After you successfully open a website project in Static Site Generator, it checks a file to see if you’ve finished the initial setup or not. If not, then it will run the functions to set the settings input and about input. There’s some basic information that the site needs before it can really do anything. You need basic stuff like email address, website name, and some social media links. The website name is not the same as the project name. Project names should be things like alans_cool_website and then the title could be something like “Alan’s Cool Website” because of issues with spaces in command line arguments, and issues with how operating systems handle special characters.
Because a user could hit ctrl+c or close the terminal window after setting up the project, there is no guarantee that they entered in all of the information in order to configure the project. And if they close the program when it’s still getting input, it won’t have written anything to the project’s JSON configuration files, settings.json and about.json, which means none of the settings would get saved at all. So that’s why it does the check when you first open a project. When the project successfully has the initial setup completed, the finished_intial_setup.txt file has its contents changed to reflect that it no longer needs to go through the setup process. Then, the next time it’s opened, it will skip the initial setup stuff. The user can always change the settings later, but that is separate from the setup module, and the settings would be changed one at a time rather than getting input for all of them. This is useful is the person making the site decides that they want to change the name or links or whatever later.
Also, one design choice I had was to exclude the author name from the settings.json. This is deliberate, because I want SSG to let people make sites that have multiple contributors. So every time you write an article, you have to specify who the author is. It might be slightly inconvenient for people who don’t have multiple people writing for their site, but it’s great if a group of people all work on the same project. And it’s still more convenient than writing a site from scratch, even with having to enter the author name every time.
Today, I finished the initial setup module. Now I’m going to start working on the article module. The setup module is important because all projects need to be set up with some basic information. But once that’s all said and done, there are a few key areas I need to work on. What I’ve decided to work on now is to do the article module/menu because then you can deal with articles, or I guess you could call them posts. Articles are an essential part of a site, whether it’s made with Wordpress, Jekyll, SSG, whatever. People want content, and people want a timeline. The article module by itself only deals with getting user input and saving it to JSON files.
Across all files, not counting some redundant stuff, SSG is about 2,000 lines of code. It’s far from finished though.
Lately, I’ve disabled app notificatons on my phone, turned it to silent, and stopped checking it as often as I used to. I also don’t read the news or check social media as much as I used to. This really helps me to get more programming and learning done. Small interruptions here and there can really disrupt your productivity. Some people act weird when you don’t respond immediately, but I think it’s healthier and better for getting things done if you’re not spending so much time on a phone or social media platform. I would rather write code or learn a new computer science topic instead of always being immediately available for everything.
Push notifications aren’t a bad idea in general, but many apps misuse them in order to get you to spend more time within the app. But after a while, if apps give you too many notifications, you just start to tune them out, or even uninstall the app because it’s annoying.
Windows Server progress
I finished memtest86+ on it, and surprisingly, all the old RAM sticks were just fine. So it has 12GB of RAM. It’s a weird amount, and it will only be running in single-channel mode, due to being an odd number of DIMMs. But that’s okay for a low-usage server like this, which is primarily intended to be used educationally, not really at high load.
After finishing memtest86+, I used a tool called Rufus to make a bootable USB installer for Windows Server Essentials 2019. I use a KVM so I can control multiple computers with just one keyboard, video output, and mouse.
I installed Windows Server Essentials 2019 on it. I installed it on the SSD and then set up the 2x 500GB drives in a mirror, using Windows’ Computer Management program. Everything seems good enough.
I had to research Windows 10 versions. You need at least Windows 10 Pro in order for a workstation to be able to connect to a domain. If I want to use some computers with it, I might have to change their versions/licenses. I use a site called Kinguin to buy Windows keys.
Even though a KVM is convenient, I might just end up using some sort of remote access softare, because I want to move the server somewhere away from my main computer desk. This means the performance will be slightly worse, but nothing too terrible. And it’s always good to get experience with important software, such as RDP or VNC or something.
Factory design pattern
Today, I learned about another kind of design pattern: factory. The example in the video tutorial I watched said it’s a method that returns one of several possible classes that share a common super class. The example they used was creating a new enemy in a game. So basically it’s an easy way to instantiate objects. I could easily see how the singleton design pattern could be useful in my SSG program, but I’m actually having a hard time seeing how I could implement a factory into it. But maybe that’s just because SSG is a relatively simple program. I could definitely see how it could be used in a game, where you can do some randomization instead of having instantiation be the same every time, and it’s useful when you don’t know ahead of time what class object you’ll need. In many cases, you can use different subclasses of the same superclass, though it doesn’t fit for every situation.
One thing I’ve noticed about learning design patterns is that I try to think about “how can I implement this in my current project?” and sometimes I can’t see a good way to do it where it’d be useful instead of forced/shoehorned unnecessarily. And I think that’s perfectly fine – not everything needs to be overengineered.
In some cases, a factory might involve instantiating classes that inherit from an abstract superclass. The video I watched on this topic gave the example of an EnemyShipFactory design pattern that instantiates objects from the UFOEnemyShip and RocketEnemyShip subclasses of the EnemyShip abstract superclass.
The next design pattern I will learn is abstract factory, but that can wait until another day.
Other Python and programming topics
I learned about inheritance and polymorphism in Python. Those aren’t new topics to me, but I hadn’t done them in Python yet. I also learned about querying JSON APIs in Python. Maybe, at some point, I will write a test program to query one of my Wordpress sites’ built-in APIs. Interacting with APIs is very important, because you won’t be using everything yourself, and libraries are only for code. APIs are useful not only for features and code, but a lot of APIs are about data. API interaction is more important these days, as more and more programmers are network-connected instead of being isolated and offline-only.
One example of how you could combine APIs with Python is to use the OpenWeatherMap API to get weather data, and then use the Tweepy and the Twitter API in order to make a Twitter bot to tweet out the weather from the weather API.
Another Python topic I looked up today is recursion, using the Fibonacci sequence as an example problem to solve with it.
I also learned about lambda expressions/anonymous functions in Python. I used them a lot in Java, but I haven’t really used them in Python yet.
Other stuff I researched today:
- Resume editing
- Sass CSS preprocessor
- Router Entware/Optware – basically apps/packages you can install on DD-WRT or OpenWRT routers
- SSH honeypots
- Manifest files
- A cool way to pipe output from one command line program as a command line argument for another program.
- Imposter syndrome in tech
Today, I downloaded a Bitnami Django stack and set it up on my ESXi hypervisor.
The process of downloading a premade Django VM and then installing it in ESXi only takes a couple of minutes. It’s much faster than setting everything up manually. Hypervisors are great in the sense that they allow you to set up many virtual servers quickly and easily without having to always buy new hardware. It’s true that I currently have two bare metal servers – one for FreeNAS, and the other for Windows Server. But most of my servers are virtual machines on my VM server. I feel very comfortable with my virtualization skills, though I think I need to get more into containers instead of just VMs.
New git repos
I made new git repos for my notes repositories for learning SQL and C. These are just like my gitUdemyLearning and misc_python_1 repos: just educational notes and testing rather than being fully-fledged apps.