
## Introduction

Hey guys, I just deployed a new version of my website! I'm excited because this version adopts modern best practices like unit testing, CI/CD and serverless! The main benefits of course being reduced hosting costs and enhanced scalability.

I developed my last website in Ruby on Rails, hosted on a virtual machine with DigitalOcean. I've been wanting to move to a serverless setup for a while now, and I finally made the leap.

### Serverless

The site now resides on AWS, utilizing an S3 bucket for file storage (HTML, CSS, JS, assets, etc. ) and CloudFront for distribution on my custom domain. A key reason for this upgrade is the automation of deployment; thanks to Github Actions, each new release triggers the build of SEO-optimized HTML, which is then automatically uploaded to AWS S3.

<pre class="mermaid">
graph LR
    User -->|www.rgonzalez.tech| CF[Amazon CloudFront]
    CF -->|Serves Files From| S3[(S3 Bucket)]

    Github -->|Push Changes| GA[GitHub Actions]
    GA -->|Build and Upload| S3
</pre>

This setup is not only cost-effective, but also highly scalable. I can now handle traffic spikes without any manual intervention. Plus, the development workflow is much smoother, as I can now focus on writing content & upgrading my website and not worry about server maintenance, scaling, or deployment.

### Markdown

This shift isn't just about infrastructure. It also transforms how I manage my blogs. I've adopted markdown for writing blog posts. I have a directory in my github repository where I write my blog posts in markdown. When the website gets rebuilt (on a new release), the markdown files are compiled into static HTML files.

This approach allows for a streamlined workflow where I can focus on content creation in my preferred text editor, without the overhead of managing a database or wrestling with WYSIWYG editors.

I can also make use of powerful features in Markdown, such as code blocks and mermaid diagrams. For example, here's a simple mermaid diagram that explains the deployment process:

<pre class="mermaid">
graph LR
    A[GitHub Repository] -->|Publish Release| B[GitHub Actions]
    B -->|Build and Upload| C[AWS S3]
</pre>

In order to draw that diagram, I just had to write the following markdown:

```html
<pre class="mermaid">
graph LR
    A[GitHub Repository] -->|Publish Release| B[GitHub Actions]
    B -->|Build and Upload| C[AWS S3]
</pre>
```

## Rails to Next.js

The move from Rails to Next.js was a big one. I had to rewrite the entire frontend, but it was worth it. I'm happy to use React for my website and thanks to Next.js, I can now take advantage of static site generation. This means that my website is now faster, more responsive, and more SEO-friendly.

### Design

I took the opportunity to redesign the website. I've always been a fan of minimalism, and I think the new design reflects that. I've also made sure that the website is fully responsive, so it looks great on any device.

This redesign was actually generated with the help of GPT. I explained the kind of design I wanted to ChatGPT and I asked it to write the HTML & CSS to help me visualize the design. It was a great starting point.

![](/images/blogs/new-site-2024/AI_HTML.png)

I spent some time in draw.io to create a wireframe for what I wanted the website to look like. (Yes, the background image is also generated by GPT!)

![](/images/blogs/new-site-2024/wireframe.png)

Then, I remixed that with a design generated by the AI over at [UIZard](https://uizard.io/).
The end result is a website that I'm proud of. It's clean, modern, and easy to navigate.

### Migrating Content

Migrating content from my old website to the new one was actually rather simple thanks to the help of ChatGPT and it's data analysis feature. 

I was able to upload the `.sqlite` database directly to ChatGPT and it was able to leverage python to query and manipulate the data from the database for me. I asked it to create an excel sheet containing all the blog posts and their metadata.

![](/images/blogs/new-site-2024/AI_Database.png)

Then, I asked it to use the metadata to generate markdown files for each blog post. It was a huge time saver.

![](/images/blogs/new-site-2024/AI_Convert_Markdown.png)

The only remaining task was to get the images & uploaded files from the old website to the new one. Unfortunately, rails active storage doesn't have a direct way to download all the files. So, I had to write a rails script to download all of the files from the old website.

```ruby
ActiveStorage::Attachment.all.each do |attachment|
  puts "------------------"
  puts "Processing attachment: #{attachment.id}"
  puts attachment.filename
  puts attachment.record_type
  puts "------------------"

  # Get the image data
  blob = attachment.blob

  # Get the record type
  record_type = attachment.record_type

  # Get the filename
  filename = attachment.filename

  # Get the extension
  extension = filename.to_s.split(".").last

  # Get the record id
  record_id = attachment.record_id

  # Create the folder for the record type
  folder_name = record_type.downcase.split("::").last
  folder_name = "#{Rails.root}/tmp/images/#{folder_name}"

  # Create the folder if it doesn't exist
  FileUtils.mkdir_p folder_name

  # Create the new filename
  new_filename = "#{filename}.#{record_id}.#{extension}"

  # Create the new path
  new_path = "#{folder_name}/#{new_filename}"

  # Write the image to the filesystem
  File.open(new_path, "wb") do |file|
    file.write(blob.download)
  end
end
```

## Lessons Learned

I learned a lot from this update. I learned that less is more and that being able to quickly iterate on a design or a project is a huge advantage. I have really enjoyed spending time developing with modern AI-driven tools and I'm excited to see what the future holds. I am now planning on using this website to write more blog posts and document my experiences with modern development practices, tools, and challenges.

I hope you enjoy the new website. I'm looking forward to writing more blog posts and sharing my thoughts with you all. Stay tuned for more updates!


Hey, what's up guys? It's your boy *back at it again with...* another post. This time, I'm thinking of updating my website and it's design. I've been hosting my blog on a VM running Ruby on Rails + Apache for a while now, but I feel like it's time to take things up a notch.

## The Design
---

So, here's what I'm thinking: I want to hire a designer to create a responsive web design that'll make my website really stand out. I'm still on the fence about this idea, though, because I recently signed up for Udemy's subscription and I'm thinking about taking some UI/UX/Design classes to improve my own skills. We'll see. It might be worth it to invest some time learning that and doing it myself rather than outsourcing that. We'll see 🤷‍♂️

![](/images/blogs/paintbrushes.png)

## Serverless
---

![](/images/blogs/the_cloud.png)

I'm really excited about using serverless technology for my personal blog. I've seen the benefits it can offer through my professional work, and I understand the benefits it provides for web development. Automatic scaling and cost-effectiveness is too good to ignore 📈🤑. I'm looking forward to saying goodbye to apache configurations and hello to Github actions, lol.


I'm confident Azure, GCP and AWS can all handle the infrastructure, so I just need to pick the right one for me. I'll figure that out later 🤦‍♂️. I use Azure at work, but I've worked with GCP and AWS for hobby projects before, and I feel like it would be neat to really learn these other platforms. I've really come to enjoy working with a CI/CD workflow. It's a game-changer in terms of deploying. Working with FaaS, unit testing, and CI/CD has been loads of fun. Updating code is so much easier.


In conclusion, I'm excited to revisit my personal website as a project. With some new web design and serverless tech, I think this website will be a better showcase of my work and ideas. It's going to take some work, but I'm up for the challenge. Now I just need to set aside the time 😅


Thanks for joining me on this journey!





Hey everyone, it's me again, the nerdy developer working on a random co-op game. It's been a minute since my last blog post, but I finally got some time this weekend to add some more features to this new project. So far, we've got a main menu and a character controller as well as the spawn room. It's now time to start adding some gameplay, so I need to build the interactive systems.


As I mentioned in my last post, I've been getting some help with 3D art; I'm collaborating with a Redditor from Spain who's doing an amazing job creating some cool assets for the game. But for the interaction system, it's just been me and Godot.


![](/images/blogs/Progress.gif)


## Interaction System
---


Now I'm planning the interaction system for this game, and I wanted to share my progress with you all 😁.


I've been thinking a lot about how to create a flexible interaction system that will work will with my future plans for players and the game world. That's why I'm building an interaction system that will allow players to walk up to objects and interact with them using (A) or [SPACE], both with tap and hold interactions. I love that godot is lightweight and flexible with it's script systems.


Of course, with great flexibility comes great responsibility 😴. Sometimes, it can be exhausting to design a system from scratch. At the same time, it's a blessing to have the freedom to craft the interaction system exactly as I need it. I think I've designed it well enough to be extendable in the future, but I'm curious to see how this system evolves and how it contributes to the overall gameplay experience.


I've been working on making the interaction system generic enough so that we can build any kind of interactable object. Players will be able to walk up to an object to interact with it, or potentially interact with it from range in the case of the mage role. Some objects might require a tap interaction while others may require a hold interaction. I've designed it so that we can build doors, levers, rocks, and anything else we can think of that a player may need to interact with. Plus, the system is co-op friendly, so sometimes one player can interact with an object, but another player cannot.


So, How?


## The System
---


![](/images/blogs/engine.png)  

*Art generated by Stable Diffusion*


The system is designed to leverage Godot's signals and duck-typing, so we can easily add new interactable objects and handle complex interactions. For example, we can use the **has\_method()** function to check if an object has the **interact\_start(node)**, **interact\_end(node)**, or **selected\_by(node)** functions. That will let us know that the object is interactable. "If it quacks, it's a duck!", lol.


Some objects will only be interactive when approached by certain roles or items. It's been a challenge to design the system to be extendable for future development, but the way that we handle this is to give interactive objects a **selected\_by(node)** function that returns true or false. When a player tries to highlight an object, the player will call **object.selected\_by(self)** and if the result is true, then the object can be interacted with by this player.


If the player is allowed to highlight the object, then the player will select the object and store it in the player's **selected** variable. When the player interacts with the object by pressing the A button, the object's **interact\_start(node)** function will be triggered. When the button is released, we will call on **interact\_end(node).** This will allow every object to determine whether it should do something on keydown, or keyup. I also thought about how to handle situations where multiple players are trying to interact with the same object at the same time. To handle this, we've added an array called **players\_selecting** to each interactable object. When a player calls **object.selected\_by(self)**, the object will store the player in the array and subscribe to the player's **selection\_changed** signal. If the player's selection changes to another object, the object will remove the player from the array. If the array still has at least one element, then we highlight the object so that players know they can interact with it.


Right now, I'm still in the planning stages and haven't yet implemented the system in Godot. But I'm making progress, and I'm hoping that it all goes according to plan. Once it's implemented, I'll be testing and refining it to make sure it feels natural and intuitive for players.


That's it for this update, nerds!! Stay tuned for more updates on game development. And if you have any questions or suggestions, feel free to drop them in the comments. Thanks for following along!





I've been busy for a while now, but I'm back again and I'm excited to share what I've been up to!


I recently picked up the Godot engine since it had caught my eye and I became curious. I've been enjoying my time with it so far. First, I decided to rebuild my AnimeMotion game in Godot and I gotta say, even though the engine requires you to spin up more solutions on your own, it has been very rewarding! I re-created the movement and camera controller in Godot and it was a lot of fun. Then, I stopped around the time that I had to build out the air dash and get started on combat systems 😅


Now, I'm working on a *new* project where I'm collaborating with a redditor skilled in 3D modeling and animation. He's from Spain! We're creating a 3D, top-down, fantasy adventure co-op game where you have to work together to succeed. The art style is very similar to Overcooked so far! It's been a blast to be able to cooperate with another creative person. 🎉


![](/images/blogs/image-20230209235031-1.png)


I'd definitely recommend Godot. The engine is easy to learn and the documentation is great. It's open-source and very lightweight, smaller than 100Kb!!! Doesn't require a lengthy installation.... Unity 📢


I was able to pick up the scripting language in an afternoon. There's a lively community and I've found that it's a great platform to experiment with. I really, really like the node-based system. It's really flexibile and I'm loving it!


I'm looking forward to finishing this co-op project and seeing what else I can create with Godot. Maybe I'll join a gamejam and write about it? Not sure. But I am sure I'm not alone in enjoying my time with Godot. I see all kinds of posts on Reddit and Youtube talking about the new Godot 4 release that's coming out 🤩.


✌ Looking forward to seeing all the cool projects Godot will create in the coming year! Happy nerding everyone! 💻





So, you know that game project I started just about a year ago, then I stopped working on it? I picked it back up again for a bit... and now I'm putting it back down. I just cannot seem to listen to my own plans / advice. 🤦‍♂️

## NEW CHANGES!
---

Since my last blog post talking about my game project, I decided to work on some of the refactoring changes that I had spoken about before. I never built a level picker or even any other levels. I got a bit too busy and have been too busy with work so unfortunately I never polished this game. In fact, that's where I'm stopping... again (Polish Stage).

I rewrote the code that handles the Wave management and the projectile/boomerang code as well.

## Test Driven Development
---

I've refactored the code for the enemy wave controller, the UI and the projectile system so that it can be properly unit tested. You can read more about Unit testing and clean code practices by searching up Uncle Bob. The developers of Factorio wrote a blog post talking about clean code practices as well. I figured I could also implement TDD into my personal projects.

![](/images/blogs/image.32.png)

```csharp
[Test]
public void OnWaveStartItQueuesAWaveStartMessage()
{
    _stubBroadcaster.StartWave();
    AnnouncementMessage currentMessage = _announcementComponent.GetQueue().First.Value;

    Assert.AreEqual("Wave #1 Started!", currentMessage.GetMessage());
    Assert.False(currentMessage.IsPriority());
    Assert.AreEqual(1, currentMessage.GetLifetime());
}
```

## Summarize
---

I refactored the boomerang system and the Wave UI system. The boomerang system is functionally the same, but it's been refactored to leverage interfaces more heavily. This allows the components to be loosely coupled. The Wave UI has more components like the enemies remaining and the wave history. Maybe I'll keep working on this project eventually. No promises. 3D might've just been a bad call for a solo project, lol.

![](/images/blogs/WaveUI.gif)


I took a vacation from work in the first week of September this year. I set a challenge to develop an app during my vacation, which was 12 days long. The goal was to develop an app and launch it within that time limit. The point of the challenge was to force myself to limit my scope because I've noticed that I tend to bite off more than I can chew.


I decided that my first app would be a maze-racing app. The whole point of the app would be that you would queue up and try to solve a maze before the other members in the race. I figured this scope was small enough because the game was 2D and didn't have any physics involved.


I was wrong. My multiplayer design needs a lot of infrastructure.

Did any of it get done?


## During the vacation
---


I implemented [Prim's algorithm](https://en.wikipedia.org/wiki/Prim%27s_algorithm) to generate mazes of X by Y dimensions. The maze would then be drawn on screen with the origin to the top left (0,0) and the goal to the bottom right (x,y), I also implemented a cursor that can navigate the maze.

At this point the gameplay loop of "solving a maze" was done! This is where I should've stopped, polished what I had and released the app as a single-player procedurally generated maze app so that I could meet the time limit.

![](/images/blogs/image.30.png)

Unfortunately, I didn't. I wanted to stick to the original idea, which was a real-time multiplayer maze racing game. Even worse, I wanted it to be a successful multiplayer app.


## My Multiplayer Design
---


As I mentioned above, I wanted this app to become successful. I did a bunch of research to see what that even means. I decided it was user retention and Rewarded Video Ad/IAP revenue.


This influenced the design of my app to include ranking, coins, skins, a store page, and profile stats.


I got right to work developing an app menu to support the features I just listed and as I made more progress I realized that this is not what I enjoy about videogames. I really enjoy creative mechanics and themes, not grinding for currency to purchase a shiny skin.


As a result, I've decided to stop right here on this app. The menu is unfinished and the only button that works is the play button, but that's okay. I want to focus on making games that are fun to play, even if you only play the game once. 


![](/images/blogs/AppMenu.gif)


Then, why make games?


## Games for FUN
---


This is where my current decision comes into play. I keep making decisions about my projects based on the project's chances to succeed financially, which is stupid. I love videogames because they are a form of art, not because they are profitable. I think it's time that I shift my mindset accordingly.


I'm going to keep making games, I just need to drop the desire to immediately profit off of them. I hope that one day I capture an audience that enjoys my games, but until then I can take my time and explore different themes and mechanics.


I figure I should at least make my prototypes playable, so I plan to release my games right here on my website! I figure I can build my games for WebGL and deploy them right on my projects page. This will allow me to focus on small, but interesting ideas without the need to seek approval from any storefronts. I can also drop the pressure to make money or retain users.


Maybe I'll start a patreon or just add a Donate button if anyone wants to support me. I'll update you guys as I make more progress.


Thanks for reading,





Hey guys,


I've been working on my game prototype lately and I've come to realize that this project is going to take a long time to produce. It's important to me that this game is:


* Developed in 3D
* Has interesting synergy between movement & combat with the enemy AI.
	+ Feature boomerang gameplay
	+ Feature sword/hack-n-slash gameplay
	+ Movement is based on acceleration/momentum
* Feature a beautiful aesthetic
	+ Well designed art-style / 3D models
	+ Animations
	+ Particle Effects
	+ UI / HUD Art
* Procedurally generated waves of enemies


The plan was to make this game a single-player 3D rogue-like style game where you fight waves of enemies for as long as you can. I don't want to compromise the scope of the project, but I also don't want to develop a game on my own for several years. I think working on a project for a week, then moving on is more reasonable to do alone instead.


As a result, my new plan is to start releasing apps/products at a much smaller scale, but far more frequently. In fact, I plan on spending no more than 1 week taking each idea from concept to launched on the App store/Platform of choice. This should force me to scale down the scope of my projects and therefore gain more practice taking an idea to production. Once I've got a bit of an audience, or maybe some capital I might come back to this project.


## Wrapping Up The Prototype
---


I'm not done working on the prototype yet, though! I won't start my new week-long iterations until next week. I plan to "finish" the prototype by then. This just means refactoring the player's code just so that it's easier to read and then designing a new menu with a level picker. All of the systems have already been writen, so now we just need to polish the aesthetics and design a couple of levels as a proof of concept. I think I can do this by the end of the week 🤞


My latest changes to the code had me rewrite the enemy AI script. I initially built my first enemy using only 1 script. This script controlled the nav behavior, HP behavior, combat and wind-up. I went ahead and split them up into their own code files so that adding variety to enemy behaviors will be as simple as removing the base script, to be replaced with a derived script.


![](/images/blogs/image.28.png)


Once I refactored the code to facilitate variety in enemy behaviors, I implemented the shooter enemy. This enemy shares the exact same follow, HP, and Wind-Up script that the launcher enemy has. The only difference is that I removed the "Launch" script and replaced it with a "Shoot" script. Adding variety to enemy behavior is a simple request now, considering that their behavior is made up of lego-pieces of code. Developing a turret enemy would simply have me add the "Shoot" script, along with the wind-up and perhaps an "HP" script so that it can die. The only difference between a turret and a shooter enemy is that the turret would lack a "Follow" script.


![](/images/blogs/0.9.1.gif)


I *might* add more enemy variety, depending on what the levels I design need. Could also tweak the values that control the enemy's speed, damage, knockback, etc. That's not my top priority, though. In order to consider this prototype complete, I need to clean up the code that controls the player's movement, combat, etc. just because I wrote the code but never cleaned it up for readability. I expect that refactoring the player's code will only take a day.


Once the player's code has been refactored, I will clean up the main menu and design a level with a few waves in it. The reason that I want to develop at least one fun level is that if I ever share this prototype with an interested party, I want it to communicate the gameplay that I had in mind when designing this prototype. If I have more time, I might even design another level or develop more variety in enemies. In truth, I'd really love to work on this project full time. I'm just going to leave it at the prototype phase for now, but I'm going to leave it in a place where I can easily pick it back up and continue the project.


I'll make sure to release the demo once I've developed the changes this week. :) Thanks for reading.





Hey ya big nerds!


If you stop by my blog every once in a while, you'll notice that I have not posted a blog in a minute. That's because I had to take a step back and really think about what I was spending my time on.


As a result, I decided to dive into game development. It's been my dream since I was a kid, but now I am equipped with all of the skills and tools that I need to develop a videogame. I've been developing a prototype over the last few months, not much time invested though. I've been working on this idea while watching meme compilations & netflix with my GF.


## The Design
---


I personally love games where movement is a requirement to do well and speed is both a blessing and a curse. I also took inspiration from anime. Characters in anime tend to have the ability to move so gracefully, yet powerfully. Watching a master do their work is almost like watching a dance or performance and I feel the need to capture that in a videogame experience.


I've never personally watched Naruto, but I love this fight scene where a single person fights an army and uses grace & tactics to defeat overwhelming odds. The first 2 and a half minutes of the video are great examples.  

  


 I started with a design document that detailed my game's key pillars: **Speed** and **One-Versus-Many**. I use these key pillars as reference whenever it comes time to make a decision. For example, should the player accelerate? What about during a jump, can they change their velocity? etc. etc.


## The Prototype
---


Although I've been working on this project very casually, I feel that I've made some real progress towards something resembles "fun". Throughout development I have changed my original design doc. I have not, however, changed my key pillars. I started by deciding on the player's movement & went with a double jump. This was super early, when I just got the camera to work in conjunction with the player's movement/velocity.


**First Build:**  

![](/images/blogs/0.0.1_Movement.gif)


Once I got he camera working, I fleshed out the movement controller for the prototype. I also decided that a mid-air dash would be a critical part of the player's moveset.


**Third Build:**  

![](/images/blogs/0.3.0_PrototypeRoom.gif)


I tweaked the movement variables around until I got a feel that I enjoyed. I've spent quite a bit of time just running around the prototype map that I've laid out.  

Once I got to designing the enemy AI, I had to make decisions about the enemy behavior. I decided to focus on melee for now and develop ranged enemies later.


Since the game is focused on speed, the enemies will penalize you for standing still. They will run up to you, charge up and finally pounce once their charge-up is complete. If you walk too far while they are charging up, they simply start chasing you again. I also wrote a simple healthbar for enemies as a placeholder. It's literally a red cube that scales down on the X axis.


**Fourth Build:**  

![](/images/blogs/EnemyWindup.gif)

*At the point of recording this GIF, I had not written the "pounce" part of the code.*


I'll go ahead and skip to Build 8, which is the build I'm on now. I wrote a pretty flexibile wave manager script that allows me to control which GameObjects will spawn, where they will spawn, and in what order/quantity. This is going to be great. I've decided that for the prototype, I won't build any procedurally generated waves or levels. I'll simply create a single level with 10 waves.


I also gave the player a sword, and I've given them 2 boomerangs. The boomerangs come back if they hit an enemy or if they've been in flight for a while. I still need to add knockback, stuns, and particle effects so that the boomerang and the sword have more impact when they hit. (Please disregard the healthbars, those are just placeholder.)


**Eighth Build:**  

![](/images/blogs/0.8.3.gif)


None of these decisions are set in stone, though. I'm still in the prototype phase. Not even alpha. This game doesn't have a name yet, nor does it have any concrete lore or story/setting. I've been focused on making the mechanics enjoyable. So far, I've just called this "Project: Anime Motion". I've implemented all of the key systems that are required to consider this project a game, though. It has a main menu, a healthbar/lose condition, and a win condition (kill all enemies that have spawned). Now, begins the tweaking & tuning...


## Coming Soon™
---


Soon™ in this case really means *BRUH, IDK!* 


LOL, At this point, I only have a bunch of systems that are going to be useful. You can technically load in, start the waves and win/lose the game, but there are only 2 slow enemies that spawn and the "arena" you're fighting in is empty and bland.


My wave system works well but I need **enemy variety.** Like I mentioned earlier, I only programmed melee enemy AI. I need to refactor my code to support ranged & melee enemies while being extendable enough to eventually support sub-types of ranged & melee enemies. Once I've built out a "ranged" and "melee" prefab, I can develop a few varients of each type by tweaking some variables on my script, such as attack damage, movement speed, health & how long it takes to charge up a pounce.


 I also need to **design a level**. It's no good having a wave spawning system if there are no interesting places to spawn enemies. This one is going to be a bit difficult because I will need to account for the enemy spawns ***and*** the player's speed / verticality. I've also considered adding some level obstacles or maybe some interactable objects, but I feel like those are not critical to the game's key pillars so they might not make it into the demo/prototype.


I'd also like to add ***a leaderboard***, and it would definitely play well with the key pillars because we can time the player and encourage the player focus on their speed. However, considering that the prototype is primarily to test the game concept in execution, I don't think it's critical to the prototype. I'm just going to display the player's final time on the victory card so that players can compete among friends by sharing screenshots. By doing it this way, we can still encourage players to compete over time, without the need to write code that supports a local high-score database.


If everything goes well, I'll make some interesting & challenging enemies and a map that compliments them. I'll also create some interesting group compositions and spawn waves that are challenging, but fun to plow through. Once I prepare a level of my design, I'll release the prototype/demo and ask for some feedback. Hopefully, I'm onto something 🤞.


Thanks for reading!





Finally, I got around to upgrading the SEO of my website. Read the blog where I discuss the problem by clicking [here](/blogs/seo-upgrades-incoming).


Unfortunately, Disqus is still showing the old SEO images. I don't know how long it'll take for the page cache to clear, but I hope it's soon. I put a lot of effort into each blog post and thumbnail as a one-man show, haha.


## Next Steps
---


Now that sharing links to my blog provides a richer link, I want to include more reasons for people to share my blog posts. Let me know if you have any ideas for engaging the comments and sharability of my posts. At the moment, I've been treating my blog as a public diary and a *potentially* helpful source of knowledge for random googlers. If there are any particular topics you'd like to read about, please leave a comment :)


Lately, I've been developing a prototype for a game concept I've wanted to bring to life for a couple of years now. Unfortunately, I have not developed any art assets so all I have are a bunch of capsule objects in Unity3D with some custom C# scripts. I'm going to continue working on my game prototype until I believe the game is worth pursuing. I will definitely post what I have here when I arrive to that point.


I figure that I may want to write more tutorial-centric articles if I want people to share my posts. Again, let me know if there is a specific topic you'd be interested in.


## Life Update
---


I was working with Brandon on Potato Potato (card game) for a majority of 2020 and the first half of 2021. Brandon is going to continue to work on the card game without my assistance, but I am confident that he will be able to develop the game using what we were able to generate for testing purposes. Keep an eye out for "Potato the card game", coming soon™... haha. 


The reason I stopped working on Potato Potato is that I've decided to pursue my own game project. I recently worked on a small project that inspired me. The project itself was not inspiring, but rather the fact that I was able to make a fun, multi-player project within a weekend was. It showed me how far I've come as a developer and what I am truly capable of on my own, even with limited resources.


As much as I was enjoying working on the card game, I just know that I would much rather develop a videogame. Who knows, maybe I'll help Brandon migrate Potato Potato to mobile & steam ;)


 


Thanks for reading,  

Write to you soon :)


- Raul





I've been working on some new infrastructure at [Professional Financial Solutions](http://pfsaccounting.com) lately. It's been pretty fun and interesting stuff.


Covid has forced many businesses to embrace remote work, but luckily, we never had any problems with remote work in the first place.  

The use of remote work does complicate things in regards to system administration though.


For example, we recently started leveraging Active Directory to manage our users and computers, as well as apply a policy-driven SysAdmin solution. This means **that** all computers need to have a secure connection to the Domain controller.


I used OpenVPN to host a VPN gateway on-site and in the cloud where we are hosting a DC. This allows me to route packets between our on-site network and the cloud network seamlessly. The problem arrises with users who occasionally work from home. They need their connection to be available regardless of what network they're using.


## OVPN Connector
---


Luckily, all we need to do here is install OpenVPN Connect on the client computers. 


OpenVPN Connect comes with a binary called **ovpnconnector.exe** in the Program directory. You can read more about it [here](https://openvpn.net/vpn-server-resources/use-openvpn-connect-v3-on-windows-in-service-daemon-mode/).  

Long story short, ovpnconnector is a command line utility to install the OpenVPN Connector service. Simply create a client ovpn file and name it **ovpnconnector.ovpn**.  

Then, drop that file in the same directory as the executable.


Finally, using an elevated command prompt, navigate to the program folder and run:



```

ovpnconnector.exe install
ovpnconnector.exe start
```

Those commands will install the OVPN profile and run it as a service. This will ensure that the VPN tunnel is initialized before the user has even signed in.


It's perfect for use in an Active Directory environment.


Hope this was helpful to some of you,  

Thanks for reading!





I've spent some time on writing these blogs now and officially have more than 1 page of blog posts!  

I've also noticed that it automatically recommends posts to users under the comment section.


Honestly, this is great!  

It allows me to focus on writing blog posts and less on implementing a "recommended posts" system.


This gave me the idea to look into more service providers, so I may incorporate more.  

Let me know what you think, I am undecided there.


## But, there's a problem...
---


At the moment, all of my posts have the same thumbnail and the same description!


![](/images/blogs/SEO.png)


The reason for this is because when I wrote the header code for my website, I hard-coded values into the **<head></head>** section.  

Specifically, I hard-coded the **<meta>** tags.


If you're familiar with web development & HTML, you are aware of the **<head></head>** section of most web pages.  

For those who are not, this is the section where you detail information about the web page. This typically includes stuff like:


* Page title
* Javascript code to run
* Style code to apply
* Image, Audio, Video assets to load
* *meta data*


The meta tags store information about the page you're visiting.  

On the modern web, we take advantage of [OpenGraph](https://ogp.me/) to allow services to thumbnail our web pages on their platforms.


Whenever you share a link to one of my web pages on Facebook (for example), Facebook queries my website, reads the <meta> tags and generates the thumbnail using that data.  

This is why you see youtube thumbnails automatically appear when you share their links on Facebook and Discord. 


*The beauty of OpenGraph and HTML Meta tags.*  

Let me know if you guys have any questions!


*Also*, I'll fix that isssue **Soon™**  

Thanks!





Most people are using cloud storage providers as their file hosting solution. This makes sense because most people are storing simple file types like Word docs, Powerpoints, Spreadsheets, etc. and they are storing them in such small quantities that it does not incur any costs.


I believe [Google Drive's free](https://www.google.com/intl/en_in/drive/) tier provides 15GiB of storage for free! This has been great for simple file storage. I use Google Drive to write Docs and Sheets that need to be highly available and easily shared. Google Drive is awesome for collaboration. I also use Google Drive to store archived data. Files that I don't need to use anymore, but I don't want to lose.


![](/images/blogs/image.17.png)


## Reading and Writing
---


As you see above, the pricing for Google Drive is very reasonable. So, why don't I want to use it?


Well, the truth is that I would love to use it, but I want to be able to read and write to my cloud files as easily as I would interact with them locally. This is simple to do with Excel, Word docs, and other supported file types because we edit the files using their web editing tools, such as Google Docs and Google Sheets. Non-office files, however, are not as seamless. 


Binary files or files that are not associated to an office suite are simply uploaded and downloaded from Google Drive. You cannot simply edit them in the web browser. This is not ideal as we must download the file, apply our changes, and then re-upload it to overwrite the old cloud copy.


It's possible to use the [Google Drive sync software](https://www.google.com/intl/en_ca/drive/download/), but in my experience that software is not as customizable as I had hoped and it was slow when downloading/uploading files.


## LAN Storage
---


I recently wrote a blog post about [turning your Rasberry Pi into a Network Attached Storage device using Samba](/blogs/make-a-file-server-with-rasberry-pi). Unfortunately, my external hard drive pooped the bed about 2 days after posting that blog.  

I shouldn't be too surprised. This is the same hard drive that pooped the bed back in 2015, so I think this HDD is the problem.


I realized this meant I would have to either buy a new External hard drive, or maybe some USB to SATA adapters for some hard drives that I have lying around.  

I came to the conclusion that it might be best if I simply buy a NAS instead.


I figure that purchasing a NAS will require less management from me than building my own. Also, using SATA rather than USB will provide faster read and write speeds to the HDD.  

This means that the rasberry Pi can focus on hosting VPNs, DNS servers, and other micro services while the reading and writing is handled by the consumer-grade NAS device.


I'll let you guys know when I decide to buy one and what I based my decisions off of.


### Use Case


I plan to set up the NAS with 2 Hard Drives. Ideally, the NAS will support RAID 1, so that we can prevent disasters such as a hard drive failure.  

We'll split up the disk(s) with 2 partitions. One as an Archive partition and the other as a "working set" partition.


The archive partition's purpose is to store files that I don't plan to access for a long time.  

We will also store device backups in that partition. These backups will be generated using [duplicity](http://duplicity.nongnu.org/).


Conversely, the other partition will be mounted to my devices as the set of files that I will be reading and writing to often. These will likely be assets and documents that are related to an active project.  

I can access these files remotely thanks to the VPN server that I set up on my other [blog post (read more here)](/blogs/host-a-vpn-on-your-rasberry-pi).


## Back Up Your Backups!
---


*Ah, disaster recovery.*   

Even if you back up your files, you gotta back up those backups!.


Our NAS will have protection against hard drive disasters via RAID 1. If you're unfamiliar with RAID 1, we're simply using 2 Hard Drives to act as 1. The second hard drive exists as a backup of the first.  

The benefit of RAID 1 is disaster recovery and also faster HDD reading. This solution provides secure, reliable, and fast file access as well as locally managed backups.  

The problem now is that all of our eggs are in one basket! if someone breaks or steals this NAS, all of our backups and shared files will be gone.


We've gone full circle here!


The final step is to back up our NAS to a cloud or offsite provider daily/weekly. I haven't looked into these providers yet, but I will perform some research to determine the most  

cost-effective file archive/backup service. Technically, it doesn't even need to be a cloud service. As long as you are back uping to an off-site storage, that should be enough.


The cloud backup service does not need to be read or write quickly. Nor do I need frequent access to the files. It simply needs to be realiable and cheap.  

We don't expect to use the cloud storage provider. It will only come into play if anything happens to the NAS.


If you have any questions about file backups and storage, please let me know.  

Hope this was helpful!!





As I discussed in [my last post](/blogs/make-a-file-server-with-rasberry-pi), I set up my Rasberry Pi as a network attached storage. It's currently running a Samba server so that it behaves like a simple Windows file share.  

![](/images/blogs/image.15.png)


Each share has custom permissions and they require password authentication.


At the moment, this file server is only available locally, and I have absolutely no intention of opening it to the internet.  

So, I am going to set up this same rasberry Pi as an OpenVPN server.


I don't have a static IP address, so I will also configure my pi with a DDNS service.  

Ideally, a ***free*** DDNS with Linux support.


## No Static Ip? No Problem...DDNS
---


If you're not familiar with networking and stuff like DNS, you may want to read into it before trying to set up a VPN.


To summarize, internet addresses tend to use IPv4, so you see addresses that look like **1.1.1.1** and **172.217.4.174**.  

However, when you use the internet, you typically just type *google.com*, or rgonzalez.tech into your address bar.  

That is the purpose of the DNS service. It translates domain names into IP addresses.


Unfortunately, the IP address that my internet service provider gives me tends to change every few weeks, so simply pointing a domain like **vpn.rgonzalez.tech** to an IP won't be enough.  

I could purchase a static IP address to resolve that issue, but my VPN server won't be mission-critical and I'm trying to keep my costs low, so a DDNS should suffice.


I'm going to go with [NoIP](https://my.noip.com/), simply because I have used them before and I don't mind clicking on their email once a month for free service.  

🤔 Make me wonder if they're simply milking me to teach robots about fire hydrants and traffic lights.


Anyway, I went ahead and set up my new DDNS address and got started setting up their [**DUC** (**Dynamic Update Client**)for Linux](https://www.noip.com/support/knowledgebase/installing-the-linux-dynamic-update-client/).  

It was pretty simple, but that's probably because I already had **make** installed on my Pi.


## Prepping The Network
---


Now that I am going to use this Rasberry Pi as a host, I need to configure it with a static local IP address.  

Again, this Pi has Ubuntu 20.04 installed. I just need to change the IP to static and configure the subnet/gateway.


In the future, I may also have this lil' Pi run a DNS service. We'll see how it handles routing VPN traffic first, I have another Pi ready to go in case we need more juice.  

(Could also use my Windows gaming PC to host some stuff on Hyper-V. I never really turn it off anyway. We'll see.)


I digress, the point is that this Pi needs to be set to static. I'm accessing the Pi over SSH, so I don't have access to the GUI.  

I followed [this guide](https://linuxize.com/post/how-to-configure-static-ip-address-on-ubuntu-18-04/) to change the local IP and it worked flawlessly. 


### Port Forwarding


So, the Rasberry Pi's IP won't randomly change anymore, so we can reliably point network requests to this rasberry Pi when they arrive to our home. I'm going to host the VPN service on the default port, which is 1194 & I'm going to use UDP, so I simply create the port forwarding rule in my router and move on!


If you need any help with port forwarding, I suggest you google for your router's model number. For example, if you're using a **Spectrum F@stModem**, you should enter:


**Spectrum F@stModem + "Port Forwarding"**


into google. Also, this link might be helpful: <https://portforward.com/>


You can also use online [network testing tools](https://ping.eu/port-chk/) to see if the ports are open.  

Keep in mind that if you forwarded the port, but the rasberry Pi isn't hosting a service on that port yet, the port will read as "closed".


In fact, the way that I'm going to set up my VPN auth, it will always read as if the port is closed. (Extra security!)


## Hosting a VPN
---


So, the first thing I'm going to do with my rasberry Pi is enable **IP forwarding**. That is what allows a computer to behave like a router.   

Typically, when a computer receives a network packet that does not belong to it, it simply drops the packet. If you enable IP forwarding, the computer will attempt to send it through the router/interface it was destined for.


I'm enabling this because I want any device on my home's network to be accessible through my VPN connection. Vice versa, I want devices on my home network to communicate with my VPN clients seamlessly.


Here's a quick article that shows you how to [enable and disable IP forwarding in Linux](https://linuxconfig.org/how-to-turn-on-off-ip-forwarding-in-linux).


Now that our Rasberry Pi can behave like a router, we can install openvpn and create a PKI followed by the server configuration.  

This stuff is covered by the [OpenVPN HOWTO 2.x guide](https://openvpn.net/community-resources/how-to/), so I won't explain it further. I genuinely suggest you set aside an evening to read the HOWTO thoroughly.


### Creating the PKI


Following the guide that I linked above, we use [easy-rsa3](https://github.com/OpenVPN/easy-rsa/blob/master/README.quickstart.md) to create a PKI and Certificate Authority.



```

 ./easyrsa init-pki
 ./easyrsa build-ca
```

The certificate authority is literally just a Certificate and a Key that will create all further server cert/keys & client cert/keys.  

Once the CA is built, we can use easyrsa to build the server certificate/key & any new client certificate/keys. Also, don't forget to generate a ta.key for tls-auth.  

it'll make your server more secure.


My favorite part of this step is generating the DH.pem file.  

![](/images/blogs/image.16.png)


### Creating the Server Config


To be honest, we're not really "creating" the server configuration so much as we are modifying a template.  

When you install OpenVPN, you might find the server.conf file in **/usr/share/docs/openvpn/sample-config-files/**


Again, the guide above mentions this stuff, so I'll move on. You can copy the sample server.conf file and modify it to suit your needs.


The important changes in my case were:


* The location of my certificate & key files
* I decided to change my tunnel subnet to **172.20.0.0/24**
	+ I doubt I'll run into a WiFi network using that subnet address, 😅
	+ change **server 10.8.0.0 255.255.255.0** to **server 172.20.0.0 255.255.255.0**
* I'm also going to inform clients that this server can contact the **192.168.254.0/24** network.
	+ Add the **push "route 192.168.254.0 255.255.255.0"** command to the server config.
	+ This will allow clients to communicate with any device on my home network.
	+ It will also require that I create a static route in my home router.
		- Route **172.20.0.0/24** to my Rasberry Pi
		- This is because all client devices will be in the tunnel subnet.


That should be enough to run this server. I wrote a quick shell script to run openvpn with my server.conf file.



```

#!/bin/bash
/usr/sbin/openvpn /etc/openvpn/server/server.conf
```

Last thing to do was add that script to crontab. @reboot /etc/openvpn/server/launch\_vpn\_server.sh


## Connecting To the VPN
---


Just as the HOWTO guide mentioned how to create the server certificates and other important files, it also explains how to  

create client certificates and client configuration files.


The first device I'm going to add is going to be my phone. Luckily, OpenVPN has a free mobile app called "OpenVPN Connect".   

The same name is used for the windows desktop OpenVPN client.


We generate the client certificate/key and configure the client.conf file to use these credentials. Finally, we rename client.conf to client.ovpn.  

To simplify the installation of these configuration files, we're going to embed the certs and keys into the .ovpn file. You can [read more about that here](https://samuel.dalesjo.net/2018/02/05/embedding-certificates-in-ovpn/).


Finally, install the .ovpn files into your devices and you're done! You have a private VPN.


Please, let me know if you have any questions.  

Thanks for reading!





I've been doing a lot of networking & security lately for my day job. This includes stuff like creating static routes, and a simple PKI for an OpenVPN Site-To-Site implementation.


This has been a great learning experience for me, as I finally put a lot of theory into practice with regards to networking.  

I'll write a post about **OpenVPN** *Soon™.*


## My Own Cloud
---


Before I concern myself with creating a personal VPN for my own home, I figure I should first host a service that I want to make accessible from the WAN.  

The first thing I'm going to do is create a Samba share, AKA a Windows **N**etwork **F**ile **S**hare.


Nothing fancy. I have a rasberry pi that is currently hosting a few IRC & Discord bots. I also have a 1TB USB HardDrive lying around, not being used for anything.  

The story with this hard drive is that I originally bought it as a teenager so that I can develop games "on the go". The plan was to host all the assets & code on this single hard drive... with no backups.


As you can guess, the hard drive pooped the bed during my 3rd learning project. I learned the value of backups after that, hahaha.  

Mind you, this was before I learned about Git & source control.


### Let's get started


We can create a file share using practically any computer. I've plugged my USB hard drive into my Raspberry Pi.  

My Pi is running ubuntu 20.04 LTS at the moment.


First thing I had to do was figure out how to mount this thing as a storage device on my Pi. 🤔


### Mounting a USB HDD on RasberryPi/Ubuntu


I immediately ran into an issue with my Pi not detecting the USB hard drive. I think this was caused because I had plugged in the HDD and left it alone for several minutes.  

I noticed that the HDD was blinking in a very slow pattern. So, I unplugged it and plugged it back in and it worked :)


I was finally able to see it represented in the output when I type **lsblk**  

![](/images/blogs/lsblk.png)


The next step is mounting the disk to a folder in /media. I'm going to call mine **external\_drive**. So, I made the folder in /media.  

Mounting the disk is as simple as typing:



```

sudo mount /dev/sda1 /media/external\_drive/
```

but we want it to stick around even after we reboot the Pi. So, what we actually want is to mount it automatically on startup.  

According to [Linux Config](https://linuxconfig.org/howto-mount-usb-drive-in-linux) , we should use the disk's UUID, which we can get by entering:



```

ls -l /dev/disk/by-uuid/*
```

In my case, I can see that the disk I'm trying to access is UUID FC8...  

![](/images/blogs/uuid.png)


So, all I have to do now is edit the **/etc/fstab** file, which you can read up on [here](https://man7.org/linux/man-pages/man5/fstab.5.html).  

I ran into some speed bumps, but after formatting the HDD to ext4, auto-mounting was successful. 👍


### Hosting a Samba Share


This is the fun part. Now that the HDD is in place to be used by the Rasberry Pi, we can host an SMB server.  

I created some users via **sudo useradd <username>** and then set their Samba password using **sudo smbpasswd -a <username>**


Finally, created some folders in the /media/external\_drive directory and updated the **/etc/samba/smb.conf** file to include these lines:



```

[Public Files]
path = /media/external\_drive/public
writable = yes
browsable = yes
read only = no
guest ok = yes

[Our Share]
path = /media/external\_drive/our\_share
writable = yes
browsable = yes
read only = no
guest ok = no
valid users = user1 user2 

[My Files]
path = /media/external\_drive/share1
writable = yes
browsable = yes
read only = no
guest ok = no
valid users = user1 

[Your Files]
path = /media/external\_drive/share2
writable = yes
browsable = yes
read only = no
guest ok = no
valid users = user2 
```

Finally, restart all of the services for SMB.



```

sudo systemctl restart smbd.service
sudo systemctl restart nmbd.service
sudo service smbd restart
```

If you need any help, check out the article that guided me the most: [link here](https://help.ubuntu.com/community/How%20to%20Create%20a%20Network%20Share%20Via%20Samba%20Via%20CLI%20(Command-line%20interface/Linux%20Terminal)%20-%20Uncomplicated,%20Simple%20and%20Brief%20Way!).  

If you have any questions or want me to elaborate on any of this, please leave a comment!


Thanks for reading :)





I've been working on a secret project lately. It's something that I plan to support & deploy properly. That's not what I'm writing about today, though!


In my research to properly support & deploy this project, I've been learning a lot about docker & containerization, as well as Kubernetes as a solution to deploy these containers. In the mean time though, I'm still prototyping this idea so I figured I'd look for something simple.


### Google Cloud


Once my code was updated to support Docker, I googled around and decided to try Google Cloud. So far, it's been great. At first, I had a hard time understanding the differences between the Compute Engine, Kubernetes Engine, and "Cloud Run" services. After a weekend of reading, I finally understand the need for each service. I decided I'd spin up a small VM in the Google Compute Engine service running a container-optimized OS. 


Once I did that, I left it alone and moved on to other stuff, like implementing some CI/CD with Google Cloud Build + Github. That was pretty simple to set up.


By the time I was done preparing Google Cloud for my code, I had created a:


* **Cloud Build**
	+ Google **Container Repository**  
	
	 (To Store Auto-Generated Docker Containers)
* **CloudSQL** server
* **Compute Engine** VM instance
	+ *E2-Micro*
	+ *Running the latest image in my repository*


My lovely girlfriend brought up pricing, trying to gauge how much I was investing in this prototype.   

It was my understanding that hosting my code would be free, as is Google's *Free Tier*, but that lead me into some more research.


### Free Tier


Looking at the [Free Tier](https://cloud.google.com/free) link, I noticed that they had no mention of CloudSQL. Even on the CloudSQL "Pricing" page & service page, there was no reference to a free tier.


Their free tier page does, however, mention "Firestore", which might be good enough for what I need and it says I get 1Gb of storage for free. I think that should be plenty considering I don't expect this thing to blow up overnight.  

The best part about using cloud services though, is that I won't have to worry in case it does!


So, I'm going to change my approach and store my data on their document-based databases instead of SQL. I also noticed that my VM was too large! Their free tier only supports up to 1 f1-micro VM, so I had to size that down.


Their offerings are very generous, though, so I should be able to build my tech off this till it becomes popular!


 


*Thanks for reading!*





I've been playing Cyberpunk 2077 lately. A good friend of mine purchased it for me in late December as a birthday gift.


![](/images/blogs/Cyberpunk2077Logo.gif)


I won't lie, I have definitely been enjoying the game and the fun mechanics that the game offers such as the *Quickhacking* feature.  

I don't spend a lot of time on gaming anymore, so it's taken me a while to really absorb the game.


## Preface
---


Much like everyone else born after the 90s, I've been gaming since I was a kid. Now that I've got more responsibilities and hobbies, I don't set aside a lot of time to enjoy gaming as much as I used to.   

I have spent ~36 hours on the game over the course of a month and a lot of that time has been spent on side missions and random events. Personally, I love role playing games that allow me to tackle my character's problems however I believe my character would. Whether that is talking my way through something, stealthing through it, or just going in guns blazing. Games like [Fallout: New Vegas](https://store.steampowered.com/agecheck/app/22380/) really capture that feeling.


This game does have skill trees and some exclusive dialogue choices, but they are somewhat lack luster. Oh, and the bugs...



## My Opinion
---


While I do enjoy the game, it feels like I'm playing a game in it's late "early-access" phase. There are numerous glitches with a lot of the systems that are supposed to run in the background.


For example, you will often run into duplicate NPCs or duplicate cars that are following the exact same pathfinding. Also, I've ran into issues where a random NPC is walking down the street, and as I go to approach them, they just fade out of existence right as I approach. Seeing something that you were going to interact with simply fade out of existence feels cheap and it feels like the game robbed me of an interaction. 


### The Good


There are many things that Cyberpunk does right. The side characters and their story lines are actually interesting and they give you a chance to get to know each of the side characters and the things that they value. This is great for roleplaying and feeling invested in the quests and their characters. I love the driving sequences when you are the passenger seat. It's nice that the game gives you the opportunity to simply take in the sights and have an optional conversation with the character in the driver's seat. 


CDPR also did a great job designing the levels in such a way that they can be completed in your preferred style of gameplay. Your "gameplay style" is really a question of what weapons/cyberware you choose and how much you're going to invest into Quickhacks. You also have the option of choosing between stealth and combat.


You will also find objectives that will require you to solve a problem, such as getting past a blocked door. There are clearly many ways you can go about this, but the game might present you with an option to shove the door open **IF** your strength level is high enough. The same thing applies to some of the other stat points.


Mechically, the game feels great to play. The gunplay was done really well, especially when used alongside Quickhacks. The melee combat is definitely better than Skyrim's, haha. The NPCs react to being hit, so it doesn't just feel like you're striking a hitbox. Finally, there is always something to do or work towards. Often times, *TOO much* to do. It can be overwhelming, but luckily the game gives you space to play however you like and take your time.


### The Bad


Continuing from my last point, there are many times when your character has the opportunity to say something that is only available to you because of a stat point, like your "Cool" stat or "Reflexes" stat. Often times, your character's origin story will also come into play, but it all feels very shallow. Most of the time, these exclusive choices do not actually contribute to your story, but rather allow you to say something witty or *display your tech knowledge* by blathering on about some techno-babble before being forced to pick the "Continue the story..." dialogue option, anyway. The outcome of these conversation rarely ever depends on your exclusive dialogue choices.


Also, ***THE BUGS! Plenty could be said... about the immersion breaking bugs...*** but I won't waste your time. You can view the video I linked all the way at the top if you wanna see how bad it gets. To summarize, the developers have built this grand world to immerse you, but the systems that it relies on break often which means that it often fails at the very thing it sought out to do.


to summarize...


## Ending note
---


![](/images/blogs/CyberpunkPatchesMeme.jpg)


While the concept was great and the design was streamlined for immersion; bugs frequently break that immersion. When that immersion is broken, the game feels like an aesthetically pleasing shooter/stealth game with surface-level dialogue branches and many unaddressed bugs and AI glitches. It's a high quality product that often fails to immerse you as much as it could. *Darn, Bugs!*


This is **NOT** to say that the story or sidecharacter questlines are surface-level stories. They touch on intersting topics about Sci-Fi, your "soul",  AI, etc.


I am simply saying that the "decisions" you make, only make surface-level changes to the story.  

Frankly, I'd wait for all the bugs to be fixed before spending any more than $30 on this game. If the bugs have been fixed, I could see this selling for $40 in a year or two.


Thanks for reading :)  

Let me know what you think in the comments.





I am working on a card game with a co-worker, we started on this together last year.


Once the game loop was designed we made a list of the cards that would need to exist, as well as their text, names, etc.  

We had practically made a spreadsheet. From that, we agreed on a prototype card layout.


![](/images/blogs/PotatoCards.png)


I started by opening up GIMP to begin editing a ton of card images together. I got through 2 cards and realized I'd much rather automate this.


Manually updating the cards and creating the printout was too much work for me.. It would also allow human error to introduce inconsistencies and...  

*it would require Quality Checking.*


So, my excuses are: 


* **Inconsistant**. If I don't use a style guide or some other pre-defined ruleset, our cards will be inconsistent.
* **Tedious**. A lot of the clicking is going to be repetitive and will need to be done every time that we change a card.
* **The Print-Out**


### My Solution


I wrote a program in C# that takes a path to a CSV file as input.


This is the master spreadsheet and it references other spreadsheets relative to it.  

  - images/*.png  

  - spreadsheets/*.csv  

  - card\_data.csv


spreadsheets?
## generating cards
---


So, using the text, color codes and names in the spreadsheets, I can generate the cards.


![](/images/blogs/PotatoCardGen.gif)


I do this for every card of every type and generate a bunch of images.


If you're wondering how what the code looks like, we're using the Graphics class in C#.



```

Bitmap canvas = new Bitmap(CARD\_WIDTH\_PIXELS, CARD\_HEIGHT\_PIXELS);
Graphics marker = Graphics.FromImage(canvas);

```

Then, when we need to draw to the canvas, we use the **marker** object. When you want to save the image, you save the **canvas**.


So, the Header panel and icon is drawn like this:



```

Brush panelBrush = new SolidBrush(BACKGROUND\_COLOR);

Rectangle headerPanel = new Rectangle(0, 0, CARD\_WIDTH\_PIXELS, PANEL\_HEIGHT);
marker.FillRectangle(panelBrush, headerPanel);

// Add the Card Type Icon
string iconPath = "<icon path>";
Image cardTypeIcon = Bitmap.FromFile(iconPath);

Rectangle iconPos = new Rectangle(PANEL\_MARGIN\_PIXELS, PANEL\_MARGIN\_PIXELS, PANEL\_MARGIN\_PIXELS*3, PANEL\_MARGIN\_PIXELS*3);
marker.DrawImage(cardTypeIcon, iconPos);

```

You can find more info on the [Graphics](https://docs.microsoft.com/en-us/dotnet/api/system.drawing.graphics?view=dotnet-plat-ext-5.0) and [Bitmap](https://docs.microsoft.com/en-us/dotnet/api/system.drawing.bitmap?view=dotnet-plat-ext-5.0) classes online.


Put it all together and what do you get?
## The results
---


Finally, once the code runs and all the cards have been generated, we can add them to the printout as many times as necessary. The code will automatically position the cards so that they are printed and ready to play!







If you guys want me to elaborate any any of it, leave a comment or contact me!


Thanks for reading!





Once I was able to connect to [Twitch](http://twitch.tv/) using IRC, I realized I could automate a ton of stuff.


To learn about connecting to Twitch using IRC, check out my other blog post, linked [here](/blogs/twitch-chat-uses-irc).


## Where did I start?
---


Once I had connected to Twitch chat using my [HexChat](https://hexchat.github.io/) client, I realized that I could see when people join a twitch chat, as they join. 


![](/images/blogs/TwitchIRC.png)


Out of habit, I started greeting people as they come in to the twitch chat. This lead me to research the [HexChat API](https://hexchat.readthedocs.io/en/latest/developers.html#scripting) and then to figure out the best and easiest way to automate some of this behavior. I decided that a lua script would be the easiest way to start.


Further research on the scripting API lead to me to understand that HexChat has a "hook" system. So, you simply write a callback function and assign it to the corresponding hook.


### Hooks


I used the hexchat.hook\_print function to listen for "Join" events. I had a hard time learning how to use the hook\_print function. I didn't understand what options were available for "event". After some further reading, I realized that the events being referenced here were the events listed under HexChat's "*Settings*" > "*Text Events*" menu.



```

  local function on\_join(hook\_info)
    -- Code Here
  end

  hexchat.hook\_print("Join", on\_join)

```

### Remembering Users


I noticed that often times, users would join and leave frequently. I don't know whether this was genuinely a disconnect/reconnect issue or just the Twitch IRC servers bugging out. I figured I shouldn't auto greet users that I've greeted previously, so I decided to store a file with the channel name and the chatter's username.


This would allow me to effectively remember whether this combination of channel-user has been greeted before. Unfortunately, Lua has limited file system support so I kept it rather dumb and moved on to building this in Ruby.


**LUA CODE:**



```

hexchat.register("auto-greeter", "0.1", "Welcomes new members to chat channels ")

local greetings = {
   "Enjoy your stay!",
   "Thanks for joining!",
   "We're happy to have you!",
   "One of us... One of us... gooble gobble",
   ""
}

local function file\_exists(name)
   local f=io.open(name,"r")
   if f~=nil then io.close(f) return true else return false end
end

local function touch\_file(path)
   new\_file = io.open(path, "w")
   new\_file:close()
end

local function on\_join(info)
   local nick = info[1]
   local chan = info[2]

   local file\_dir = "C:\\Users\\Raul\\AppData\\Roaming\\HexChat\\" .. chan .. "\_" .. nick .. ".mem"
   local isMet = file\_exists(file\_dir)

   if isMet == false then
      touch\_file(file\_dir)

      -- Random Greeting
      local rand\_num = math.random(#greetings)
      local rand\_greeting = greetings[rand\_num]

      -- Reply to Message
      hexchat.command("MSG " .. chan .. " " .. "Welcome, " .. nick .. "! " .. rand\_greeting)
   end

   -- Allow this Event to trigger other responses
   return hexchat.EAT\_NONE
end

hexchat.hook\_print("Join", on\_join)

-- Debug Message. Lets you know this function was hooked.
hexchat.print("Auto-Greeter Has Been Hooked!!")

```

*(For flair, I randomly pick a greeting message)*


### Loading Into HexChat


Once the code was written, we simply had to load it into HexChat. The easiest way of doing that is to type: 



```

/load "file\_path.lua"
```

into HexChat. This will load the lua script to HexChat. The downside to this is that you will need to do this every time you open hexchat. If you want to load it automatically, you can add it to your HexChat **addons** folder. I believe that is found in your %AppData% folder on Windows. 


## Going headless!
---


There were a few issues with the HexChat plugin. Some could be solved by using the Python API rather than the Lua API. Could also install a lua file system library. I, however, am more comfortable with Ruby, and I realize that IRC is a very old protocol, so most languages should support scripting for IRC servers.


At this point, I decided to port the code above to a Ruby script. This way, I:


1. Do not rely on HexChat
2. Have full control over the IRC input/output
3. Can container-ize my script and run it headlessly
	1. Specifically, using one of my Raspberry Pis


This means that I'll be able to automate my account even if my laptop or desktop is offline.


### The Code


[Github Link Here](https://github.com/RGonzalezTechTips/Simple-Ruby-IRC-Bot)


I made a few changes to the HexChat script.


* No Longer Triggers on "Join"
	+ I received some criticism for disturbing the twitch lurkers.
	+ If you are unfamiliar, a twitch lurker is a viewer that joins to watch your stream, but they do not interact with the chatroom.
	+ It now triggers when it receives a message event, instead.
* It recognizes when you've returned
	+ We use a different array of phrases when a user returns


Just like before, it "remembers" users by creating a directory/file in the "**./met**" directory. We're able to use the file system to determine when the file was originally created, and therefore how long it's been since we last greeted them. 


My next blog post will detail how I used C# and some spreadsheets to generate the cards printout for [Potato, the card game](https://rgonzalez.tech/projects/potato-the-card-game)!


I hope you found something in this blog post helpful :)  

Thanks for reading!





A few days ago, I was trying to watch one of my favorite twitch streamers but the stupid platform kept going down!


I noticed that one user was unaffected by the downtime. He was capable of chatting in the livestream even though all of our chatrooms had gone offline. He mentioned a small tidbit about how he was chatting and watching from his Playstation.


🤔 This got me thinking about the chat servers on the [Twitch.TV](https://twitch.tv/) website. I did some more research on their chat system and it appears that they are using a variation on IRC and OAuth. Some quick reading on [IRC and Chatbots](https://dev.twitch.tv/docs/irc/guide#connecting-to-twitch-irc) lead me to believe I could fix my chatting issues by connecting directly to the twitch IRC server.


![](/images/blogs/rubbing_hands.gif)


## Getting My Credentials
---


Reading the Chatbot/IRC docs provides the server address, port and all other connection information you would need.


The instructions include a step that tells you how to get your OAuth token using a community-made webapp.


For some reason, I did not feel comfortable using this public app to print out my OAuth token. I figured I could probably get my token by reading the auth headers in the network traffic. I was right.


You can use your browser's DevTools to watch your network traffic and read the messages being sent accross the irc websocket in your browser. I won't tell you exactly how, but this might help?







## Connecting Via IRC
---


I decided to go with [HexChat](https://hexchat.github.io/), but any IRC client will do. Simply enter the information we just got and you're on!


![](/images/blogs/HexChat.png)


**irc.chat.twitch.tv** is the IRC server address, **6697** is the Port. This is the data that we retrieved from the IRC document (linked above).  

Notice that ***SSL is enabled*** because we're using port 6697, the SSL port mentioned in the doc.


At this point, you're good to go. "Twitch" is the server that you connect to. Going forward, you can join channels.   

For example, if your friend's livestream account is "**SuperFriend420**", simply lowercase the username and add a "#" sign ahead of it.


To chat in the **SuperFriend420** livestream, join the channel **#superfriend420**.


 


## A Whole New World
---


This opens a lot of doors for automation. IRC has existed for a long time and so it is supported by many clients.


**HexChat** has an API for Lua, Python, etc. which allows you to react to chats in real-time. You can practically turn yourself into a bot so your automations can exist in any channel you join.


You won't need to 🤞 that your favorite streamer has your favorite bot in their chatroom. *YOU* are the bot.


*Additionally*, this allows you to see when users join and leave your chatroom and plenty of other IRC benefits.


![](/images/blogs/TwitchIRC.png)


Pretty interesting, I think.


I'm going to mess around with automating some of my behavior.  

Thanks for reading! :)




Raul

About Me

Latest Blog Posts

Leveraging Next.js and Serverless: New Website

Time for a website redesign 🤔?

Building an Interaction System in Godot: An Update

Raul G.