DEV Community: Kyle Schwartz

Migrate a JS Project to Yarn Berry

Kyle Schwartz — Tue, 28 Dec 2021 14:56:46 +0000

For this tutorial, you will need a starting project. The assumption is that there is no lock file and the project is not installed (i.e. no node_modules folder). I will be starting with a blank Vite project. You can create one by running the command below and following the prompts:

yarn create vite

If you don't have Yarn berry installed, run one of the following commands:

# Node >=16.10 - May have to run as admin
corepack enable

# Node <16.10
npm i -g corepack

Once completed, open the project. Now, run the following commands in sequence:

yarn set version berry
yarn
yarn dlx @yarnpkg/sdks vscode # If using vscode/vscodium

All done!

Shuffle JavaScript Array in 1 Line

Kyle Schwartz — Mon, 06 Dec 2021 21:22:01 +0000

const shuffle = () =>
    arr
        .map((e) => [e, Math.random()])
        .sort((a, b) => a[1] - b[1])
        .map((e) => e[0]);

Although slightly longer than other implementations, it maintains an even distribution.

let results = {};

for (let i = 0; i < 100000; i++) {
    const a = shuffle();
    results[a] = results[a] ? results[a] + 1 : 1;
}

console.log(results);

// { '3,2,1': 16647,
//   '1,3,2': 16877,
//   '2,1,3': 16781,
//   '1,2,3': 16510,
//   '3,1,2': 16856,
//   '2,3,1': 16329 }

Installing Jellyfin on LG webOS

Kyle Schwartz — Sat, 06 Nov 2021 13:49:10 +0000

UPDATE 2022-11-29: Jellyfin has added their app to LG's WebOS store for newer TVs (2021+). For older ones (WebOS < 6), the following instructions should still apply.

Clone https://github.com/jellyfin/jellyfin-webos
Open WebOS TV SDK
cd into directory
Run ares-package org.jellyfin.webos
Open the Developer Mode app on the TV and login
- Get the Developer Mode app
- Get a developer account
Turn on Dev Mode, TV will reboot
Re-open the Developer Mode app
Turn on Key Server
Run ares-novacom --device tv --getkey, passphrase is found on screen
Run ares-install --device tv --list, a blank line means everything works
Run ares-install --device tv ./org.jellyfin.webos_X.X.X_all.ipk replacing the Xs with the requisite version number
Done!

If you need to re-add tv to the device list

Run ares-setup-device and remove tv
Run ares-setup-device and choose add
Use the following settings:
- Device Name: tv
- Device IP address: IP found on screen
- Device Port: 9922
- SSH user: prisoner
- Description: optional
- Select authentication: password
- Password: leave blank

Using Packetriot to Access OctoPrint Remotely

Kyle Schwartz — Wed, 12 May 2021 15:25:50 +0000

Introduction

Welcome! This is a pretty straight-forward tutorial on setting up Packetriot. Although intended for OctoPrint, this is pretty software-agnostic, and thus, can be followed for pretty much any local service. I will be doing this on an Orange Pi PC+ running Armbian (Debian) Buster. This should work for any Debian-like OS such as PopOS and Ubuntu.

For this tutorial, you'll need:

A Domain
Basic Linux knowledge
A Packetriot Account (Get one here)

If you're not running a Debian-like OS, additional download instructions are available here.

If at any point you get stuck, check out Packetriot's incredibly extensive documentation here.

Now, let's get into it!

Install Packetriot

Install the GPG key and make sure apt can work with GPG signed packages and HTTPS sources.

sudo apt-get install apt-transport-https gnupg -y

wget -qO - https://download.packetriot.com/linux/debian/pubkey.gpg | sudo apt-key add -

Setup the Packetriot source in the apt sources directory.

echo "
deb [arch=amd64] https://download.packetriot.com/linux/debian/buster/stable/non-free/binary-amd64 / 
deb [arch=i386]  https://download.packetriot.com/linux/debian/buster/stable/non-free/binary-i386  / 
deb [arch=armhf] https://download.packetriot.com/linux/debian/buster/stable/non-free/binary-armhf / 
deb [arch=arm64] https://download.packetriot.com/linux/debian/buster/stable/non-free/binary-arm64 / 
" | sudo tee /etc/apt/sources.list.d/packetriot.list

Update the apt sources and install the client.

sudo apt-get update

sudo apt-get install pktriot

Add a Domain

In order to have a static domain, you'll either need to have a paid plan or provide your own domain. In this tutorial, we'll opt for the latter.

On the Packetriot site:

Go to Domains -> Custom
Click Verify Domain.
Enter your domain name.
Click Submit.

Now, Packetriot will show you a long string of text that you need to copy. In your DNS config in your domain registrar, you will have to create a new TXT record. Paste in the TXT Record from Packetriot and wait a few minutes. Once the domain is verified, continue to the next step.

Enable Packetriot

Now that our domain is setup, we will actually create the tunnel. First, we enable the Packetriot service so it'll start on the server's startup. Then we will login and create the tunnel. When you run configure, make sure you choose option 1.

sudo systemctl enable pktriot
sudo -u pktriot pktriot configure

After the configuration has been run, you'll be provided with an IPv4. Take this and create a new A record in your domain registrar's DNS, same as before with the TXT record. This must be the same domain as registered earlier.

Next, we will run the following command to tell the tunnel where to connect. The first part is not a typo; it is telling the server to run the command as the pktriot user, and then running the pktriot command. Change example.com and 80 to the relevant fields. I.e. your domain and port, respectively. For OctoPrint, this will be 5000 if no reverse proxy has been setup. Additionally, change 127.0.0.1 to the appropriate IP if not being run on the same server as the server you are tunneling.

sudo -u pktriot pktriot tunnel http add --domain example.com --destination 127.0.0.1 --http 80 --letsencrypt

And finally, start the Packetriot service. This may take a few minutes to get the SSL certificates. Once it is done, you can go to your URL and access your server from anywhere!

sudo systemctl start pktriot

Thank you so much for following allong with this tutorial! If you liked it, had any problems, or found any errors, let me know in the comments down below!

Have a great day!

Setting Variables Using Switch Statements In JavaScript

Kyle Schwartz — Tue, 19 Jan 2021 17:01:32 +0000

Here is a snippet showing an example of how this would be done. Each animal has a favorite color which we will determine using a switch statement.

const animal = "dog";

const color = (()=>{
    switch (animal) {
        case "cat":
            return "yellow"
        case "dog":
            return "blue"
        case "fish":
            return "green";
    }
})();

console.log(color); // "blue"

And a similar example with parameter passing:

const color = (animal) => (()=>{
    switch (animal) {
        case "cat":
            return "yellow"
        case "dog":
            return "blue"
        case "fish":
            return "green";
    }
})();

console.log(color("fish")); // "green"

And just to show how much cleaner the above version is, here is an example using an if/else statement:

const color = (animal) => {
  if (animal === "cat")
      return "yellow";
  else if (animal === "dog")
      return  "blue";
  else if (animal === "fish")
      return "green";
};

console.log(color("fish")); // "green"

While, yes, the if/else version is technically shorter, you repeat yourself, it is less readable (subjectively), and it is inherently slower (at scale). With switch statements, the expression is only evaluated once, but with our if/else it is evaluated 3 times. Although, this example is short, so it may not affect performance. GeeksforGeeks has a great article comparing the two: https://www.geeksforgeeks.org/switch-vs-else/

Happy coding!

Install Docker (Compose) on Oracle Linux 8

Kyle Schwartz — Tue, 22 Dec 2020 01:05:12 +0000

Welcome! I couldn't find a simple, straightforward tutorial on how to do this. Everything written here is a summation of the Docker documentation on CentOS, as that is what OL is built upon. I will be using Oracle Linux 8 build 2020.12.17-0, for reference. Now, on to the tutorial!

Installing Docker Engine

First, install yum-utils

sudo yum install -y yum-utils

Next, we add the Docker repository

sudo yum-config-manager \
    --add-repo \
    https://download.docker.com/linux/centos/docker-ce.repo

Now, we can install Docker. Answer y for all prompts.

sudo yum install docker-ce docker-ce-cli containerd.io

Lastly, start Docker

sudo systemctl start docker

Optionally, run the hello-world image to make sure everything works

sudo docker run hello-world

Installing Docker Compose

First, download Docker Compose

sudo curl -L "https://github.com/docker/compose/releases/download/1.27.4/docker-compose-$(uname -s)-$(uname -m)" -o /usr/local/bin/docker-compose

Add execution perms to the download

sudo chmod +x /usr/local/bin/docker-compose

Lastly, we need to link to the docker-compose command

sudo ln -s /usr/local/bin/docker-compose /usr/bin/docker-compose

Now, everything should work. Run docker-compose --version to confirm.

And that's it! Easy, right?

Thank you for joining me on this adventure and I hope this was some help!

My Unnecessary Adventure in (Re)Building a Media Sever

Kyle Schwartz — Fri, 30 Oct 2020 22:13:50 +0000

Welcome! Strap in because this is gonna be a long ride.

To preface, I had an existing media server that I broke some permissions and reinstalling was the only option. I had an existing RAID 0 setup with lots of data I did not want to lose, plus a standalone SSD. So, without further adieu, here we go!

Installing Ubuntu Server

I chose Ubuntu Server because Ubuntu has a great deal of community support and I have no need for a desktop GUI.

Download the ISO from here. You may have to choose 'Option 3' to actually get a download. At the time of writing, I'm using 20.04.1, so take that into account.

My tool of choice for burning ISOs is balenaEtcher.

Plug in the USB, boot to it, and follow setup. Make sure you enable the SSH server AND NO LVM ON THE SSD. Check to make sure the full capacity of the SSD is mounted to / and then you're good to go. Finish the setup, remove the USB, and reboot. Now we'll switch to SSH.

Configuring SSH

Port 22 sucks, but we need to login with it. Connect to the server on port 22 with the credentials you added during install.

Open the port in the firewall

sudo ufw allow 8364/tcp

Open the SSH config file and change the port used

sudo nano /etc/ssh/sshd_config

Find the line containing Port 22, uncomment, and change to Port 8364. Save (ctrl x).

Restart the SSH server and then reconnect on your client.

sudo systemctl restart ssh

Drive Mounting

Find the UUID of the drive:

sudo blkid

Add your drive to /etc/fstab (BE CAREFUL, THIS CAN BRICK STUFF):

sudo nano /etc/fstab

E.g:

UUID=ccc56606-385e-4a16-89c4-7f1513ab1641 /home ext4 defaults 0 0

Reboot the system and your drive your be properly mounted.

Docker

First, update your system:

sudo apt update
sudo apt upgrade

Instead of repeating things, go follow the official Docker tutorial here. Just follow the 'Install using the repository' and 'Install Docker Engine' sections.

Then, install Docker Compose by following the tutorial here.

The next part is taken mostly from Smart Home Beginner.

We will add the current user to the docker group to simplify some things in the future.

sudo gpasswd -a $USER docker

Activate the changes using the following:

newgrp docker

Portainer

Portainer is our docker management solution. Yes, I have serious issues with it but currently don't have an alternative. Maybe this will change in the future, but here we are.

Run the following commands to get Portainer up and running:

docker volume create portainer_data
docker run -d -p 8000:8000 -p 9000:9000 --name=portainer --restart=always -v /var/run/docker.sock:/var/run/docker.sock -v portainer_data:/data portainer/portainer-ce
docker run -d -p 9001:9001 --name portainer_agent --restart=always -v /var/run/docker.sock:/var/run/docker.sock -v /var/lib/docker/volumes:/var/lib/docker/volumes portainer/agent

Open a browser and go to your computer's IP with the port 9000. For me, that's http://192.168.1.50:9000. Chose a username and password, then create user.

Choose "Docker" and press "Connect".

Now that you are in the web panel, click on 'local' in the middle, and then the 'Stacks' tab on the left. Click 'Add stack' and give it a name such as "main_stack". Paste in the following config and adjust it to fit your needs.

---
version: "2.3"
services:

  organizr:
    image: linuxserver/organizr
    container_name: organizr
    environment:
      - PUID=${PUID}
      - PGID=${PGID}
      - TZ=${TZ}
    volumes:
      - ${USERDIR}/organizr:/config
    ports:
      - 9983:80
    restart: unless-stopped

  watchtower:
    container_name: watchtower
    restart: always
    image: containrrr/watchtower
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock
    command: --schedule "0 0 4 * * *" --cleanup

  radarr:
    image: linuxserver/radarr
    container_name: radarr
    environment:
      - PUID=${PUID}
      - PGID=${PGID}
      - TZ=${TZ}
    volumes:
      - ${USERDIR}/radarr:/config
      - ${USERDIR}/media/movies:/movies
      - ${USERDIR}/downloads:/downloads
    ports:
      - 7878:7878
    restart: unless-stopped

  sonarr:
    image: linuxserver/sonarr
    container_name: sonarr
    environment:
      - PUID=${PUID}
      - PGID=${PGID}
      - TZ=${TZ}
    volumes:
      - ${USERDIR}/sonarr:/config
      - ${USERDIR}/media/tv:/tv
      - ${USERDIR}/downloads:/downloads
    ports:
      - 8989:8989
    restart: unless-stopped

  duckdns:
    image: linuxserver/duckdns
    container_name: duckdns
    environment:
      - TZ=${TZ}
      - SUBDOMAINS=kyleserver
      - TOKEN=326ab158-e842-4d0e-aa20-233e0f7e51e3
    restart: unless-stopped

  minecraft-creative:
    container_name: minecraft-creative
    image: itzg/minecraft-server
    environment:
      EULA: "true"
      TYPE: PAPER
      VERSION: 1.16.2
      CONSOLE: "false"
      LEVEL_TYPE: "flat"
    ports:
      - 25566:25565
    volumes:
      - ${USERDIR}/minecraft-creative:/data
    restart: unless-stopped

  minecraft-survival:
    container_name: minecraft-survival
    image: itzg/minecraft-server
    environment:
      EULA: "true"
      TYPE: PAPER
      VERSION: 1.16.2
      CONSOLE: "false"
      MEMORY: 2G
    ports:
      - 25565:25565
    volumes:
      - ${USERDIR}/minecraft-survival:/data
    restart: unless-stopped

  qbittorrent:
    image: linuxserver/qbittorrent:14.2.5.99202004250119-7015-2c65b79ubuntu18.04.1-ls93
    container_name: qbittorrent
    network_mode: "service:gluetun"
    volumes:
      - ${USERDIR}/docker/qbittorrent:/config
      - ${USERDIR}/downloads:/downloads
    restart: unless-stopped
    environment:
      - PUID=${PUID}
      - PGID=${PGID}
      - TZ=${TZ}
      - UMASK_SET=002
      - WEBUI_PORT=4545

  gluetun:
    image: qmcgaw/private-internet-access
    container_name: gluetun
    cap_add:
      - NET_ADMIN
    network_mode: bridge
    ports:
      - 8888:8888
      - 8388:8388
      - 8388:8388/udp # Shadowsocks
      - 8001:8001
      - 4545:4545
      - 6881:6881
      - 6881:6881/udp
    volumes:
      - ${USERDIR}/gluetun:/gluetun
    environment:
      - VPNSP=surfshark
      - TZ=${TZ}
      - USER=USERNAME
      - PASSWORD=PASSWORD
      - REGION=Canada Toronto
    restart: unless-stopped

Next, run id and note your uid and pid. We will input these as PUID and PGID in the "Environment" section.

PUID=1000
PGID=1000
TZ="America/Toronto"
USERDIR="/home/user"

Add any environment variables that you need NOW as I don't believe these can be changed in the future.

Click 'Deploy the stack'.

Currently, qbittorrent has an issue with mismatched torrent names, that's why it's using a custom image version. If it gets resolved in the future, remove everything after the colon.

These are simply the containers I use for everything I do. The 'gluetun' container connects to Surfshark VPN and routes the qbittorrent container's traffic through, allowing a normal container to connect to a VPN. This can be done with any container by moving its ports to gluetun and setting its network_mode to "service:gluetun". After that, go and check each container to make sure it's working.

Jellyfin

Welp, this is gonna be a fun one. If you don't want to do hardware transcoding with a GPU, simply add a standard container to the above compose file. Otherwise, come along for this hell of a ride. There is no guarantee that some of these steps aren't redundant as I have no clue how any of this works. But, ideally, everything will work by the end of this.

Firstly, we need to install the NVIDIA drivers. I'm using a GTX 1050ti but it should work for any NVIDIA card 10-series or newer.

Run these commands first to disable Nouveau:

sudo bash -c "echo blacklist nouveau > /etc/modprobe.d/blacklist-nvidia-nouveau.conf"
sudo bash -c "echo blacklist nouveau > /etc/modprobe.d/blacklist-nvidia-nouveau.conf"

What is Nouveau and why are we disabling it? How dare you ask! Subsequently, I have no clue but it's what we gotta do.

Reboot.

Install DMKS:

sudo apt install dkms

Follow the instructions here to install the drivers.
Additional Notes:

Update the version number in the other commands
Each command needs the be run with sudo

Click 'Yes' for everything, and after exiting the GUI, reboot, then run the test commands.

Install ffmpeg:

sudo apt install ffmpeg

Next, we need to setup the NVIDIA Container Toolkit. Follow the official instructions here.

Now the GPU should be good to go! Next, we need to actually setup Jellyfin. Going back to SSH, go to your home dir (cd).

Create a new docker compose file called docker-compose.yml and fill it with the following:

version: "2.3"
services:
  jellyfin:
    image: linuxserver/jellyfin
    container_name: jellyfin
    network_mode: "host"
    runtime: nvidia
    environment:
      - PUID=1000
      - PGID=1000
      - TZ=America/Toronto
      - NVIDIA_VISIBLE_DEVICES=all
    volumes:
      - /home/user/jellyfin:/config
      - /home/user/media/tv:/data/tvshows
      - /home/user/media/movies:/data/movies
    ports:
      - 8080:8096
      - 7359:7359/udp
    restart: unless-stopped

Run docker-compose up -d and Jellyfin will be running. Counterintuitvly, go to port 8096 on your server's IP and login.

Change Jellyfin Settings

Go to Settings > Advanced > Networking. Change "Local HTTP port number" and "Public HTTP port number" to 8080.
Go to Server > Playback and under "Hardware acceleration" choose 'Nvidia NVENC'. Select which files you want to decode.
Change 'ffmpeg path' to /usr/bin

Restart Jellyfin from Portainer, and you're good to go!

Conclusion

It was a lot of work, but hopefully everything works. Remember, DuckDuckGo and StackOverflow are your friends. If something doesn't work, let me know below and I'll do my best to help.

Good luck and stay curious!

Sources

Gitea + Drone CI + Vercel = Gitea Pages

Kyle Schwartz — Sun, 17 May 2020 22:57:05 +0000

Gitea + Drone CI + Vercel = Gitea Pages

Welcome! So this article comes after a year of being annoyed, but lazy, about the lack of functionality within Gitea to produce something similar to GitHub Pages. About a week ago, I finally got fed up and decided to figure it out and put this knowledge online since no one else has written about this and that’s frustrating. After following the steps below, you’ll be able to deploy your static-generated websites from any framework of your choosing (Vue, React, Svelte, Hugo, …) by simply pushing your code to Gitea.

Setup

This tutorial assumes a couple of things:

You have basic computer literacy
You have both a working Gitea server as well as a Drone CI server.
You have a repository on your Gitea server that has all of your code ready to be deployed

If you don’t have a Gitea or Drone CI server, check out my other post on how to host these for free. Or click the links below because no one likes extra work.
Hosting Gitea on Google Cloud Platform
How to Install Drone CI for Gitea

Vercel Setup

Make an account at https://vercel.com
Go to https://vercel.com/account/tokens
Click on the Create button, then enter “Drone” as the token name
Click Create Token and save the code that is given to you. We will need this for later and Vercel will not show it again.
Click Done and continue on.

Drone CI Setup

Open Drone and open the repository you want to use
Click Activate Repository
Click Settings
Go to the Secrets section and add a new secret. Name it “NOW_TOKEN”, and give it the value of the token we got from the Vercel setup.
Click ADD A SECRET to save it

Gitea Setup

In the root of your project, create a file called “.drone.yml”. The contents of which should be as follows:

Deployment Settings

secret: required. Don’t touch this either
deploy_name: Change to what you want your Vercel project name to be
prod: Leave this if you want a production build. If you want development builds, remove this line
directory: Only use this if you pre-compiled your code. If not, remove this line

Push the changes to your repository.

Fixing Vercel

Now that you’ve pushed your code, a Drone runner should have pushed your code to Vercel! Fantastic!

Next comes fixing the issues. Vercel may or may not detect your framework correctly. If you go to your dashboard and your project is there, check the link to see if everything worked correctly. If it did, fantastic! You’ve done everything correctly and there is nothing left to setup. Either check out the Extras section below for some cool stuff or continue writing your code! Enjoy!

If Vercel did not detect your framework, follow these steps:

Open your project in Vercel by clicking its name
Click Settings in the top left
Under Build & Development Settings, chose your framework preset. If it’s not listed, chose Other
If the build, output, or development commands listed aren’t correct, click the Override toggle and manually edit the command
Click Save and push your code to Gitea once more. This should fix the issues

Extras

If you want to only update production when you push to master, check out the gist below:

Every branch that isn’t master will push to dev builds and master will now push to production!

How to Install Drone CI for Gitea

Kyle Schwartz — Sun, 17 May 2020 22:51:18 +0000

How to Install Drone CI for Gitea

Welcome! This is a simple tutorial that is heavily based on Drone’s documentation with a minor focus on Oracle Cloud. Without further ado, let’s get into it!

Prerequisites

Basic computer literacy
An Oracle Cloud account (Any provider such as AWS or GCP with work just as well), since that’s what we’ll be using here
An existing Gitea server
A domain name and access to your DNS records

If you chose to use GCP, you can follow my Gitea tutorial for the networking bits found here.

If you don’t have a Gitea sever, you can follow the aforementioned tutorial to get one setup.

It is important to run this on a DIFFERENT server than your Gitea instance. Thankfully, Oracle offers 2 VMs for free so that shouldn’t be a problem!

Generate SSH Keys

Login to your Oracle Cloud
Open the Cloud Shell by pressing the console icon in the top right corner
Wait a few moments and then run the following command in the console at the bottom of the page:
ssh-keygen -t rsa -b 4096 -C "*your_email@example.com*"
replacing your_email@example.com with your real email. Leave the quotation marks!
The console will ask you where you want to save the key. Just press enter.
When prompted to enter a password, simply leave it blank and just press enter.
Now that your keys are generated, we need to view our public key by entering the following command into the console:
cat .ssh/id_rsa.pub
This will display our public key in the console. It will begin with ssh-rsa and end with you email. Select the entire block of text, around 3 lines, including ssh_rsa and your email. Copy this, as we will need it to setup our VM and Console Connection.

VM Setup

Click on the hamburger menu (☰) > Compute > Instances
Click Create Instance
Name your instance as you choose, I prefer “Drone”
Click Change Image and select Canonical Ubuntu 18.04. Scroll to the bottom and click Select Image
Click Paste SSH Keys and paste the key you copied from the previous section.
Click Create and wait for your instance to be fully setup (The icon will turn green)
Now, write down the public IP address under Instance Access. We will need this down the line.

Networking

Console Connections

Scroll to the bottom of your instance page, and click Console Connections
Click Create Console Connection
Click Paste SSH Key and paste in the same key from the last section.
Click Create Console Connection and wait a few seconds

Virtual Cloud Network

At the top of the instance page, below the Instance Details section, click on the blue link similar to VirtualCloudNetwork-00000000–0000
On the left side, below the green hexagon, click on Security Lists (1)
Click on Default Security List for VirtualCloudNetwork-00000000–0000
Click Add Ingress Rule and add the two rules below one at a time:
— — — —
Rule 1:
Source CIDR: 0.0.0.0/0
Destination Port Range: 80
— — — —
Rule 2:
Source CIDR: 0.0.0.0/0
Destination Port Range: 443
— — — —
Leave all other fields as-is

Connecting to the VM

Using the Cloud Console once more, we need to edit the permissions of our Private SSH Key by entering the following into the console:
chmod 400 .ssh/id_rsa
Now, using the public IP address from awhile ago, enter the following command, substituting “address” for the VMs IP address:
ssh –i .ssh/id_rsa ubuntu@address

Random Config Note

Before going to the next step, I HIGHLY recommend going into the DNS of your domain and configuring drone.yourwebsite.com to the public IP of your VM. If you’re using cloudflare, DON’T PROXY, use DNS ONLY. Thank you, you may continue now.

Installing Drone CI

First, you need to install docker. Follow this tutorial from DigitalOcean. You only need to follow everything in step 1.

Once you have docker installed, we are going to open a new tab and go to our Gitea website. You might need an admin account for this part.

Click on the profile icon in the top right and click Settings
Click on the “Applications” tab and scroll down to the Manage OAuth2 Applications section
Enter “Drone” as your Application Name, and “http://drone.yourwebsite.com/login” as the Redirect URI, then click Create Application
You will now we shown your Client ID and your Client Secret. Write down or copy the Client Secret as it won’t be shown again. Client ID will always be viewable.
Now, head back to your Cloud Console (You should still be logged into the VM) and we are going to generate a shared secret by running the following command:
openssl rand -hex 16
Save the output of this command as we will need it later
Now we will download drone itself by running this command:
docker pull drone/drone:1
The easiest way to do this next step is to copy the code below into Notepad or TextEdit and edit all of the options there. Below each of the options will be listed. Just paste the value after the equal sign on that line.

DRONE_GITEA_SERVER: The address of your Gitea server.
Eg. https://git.example.com
DRONE_GITEA_CLIENT_ID: The Client ID from step 4
Eg. f8cd8dfe-29f6–48f7–9c87–8b1bcec79dbe
DRONE_GITEA_CLIENT_SECRET: The Client Secret from step 4
Eg. xpV0v74hKmDm5tulfR932TRZTVmkHWqCcZVJXG5CK4A=
P.s. Yes, it will end in an equal sign. No, this is not a mistake
DRONE_RPC_SECRET: The shared secret generated in step 5
Eg. 2d2c5fc55c5991a8cc90f75e313d5670
DRONE_SERVER_HOST: The domain of your drone server
Eg. drone.yourwebsite.com
DRONE_SERVER_PROTO: Leave as http

Once you’ve edited all of the values, copy the command and paste it into your cloud console. Now you have a Drone CI server up and running!

Creating a Runner

Runners are necessary to make pipelines function. Think of the Drone server as a marathon and a runner as a participant. Without participants, you can’t really have a marathon. A runner is the bit of code that actually does the work, the sever just gives the runner room to work.

Setting up a runner is just like setting up the server. First, we’ll pull the image off of DockerHub (This is where most docker images are hosted) and then we’ll run it with a configuration.

Pull the runner by running the following command in the cloud console:
docker pull drone/drone-runner-docker:1
The easiest way to do this next step is to copy the code below into Notepad or TextEdit and edit all of the options there. Below each of the options will be listed. Just paste the value after the equal sign on that line.

DRONE_RPC_HOST: The domain of your drone server
Eg. drone.yourwebsite.com
DRONE_RPC_SECRET: The shared secret from awhile ago. This value should be the same as DRONE_RPC_SECRET from the server config

Now, simply copy the config and paste it into the cloud console and hit enter. Now, you’ve created your first runner!

Conclusion

Now that everything is setup, you can head to your Drone server’s URL and start building pipelines. If you don’t know where to start, Drone’s documentation is where I’d suggest and can be found here. Enjoy and happy coding!

Integrating GitLab CI/CD, Flask, and Heroku

Kyle Schwartz — Sat, 04 Jan 2020 17:40:21 +0000

This tutorial assumes you already have a working flask application within a virtual environment and have your code hosted on GitLab. Also ensure you have Gunicorn installed. You can install it from pip install gunicorn.

Getting Ready

Run pip freeze > requirements.txt to grab all the requirements for running your application.

Create a Procfile file. Inside, place the following code web: gunicorn <name of the app.py file>:<app-name> where app-name is usually “app”.

Push everything to GitLab.

Connecting Everything

Create a new Heroku application. Using a short name will make later steps easier.

Create a new file in your app root called .gitlab-ci.yaml.

Here is a cookie-cutter example of a GitLab CI/CD config:

[app.py file name]:
    script:
        — apt-get update -qy
        — apt-get install -y python-dev python-pip
        — pip install -r requirements.txt

production:
    type: deploy
    script:
        — apt-get update -qy
        — apt-get install -y ruby-dev
        — gem install dpl
        — dpl — provider=heroku — app=[Heroku App Name] — api-key=$HEROKU_SECRET_KEY
    only:
        — master

Now swap out [app.py file name] with your file name and [Heroku App Name] with the name of your Heroku app.

To set the HEROKU_SECRET_KEY,

Get your Heroku API key from https://dashboard.heroku.com/account.
Within GitLab, go to Settings → CI / CD → Secret variables.
Create a new variable and set the Variable key to HEROKU_SECRET_KEY and the Variable Value to your Heroku API Key.
Set the protected switch to ON.

Push the changes to GitLab.

Conclusion

Now that everything is all setup, head over to your Heroku app URL and enjoy the fruits of your labour.

Using Githooks for CI/CD

Kyle Schwartz — Sat, 04 Jan 2020 17:35:40 +0000

A tutorial on how to to use git-scripts for CI/CD

Create the package.json

Create a package.json by running:

npm init

Install git-scripts to the package by running:

npm install --save-dev git-scripts

Add the Git Hooks

After the devDependencies section, add:

  "devDependencies": {
    "git-scripts": "^0.2.1"
  },
  "git": { 
    "scripts": { 
      "pre-push": "<command(s) to run before git push>" 
    } 
  }

Useful Bits of Code From Around The Web

Kyle Schwartz — Sat, 21 Dec 2019 20:48:23 +0000

Reading Files in JS

Text File

fetch('file.txt')
  .then(response => response.text())
  .then(text => console.log(text))
  // outputs the content of the text file

JSON File

fetch('file.json')
  .then(response => response.json())
  .then(jsonResponse => console.log(jsonResponse))     
   // outputs a javascript object from the parsed json