Star Atlas Galactic Marketplace Price Tracker

Description

Star Atlas is a massive multiplayer online grand strategy game of space exploration, territorial conquest, and political domination that takes place in a virtual gaming metaverse.

Star Atlas Galactic Marketplace Asset Price Tracker is a bot that notify users the lowest selling price of an asset in the Star Atlas Galactic Marketplace through Telegram.

Installation

1. Clone the repository using

   git clone https://github.com/gordonjun2/StarAtlas-GM-Price-Bot.git
   

2. Get a Solana RPC URL from your preferred node.
   • Eg. https://syndica.io/ (free option)
   • Copy the RPC URL into the private.json file, under the key "rpcURL".
     • For Syndica, the RPC URL will be in this format: "https://solana-api.syndica.io/access-token/<your access token>/rpc", where <your access token> is to be edited.
   • You can also have multiple RPC URLs and place them in an array. This will be useful if you have limited RPC usage per RPC URL. In addition, any RPC URL that has reached the maximum request limit will be automatically removed from the array as the bot runs.
     • The format for multiple RPC URL: ["https://solana-api.syndica.io/access-token/<your access token 1>/rpc", "https://solana-api.syndica.io/access-token/<your access token 2>/rpc", ...]
3. Get a CoinMarketCap API by signing up at https://coinmarketcap.com/api/.
   • Copy the CoinMarketCap API into the private.json file, under the key "cmcAPIKey".
4. Create a Telegram Bot using BotFather and save the HTTP API key (refer to the guide here).
   • You can choose any name and username for the bot.
   • Copy the HTTP API key into the private.json file, under the key "teleAPIKey".
5. Change directory into ./StarAtlas-GM-Price-Bot and run

   npm i
   

6. Done!

Usage

1. Run the command below to start the bot:

   node main.js -f private.json
   

2. Once the bot is ready, look for the Telegram Bot you created in Telegram by searching for its Telegram username (eg. @<bot's username>).

3. Click on /start to start the tracking.
   

4. Done! You can leave the bot running in your terminal.

   • To stop the bot, just do a Ctrl+C.
   • To keep the bot running without the need to have your computer switched on all the time, consider deploying the repository into a Cloud service (see Deploy to Cloud section below).

Usage Tips

• In the Telegram chat with the bot, type /help to get all available commands to use.
  • Currently implemented:
    • /start: Re-initialise tracking and list assets being tracked (do not need to use again)
    • /list: List assets being tracked
    • /modify: Modify assets to be tracked
      • Some assets are already added. You can remove them and add new assets here.
    • /help: To list the available commands and their description
• You can set the data retrieving interval in main.js under

  function begin_bot(){
  teleBot.telegram.sendMessage(chatID, "Tracking starts now...", { parse_mode: 'HTML' });
  console.log(getDateTime() + ' Timer starts');
  scanManager = setInterval(scan, 30 * 1000);                             // check every 30 sec
  checkAliveManager = setInterval(checkAlive, 6 * 60 * 60 * 1000);        // check every 6 hours
  scan();
  }
  

  • Change the interval value to your own preference.
  • Please see Note section below.

Deploy to Cloud

• In this repository, only Google Cloud Platform (GCP) is being used here. There are a few ways to go about deploying the program to GCP. I experimented with two ways. When testing Method 1 with GitHub Actions, I intentionally left the private keys in private.json blank. This is to prevent me from commiting my keys into this public repository. For Method 2, my keys can remain in private.json since I do not need to commit them to GitHub (there is an option to use GitHub Actions too).

• Before following the steps below, please fork this repository first.

• Method 1 using GitHub Actions and Workload Identity Federation (follow the guide here)

  1. Change directory into ./StarAtlas-GM-Price-Bot and run

     docker build . --no-cache -t staratlas-gm-price-bot
     

  2. Continue with the steps using the link above, starting from the Create a Google Cloud Project header.
     • Use Cloud Shell in the browser to run the commands.
  3. Follow Create a Google Cloud Project.
  4. Follow Create a Workload Identity on Google Cloud.
  5. Follow Create a GitHub Actions Workflow.
  6. When deploying Cloud Run for the first time, an authentication error will be shown in the GitHub Actions' Workflow (To see, click on the Actions tab in your forked repository. Then, click into the latest workflow run and then deploy under Jobs.). To fix this error, the Cloud Run service should set to allow unauthenticated access. Run the below command in the Cloud Shell:

     gcloud run services add-iam-policy-binding app \
                 --member="allUsers" \
                 --role="roles/run.invoker"
     

  7. Run the below command in the Cloud Shell to start streaming logs before the deployment is being done again (refer here):

         gcloud alpha run services logs tail app --project PROJECT-ID
     

     • Edit PROJECT-ID based on your own project ID (see the steps in the link above).
  8. Return back to the page where the GitHub Actions' workflow run is shown (refer to Step 3.) and click on Re-run all the jobs.
  9. The deployment should be successful (see image below), and you can view the logs in the Cloud Shell (refer to Step 4.).
  10. Since the private keys are not in *private.json*, the program will exit (see the image below). However, this shows that the program is being deployed correctly.
  

• Method 2 using Google Cloud Artifact Registry (follow the guide here)

  1. Change directory into ./StarAtlas-GM-Price-Bot and run

     docker build . --no-cache -t staratlas-gm-price-bot
     

  2. Continue with the steps using the link above, starting from the Set up Artifact Registry header.
     • Use Cloud Shell in the browser to run the commands.
     • Since the docker is built in Step 1., skip if required to.
  3. Follow Set up Artifact Registry.
     • You can give a different Artifact Registry name.
  4. Follow Build, Tag, and Push to Artifact Registry.
     • Since the docker is built in Step 1., do not use the build command.
     • You can give a different Artifact Registry image name
     • Tag the local docker image name (staratlas-gm-price-bot) with the Artifact Registry address, repository, and name.
  5. Follow Create a Google Cloud Run Service.
     • Make sure to use the correct port number when setting up the Google Cloud Run Service.
     • In this repository, port 8080 is used (see main.js and Dockerfile).
  6. Run the below command in the Cloud Shell to start streaming logs before the deployment is being done again (refer here):

         gcloud alpha run services logs tail app --project PROJECT-ID
     

     • Edit PROJECT-ID based on your own project ID (see the steps in the link above).
  7. Follow Deploy with gcloud.
     • Since the private keys are in private.json, the program will run (see the image below). It shows that the program is being deployed correctly.
     
  8. You can follow Automate Deployment with GitHub Actions, but remember not to commit your keys to your public repository.

Note

• The program uses the @staratlas/factory package.
  • Package documentations
• The bot does not make use of the class GmEventService in the package @staratlas/factory. Thus, it does not react immediately to new events happening in the marketplace (eg. new sell orders). Rather, it uses the GmClientService based on a fixed interval set in main.js.
  • My intention was to use GmEventService as this can probably reduce the frequency of RPC calls, but I am unsure on how to use it. Please hit me up if you know how to do it.
• I am very new to JavaScript. Please let me know if there are any errors or ways the code may be improved.
• I am also new to DevOps / Cloud Engineering. Please let me know if there are any better way to go about the deployment of the program.