add-challenge

name	add-challenge
description	Scaffold a challenge in this git repository
argument-hint	<blog-name> <programming-language> <stack>

For enums strictly follow documentation.

Add challenge

Scaffold a new challenge in this git repository. Steps are as follows

Step 1: create folder structure

1. create a folder with name format challenge-[id] where id is a short - separated name derived from challenge-name arguement
2. inside the folder created preceing create .nurburgdev directory and a file .nurburgdev/README.md
3. in .nurburgdev/README.md add frontmatter with the following format

---
title: "Your experiment title"
author: "Author Name"
authorLink: "https://github.com/your-handle"
authorTitle: "Software Engineer"
summary: "One-sentence description shown in listings"
publishedOn: 2024-06-01
tags: [redis, scalability]
intent: "challenge"
draft: true
challengeDetails:
  id: 1234
  difficulty: "medium"
  points: 100
  language: "go"
---

Rules to follow

Rules for frontmatter of README.md

1. There should be a front matter to README.md according to the specification document docs/specification-for-experiments.md
2. intent should be challenge
3. challengeDetails are required.
4. For id in challengeDetails in frontmatter - scan challenge-*/.nurburgdev/README.md for the max value in front matter. new id should be 1 more that.

Structure for challenge README.md content/body

1. In content/body of the markdown never use h1 or # heading. Use ## or ### or #### or ##### or #####
2. Body of a challenge should be around 300-500 words.

Writing guidelines for README.md content/body

Follow the guidelines provided in README_GUIDELINE_V1.md folder.

Step 2: Create .nurburgdev/experiment.toml

Read docs/specification-for-experiments.md the section Format of an experiment to understand how to format this file. Consider <programming-language> and <stack> arguements to create this file.

Rules to follow

1. all paths in .nurburgdev/experiment.toml should be from the from folder created in step 1 as base. Never from the git root as base.
2. all paths in .nurburgdev/experiment.toml should not start with ./. for example .nurburgdev/traffic.js is correct. ./.nurburgdev/traffic.js is not correct.

Step 3: Create Procfile

Read docs/specification-for-experiments.md the section Procfile to understand how to create Procfile.

Step 3: Scaffold an http project in as few files as possible

Any project created should have 1 endpoint /healthcheck api

For go project

1. go.mod
2. go.sum
3. src/main.go with a /healthcheck api
4. .gitignore

For typescript or node project

1. packge.json and package-lock.json
2. src/index.ts - use express for http service
3. .gitignore

For python

1. requirements.txt
2. src/index.py - use fastapi for http service
3. .gitignore

For java

1. pom.xml
2. src folder with package dev.nurburg with an springboot 1 file http service. The class should be named main
3. .gitignore

Step 4: Create .nurburgdev/traffic.js

This is a k6 load tests. Create a 10 second loadtest on /healthcheck api.

Rules to follow

1. Assume a HOST env var will be provided. Use this env var for creating urls.
2. HOST env var will be of the format http://host:port format. NOT host:post or http://host:port/. create the urls accordingly.

Step 5: Create .nurburgdev/apitest.tavern.yaml

This is a tavern test file. create 1 test on /heatlhcheck which just expects status code 200

Rules to follow

1. Assume a HOST env var will be provided. Use this env var for creating urls.
2. HOST env var will be of the format http://host:port format. NOT host:post or http://host:port/. create the urls accordingly.

Step 6: Setup devcontainer in the challenge folder created in step 1

Change directory to the challenge directory and run nd devcontainer --features=<comma-separated-features>. Replace <comma-separated-features> with a list of features inferred from arguements. allowed list of features are 1. postgresql 2. mysql 3. kafka 4. typescript 5. go 6. java 7. python 8. temporal

Few conventions

1. in experiment.toml env should be inline
2. in experiment.toml kafka topics should be inline