What Functionality Are You Suggesting?

To start our work on supporting articles we need a new article module that houses an article entity and article service. The article service should include the ability to:

• Get one specific article by ID
• Get all articles
• Create a new article
• Update an article by ID

An article entity should have:

• An auto-incrementing, unique and primary key id field (e.g. 1)
• A created field which contains the date at which the entity was created
• A modified field which's value updates each time the entity is modified
• A user field which's value is equal to the ID of the user who created the article
  • In the entity, we should join the article and user tables together so that it contains a full copy of the user
• A title field which's value is a string which is the title of the article
• A slug field which's value is a unique string which is the path to the article
• A status field which's value is an enum of APPROVED, PENDING, REJECTED (default of PENDING)
• A description field which is a string (max length of 250 characters)
• A content field which is a string

You can use the @CreateDateColumn() and @UpdatedDateColumn() decorators to automatically handle the dates.

Why Will This Benefit The Community?

This will allow us to create and retrieve articles throughout the API, therefore unlocking the ability to serve them across the web and Discord.

What Functionality Are You Suggesting?

The ability to get a specific user from the database via the API.

API Endpoint

GET /user/:id

Users are not required to be authenticated to access this endpoint.

Response Body

{
    id: number;
    username: string;
    created: string; // simplified extended ISO formatted date
    modified: string; // simplified extended ISO formatted date 
}

Why Will This Benefit The Community?

This functionality will allow us to display user profiles on the web and in Discord.

Resolves #18

Not entirely happy with the tests yet. ~~80~~ 90% sure it works fine, ~~but haven't had the chance yet to attach it to a controller to really test it out~~.

Overview

Please bullet point the changes you have made below:

• Adds user entity
• Adds user service with all methods
• Adds basic test cases for the service

What Functionality Are You Suggesting?

To start our work on supporting users we need a new user module that houses a user entity and user service. The user service should include the ability to:

• Get one specific user by ID
• Get one specific user by Auth ID
• Get all users
• Create a new user
• Update a user by ID

A user entity should have:

• An auto-incrementing, unique and primary key id field (e.g. 1)
• A unique auth_id field (e.g. auth0|6237590b647a36006ba3c6ea)
• A unique username field (e.g. johnsmith)
• A created field which contains the date at which the entity was created
• A modified field which's value updates each time the entity is modified

You can use the @CreateDateColumn() and @UpdatedDateColumn() decorators to automatically handle the dates.

Why Will This Benefit The Community?

This will allow us to create and retrieve users throughout the API, therefore unlocking the ability to serve them across the web and Discord.

Resolves #3

Overview

• Installed @nestjs/typeorm, typeorm, and pg (postgres package used by typeorm)
• Added database credentials to .env.example and to .env.docker.local
• Added Dockerfile and docker-compose.yaml for running NestJS app and PostgreSQL instance in docker containers with no configuration required
• Updated Readme.md with instructions for getting started
• Import TypeOrmModule into root module found in src/app.module.ts. This automatically connects to PostgreSQL on startup.

Blocked by #1

What Functionality Are You Suggesting?

We want to set up a swagger document that automatically contains all the documentation for the API. This functionality is described on the NestJS documentation here.

Why Will This Benefit The Community?

This will allow developers who are working on the frontend to easily see how to save and retrieve data to and from the API.

Blocked by #1

What Functionality Are You Suggesting?

We want to set up an ORM so that we can interact with our database. This can be achieved through TypeORM and is documented here (note this uses MySQL in the example, but we want to use PostgreSQL).

You do not need to set up any repositories as part of this ticket, just the initial connection.

The following environment variables should be used for the database connection:

• DATABASE_HOST (host)
• DATABASE_PORT (port)
• DATABASE_USER (username)
• DATABASE_PASS (password)
• DATABASE_NAME (database)

If NODE_ENV is not production then synchronize should be true.

We should also have a Docker Compose setup for running the API and database together.

Why Will This Benefit The Community?

This will benefit us by allowing us to have a place where we can save and retrieve data from.

What Functionality Are You Suggesting?

• Switch over POST /user to use API key validation so that it can be used in an Auth0 Action
• Update Open API documentation
• Hookup deployment to Heroku

Why Will This Benefit The Community?

This will make the API ready for use in production

Resolves #29

Overview

• Added GET /article endpoint to retrieve all articles
  • Added ?userId query parameter to retrieve all articles by a given user
  • Added ?status query parameter to retrieve articles of a given status

What Functionality Are You Suggesting?

The ability to get all users from the database via the API.

API Endpoint

GET /user

Users are not required to be authenticated to access this endpoint.

Response Body

[
    {
        id: number;
        username: string;
        created: string; // simplified extended ISO formatted date
        modified: string; // simplified extended ISO formatted date 
    },
    ...
]

Why Will This Benefit The Community?

This functionality will allow us to display user profiles on the web and in Discord. As we are statically rendering via Next.js we need to be able to get every user ahead of time, rather than on demand.

Since people run CI on their forks end to end tests do not run successfully, therefore blocking the merge process. For the time being, we will remove this job from CI as we investigate a better solution.

What Functionality Are You Suggesting?

Currently we are not marking any optional query parameters as optional within Swagger, therefore causing them to be labelled as required when they are not. This currently only affects this route.

This can be achieved by using the @ApiQuery() parameter decorator.

Why Will This Benefit The Community?

This means the documentation will correctly reflect the implementation, avoiding any confusion which may be caused.

What Functionality Are You Suggesting?

We should introduce a new GET /article/search endpoint which returns any articles which contain a given query. We can use PostgreSQL's LIKE operator to return the results. As we do not have much data this is a perfectly fine solution - there is no need for something like ElasticSearch.

Why Will This Benefit The Community?

This will allow members to easily search for articles, primarily through auto-complete functionality of an upcoming /article command within our Discord community.

What Functionality Are You Suggesting?

Instead of grouping all our API endpoints together we should have them split out into sections based on their module name. For example: user, article...

See documentation here.

Why Will This Benefit The Community?

This will make the documentation more organised and easier to follow.

What Functionality Are You Suggesting?

We should remove the App controller and service file which was used during setup and testing.

Why Will This Benefit The Community?

The application is now running in production and these serve no purpose.

What Functionality Are You Suggesting?

Due to issues with running end to end tests on forked repositories, we removed them from CI. However, end to end tests are a very powerful safety net for your codebase so it would be ideal to add them back. We should explore ideas on how to best do this.

test_e2e:
    name: End to End Tests
    runs-on: ubuntu-latest
    container: node:16.14.2-alpine3.15
    env:
      DB_PASS: ruv2cyHT4H78BVzK7b4DnBAHJR8dG4fG
      DB_USER: admin
      DB_NAME: codesupport-api
    services:
      postgres:
        image: postgres:12
        env:
          POSTGRES_PASSWORD: $DB_PASS
          POSTGRES_USER: $DB_USER
          POSTGRES_DB: $DB_NAME
        options: >-
          --health-cmd pg_isready
          --health-interval 10s
          --health-timeout 5s
          --health-retries 5
        ports:
          - 5432:5432
    steps:
      - uses: actions/checkout@v2
      - name: Runs End to End Tests
        env:
          DATABASE_HOST: postgres
          DATABASE_PASS: $DB_PASS
          DATABASE_USER: $DB_USER
          DATABASE_NAME: $DB_NAME
          AUTH0_ISSUER_URL: ${{ secrets.AUTH0_ISSUER_URL }}
          AUTH0_AUDIENCE: http://localhost:8080
          AUTH0_DOMAIN: ${{ secrets.AUTH0_DOMAIN }}
          AUTH0_CLIENT_ID: ${{ secrets.AUTH0_CLIENT_ID }}
          AUTH0_CLIENT_SECRET: ${{ secrets.AUTH0_CLIENT_SECRET }}
        run: |
          npm ci
          npm run migration:run
          npm run seed:run
          npm run test:e2e

Why Will This Benefit The Community?

Allow us to quickly and easily catch bugs when making changes to the source code.

What Functionality Are You Suggesting?

We should create a custom logger which when NODE_ENV is production logs to our Logtail instance and in any other environment uses the regular Node console.

A new environment variable called LOGTAIL_TOKEN should be added and provided to the Logtail constructor.

Useful Resources

• NestJS - Custom Logger Implementation
• Logtail JavaScript SDK Documentation

Why Will This Benefit The Community?

This will allow us to easily trace issues, query logs and solve bugs without the need to SSH into the server.