Chatter

Permit to register users and its settings, also manage websocket connections and messages. The app has friendly errors, validation, also a simple versioning was made.

Table of Contents

• Installing
  • Configuring
    • .env
    • SQLite
      • Migrations
• Usage
  • Error Handling
    • Errors Reference
  • Versioning
  • Routes
    • Requests
• Running the tests
  • Coverage report

Installing

Easy peasy lemon squeezy:

$ yarn

Or:

$ npm install

Was installed and configured the eslint and prettier to keep the code clean and patterned.

Configuring

The application use just one database: SQLite. For the fastest setup is recommended to use docker-compose, you just need to up all services:

$ docker-compose up -d

.env

In this file you may configure your Redis database connection, JWT settings, the environment, app's port and a url to documentation (this will be returned with error responses, see error section). Rename the .env.example in the root directory to .env then just update with your settings.

key	description	default
PORT	Port number where the app will run.	3333
DOCS_URL	An url to docs where users can find more information about the app's internal code errors.	https://github.com/DiegoVictor/chatter#errors-reference

SQLite

Store the users, settings, messages and connections. For more information to how to setup your database see:

• typeorm

You can find the application's ormconfig.js file in the root folder.

Migrations

Remember to run the database migrations:

$ yarn ts-node-dev ./node_modules/typeorm/cli.js migration:run -- -d ./src/database/datasource.ts

Or:

$ yarn typeorm migration:run -- -d ./src/database/datasource.ts

See more information on TypeORM Migrations.

Usage

To start up the app run:

$ yarn dev:server

Or:

npm run dev:server

Error Handling

Instead of only throw a simple message and HTTP Status Code this API return friendly errors:

{
  "statusCode": 400,
  "error": "BadRequest",
  "message": "Setting already exists",
  "code": 140,
  "docs": "https://github.com/DiegoVictor/chatter#errors-reference"
}

Errors are implemented with @hapi/boom. As you can see a url to error docs are returned too. To configure this url update the DOCS_URL key from .env file. In the next sub section (Errors Reference) you can see the errors code description.

Errors Reference

code	message	description
140	Setting already exists	Already exists a settings to provided id.
144	Setting not found	The provided id not references an existing setting in the database.
244	User not found	The user id sent does not references an existing user in the database.
245	User not found	Is not possible to retrieve messages from a user not found in the database.

Versioning

A simple versioning was made. Just remember to set after the host the /v1/ string to your requests.

POST http://localhost:3333/v1/settings

Routes

route	HTTP Method	params	description
/settings/:id	GET	id of the setting.	Return the user's setting.
/settings	POST	Body with settings user_id and chat property.	Create new setting.
/settings/:id	PUT	id of the setting and body with chat new value.	Update an user's setting.
/users	POST	Body with user's email.	Create a new user.
/users/:id/messages	GET	id of the user.	Return user's messages.
/messages	POST	Body with message data.	Save user new message.

Requests

• POST /settings

Request body:

{
  "user_id": "17c6f66d-a24b-45e0-8c3f-2b3411e51fe8",
  "chat": true
}

• PUT /settings/:id

Request body:

{
  "chat": true
}

• POST /users

Request body:

{
  "email": "johndoe@example.com"
}

• POST /messages

Request body:

{
  "user_id": "d01bc88b-15cb-4478-830f-edc44577d707",
  "admin_id": null,
  "text": "Lorem ipsum doolor sit amet"
}

Running the tests

Jest was the choice to test the app, to run:

$ yarn test

Or:

$ npm run test

Coverage report

You can see the coverage report inside tests/coverage. They are automatically created after the tests run.