adon-api
Express JS + Typescript + Jest development made easier. The goal is for us to easily build websites or APIs without managing the same dependencies and breaking out of our standards and conventions each time. We follow convention and configuration over code so that all future projects are handled easily.
Project stats
Running the sample
On the root of this project is a simplified sample project. Run it with the following command...
npm run start:sample
Then from your browser, check the route endpoints for these
How it works?
Technologies used:
- Express JS - our core framework is based on express and so this module will provide a library to easily kickstart an express js web service.
- TypeScript - Our projects will be based only on TypeScript so make sure this is installed on your system.
- DOTENV - Loading process.env for environment specific configurations
- Winston - default logger mechanism
Configurations
For now we reverted to use a simple hardcoded config object and injected the process.env (Environment Variables) to it
Default Configs
- API.LOGGING.LEVEL - The default logging level that gets rendered in the console (default: 'debug')
- API.PORT - The default PORT to be used by the server (default: 3000)
- API.SECURITY.CORS_ORIGINS - The CORS Origins to be allows (default: '*')
logger
The built-in logger (winston) is set to utilize the common 6 logging levels in my dev experience. You can access the logger instance as part of the ExpressApp instance's 'log' property.
// usage log.{level}( message, meta )
ExpressApp.log.info('Information here', { traceToken: 'SOME-GUID'})
ExpressApp.log.error('An error occured here', { traceToken: 'SOME-GUID', err })
const levels = {
crit: 0, // Critical errors that impact the stability of the system rendering it useless or break
error: 1, // Errors that affect a process flow that may will need attention
warn: 2, // Warnings that may affect requests or the entire service
info: 3, // Regular log informations that are non-issue but may help understand the service state
debug: 4, // debug level logs that should not be in production
verbose: 5, // verbose information that may include even PII that should never be in production
}
RouteManager
The RouteManager structures how REST routes are defined on the project. It also adds generic logs and a 'trace-token' to the request header that we can use to better debug logs for a particular request (user must make sure to add this explicitely to the meta of the logs). To define your routes, create a {*.rt.ts} file in the project's "routes" folder. It should export as default a function named route that accepts 2 parameters (the ExpressApp instance and the express router object).
import { ExpressApp, Router, Request, Response } from 'adon-api'
export default function route(app: ExpressApp, router: Router): void {
router.get('/', async (req: Request, res: Response) => {
res.send('This is the root main route')
app.log.debug('Response sent!', { traceToken: req.headers['trace-token'] })
})
router.get('/test', async (req: Request, res: Response) => {
res.send('This is the root test route')
app.log.debug('Response sent!', { traceToken: req.headers['trace-token'] })
})
}
Installation
install the following
* adon-api framework (this)
npm i adon-api --save
Redis Dependency
As mentioned above, this API depends on Redis particularly for the RateLimitter. I would recommend the use of redis for session management and other caching concerns as well. You can run it wherever and however you want. If you have docker for example (which I use for my developments) then just run the following to spin a new redis instance
docker run --name local-redis -p 6379:6379 -d redis
How to use
Create index.ts in your project root and enter the following
import { ExpressApp, ExpressAppConfig, EnvConfig } from 'adon-api'
let app: ExpressApp
async function onHealthCheck(): Promise<boolean> {
return true
}
async function onLoading(): Promise<void> {
}
async function onReady(): Promise<void> {
app.start()
}
const cfg: ExpressAppConfig = {
port: parseInt(EnvConfig.get('PORT')) || 3000,
onHealthCheck,
onReady,
onLoading,
}
app = new ExpressApp(cfg)
Environment configurations
- PORT - sets the port number where the service will listed to. (Default: 3000)
- SERVER_CORS_ALLOWED_ORIGINS - sets the allowed request origins urls. multiple URLs are divided by semicolon. (Default: '*')
Either set in your machine environment the values or create a file named '.env' in the root of the application then enter the following
PORT=3000
SERVER_CORS_ALLOWED_ORIGINS=http://localhost.com;http://bytecommander.com;http://bcomm-local.com
You can add more configurations here as much as you like and they can be accessed in code as
import { EnvConfig } from 'adon-api'
const port = EnvConfig.get('PORT')
EnvConfig.set('APP_VAR', 'Hello world')
Routes
Adding new routes is simple and organized in this framework. First create a 'routes' folder in the root.
Add new file inside the routes folder with the 'rt.ts' extension. To create a root route '/' name the file 'index.rt.ts'
You can also create routes on specific paths based on the filename. To make a route on path '/users/properties' name the file 'users_properties.rt.ts'
For this sample, lets create a file named 'index.rt.ts' and 'users_test.rt.ts' then enter the following in both files
import { Router, Request, Response } from 'express'
import { ExpressApp } from '../../src/libs/ExpressApp'
export default function route (app: ExpressApp, router: Router): void {
router.get('/', async (req: Request, res: Response) => {
res.send("WOW!")
})
router.get('/test', async (req: Request, res: Response) => {
res.send("YO!")
})
}
Run the application and use postman to invoke GET on 'http://localhost:3000/', 'http://localhost:3000/test' and 'http://localhost:3000/users/properties/