This is done so that the front and the beck development do not depend on each other and run in parallel. There is one server that returns static files and does nothing else. And there is a second server in which the logic of working with data is registered.

On the front, ajax requests can be blocked while the API is not ready, and the back is not a gentle front for developers.

The most primitive example is all on one server.

api.js

The logic of working with data

const express = require('express'); // некая база данных const db = require('./db'); const router = express.Router(); // что-то делаем router.post('/users/:id', async (req, res) => { try { const id = req.params.id; const user = await db.fetchUserById(id); res.status(200).json(user); } catch (e) { res.sendStatus(500); } }); module.exports = router; 

index.js

 const path = require('path'); const express = require('express'); const bodyParser = require('body-parser'); // Наше API const api = require('./api'); const PORT = process.env.PORT; const indexHTML = path.resolve(__dirname,'../public/index.html'); const app = express(); app.use(bodyParser.json()); // Для статических файлов app.use('/', express.static('public')); // Подключаем API // Теперь мы можем делать ajax запросы // и получить пользователя с id 1 по урлу /api/users/1 app.use('/api', api); app.get('/*', (req, res) => res.sendFile(indexHTML)); app.listen(PORT, () => console.info('Server is running on port:', PORT));