What's the reason for this library to exist?

[punkpeye]

Hey,

I am just getting familiar with notifications in postgres.

Why can't I just do something like:

const { Client } = require('pg');
const connectionString = process.env.DATABASE_URL;
const client = new Client({ connectionString });
async function listenToNotifications() {
  try {
    // Connect to Postgres
    await client.connect();
    // Listen to specific channel in Postgres
    // Attach a listener to notifications received
    client.on('notification', (msg) => {
      console.log('Notification received', msg.payload);
    });
    await client.query('LISTEN channel_name');
    console.log('Listening for notifications on my_channel');
  } catch (e) {
    console.log(e);
  }
}

What purpose does this library serve that's not covered by this?

[aldis-ameriks]

You can do that and that's perfectly fine.

The library provides a few conveniences to reduce some of the boilerplate that you may have to implement, e.g.

• handling multiple listeners over a single connection
• proper listener clean-up to avoid memory leaks
• handling reconnections
• deserializing the payload safely, avoiding prototype poisoning
• validating the maximum payload size when publishing, you can also let postgres handle the failure, however, that can lead to additional side effects which in your case may or may not matter