wearescytale/amqp-nodemailer-sparkpost-consumer

AMQP Nodemailer Consumer for SparkPost

Introduction

This is a simple consumer for a messaging queue system (using the AMQP protocol) already setup for use with a nodemailer transport for SparkPost.

Requirements

• Node 7.10.0
• An MQ server (ActiveMQ, RabbitMQ, etc)
• SparkPost API key

Installation

Clone this repository and run yarn.

Be sure to fill in your details in the configuration/configs.js file then run npm start.

To run in production you can simply run:

npm run start:production

For a development environment (automatic restarts upon file changes) run:

npm run start:dev

Configuration

The following options may be overriden in the appropriate switch case statement.

• debug (Boolean) - Turns the application more verbose. Be careful with this option in production environments because printing to stdout is a synchronous operation.
• sandbox (Boolean) - If enabled, the application won't contact SparkPost and will simply acknowledge the received message.
• amqp (String) - The URL of your AMQP server.
• queue (String) - The name for the queue the application creates and uses.
• messagePrefetch (Integer) - How many messages sent over the channel can be awaiting acknowledgement.
• sparkpost (Object) - The SparkPost Nodemailer transport options. The only required property of this object is the sparkPostApiKey.

Whenever the NODE_ENV environment variable is prod, production or staging, the following values are read from environment variables by default:

Property	Environment Variable
amqp	NMSP_AMQP_URL
queue	NMSP_QUEUE_NAME
debug	NMSP_DEBUG
sandbox	NMSP_SANDBOX
sparkpost.sparkPostApiKey	NMSP_SPARKPOST_API_KEY

Note: For the NMSP_DEBUG and NMSP_SANDBOX environment variables, only the existence of these variables is checked and the value is never used. As long as these variables have a value, their respective options will be set to true.

Adding emails to the queue

You can test sending emails through the message queue by adding the following code to the bottom of consumer.js and running npm run start:dev

Be sure to have a SparkPost API key setup in the configuration/configs.js file.

/**
    Provider
 */

// Reuse the pre-existing connection
connection.then((conn) => {
    // Create a new channel
    return conn.createChannel();
})
.then((ch) => {
    // Make sure the queue exists.
    return ch.assertQueue(q, {
        durable: true
    })
    .then((ok) => {
        // Build a transport message
        let message = {
            from: '"Sender" <sender@example.com>',
            to: '"Receiver" <receiver@example.com>',
            subject: 'Example Subject',
            text: 'Example Text',
            html: '<h1>Example Rich Text</h1>'
        };

        // Send it to the message queue as a buffer.
        // #sendToQueue accepts an optional callback to check if the
        //  message was acked or nacked
        return ch.sendToQueue(q, Buffer.from(JSON.stringify(message)), {
            persistent: true,
            contentType: 'application/json'
        }, (err, ok) => {
            if (!err) {
                console.log('Message Acked!');
            } else {
                console.log('Message Nacked!');
            }
        });
    });
})
.catch(console.error);