MessageProcessor

Message processing application built with Elixir, Phoenix, and Redis

Run It

Clone this repository locally and cd into new directory
Install dependencies mix deps.get
(Optional) Uncomment the following in config/config.exs to reduce noise in terminal where server is running # config :logger, level: :error
Verify redis is running locally at 127.0.0.1:6379
Start server mix phx.server
The application will run at localhost:4000

How It Works

Make GET request to /receive-message

The endpoint accepts the following query string parameters

message (required)
queue (optional) default queue name can be edited in config/config.exs
```
config :message_processor,
  default_queue_name: "default"
```

After making a GET request to /receive-message?message=yourmessage[&queue=somequeue]

You should immediately receive a status code of 200.

A GenServer MessageProcessor.ProcessingQueue will process the message based on the order it was received.

The GenServer will

persist your message in a Redis list named whatever you specified as the queue (or the default queue name if you didn't specify a queue param)
Add a reference to the queue of your message to a List representing a ProcessingQueue (this list is kept in the state of the GenServer)
Every second, GenServer will look at ProcessingQueue's queue kept in state, take the first item added (which is a reference to the Redis list/queue), get the first message that was added to that Redis list/queue, and process that message (print it in the server window)

Try It Out

Make sure your server is running (mix phx.server)

One Request at a Time

Try making a single GET request to the /receive-message endpoint

# --include flag will include HTTP headers of response
curl --include "http://localhost:4000/receive-message?message=message1&queue=queue1"

You should see message1 print in the terminal window where the server was started.

Many Requests at a Time

Try making many GET request to the /receive-message endpoint.

curl --include "http://localhost:4000/receive-message?message=message1&queue=queue1"
curl --include "http://localhost:4000/receive-message?message=message2&queue=queue2"
curl --include "http://localhost:4000/receive-message?message=message3&queue=queue3"
curl --include "http://localhost:4000/receive-message?message=message4&queue=queue4"
curl --include "http://localhost:4000/receive-message?message=message5&queue=queue5"
curl --include "http://localhost:4000/receive-message?message=message6&queue=queue6"
curl --include "http://localhost:4000/receive-message?message=message7&queue=queue7"
curl --include "http://localhost:4000/receive-message?message=message8&queue=queue7"

You should see the values of all message params print in the terminal window in the order they were received.

Each message will print one second apart.

Plans To Scale

This is not a production application and is not expected to scale.

If I were implementing this in a production environment, I would look into using Amazon SQS in combination with BroadwaySQS

I would also ensure that all connections are secure and using SSL.