Exploring Ahoy Analytics With Rails 7

This post will cover my setup and first look at using Ahoy with a Rails application.

Nothing more than the basics will be covered, but it will give you a look at how you can quick-start your own Ahoy-analytics application.

Source code can be found here.

Prerequisites

Getting started

We will use Rails to initialize the project demo-rails-ahoy:

# Create a new rails project
$ rails new demo-rails-ahoy
$ cd demo-rails-ahoy

# Install gem
$ bundler add ahoy_matey

# Generate install
$ bin/rails g ahoy:install
$ bin/rails db:migrate

# Generate a controller for testing our analytics
$ bin/rails g controller home show
copy

In the above, we add Ahoy with the gem ahoy_matey, use the Ahoy generate for the installation process and then migrate the database for the changes made.

Finally, we create a Home controller for use to setup a basic "show" route localhost:3000/home/{id} that we can create some basic analytics against.

In our example today, we will just be showing how to track an event with the track method.

Updating our routes

Update config/routes.rb for the following:

Rails.application.routes.draw do
  resources :home, only: [:show]
end
copy

This will enable the endpoint /home/{id} that we are aiming for.

Next is to set it up in the home controller.

Updating our controller

We will be making a very contrived example for our controller.

Update the app/controllers/home_controller.rb file to the following:

class HomeController < ApplicationController
  def show
    ahoy.track 'home#show', search_id: params[:id]
    render json: { message: 'ok' }
  end
end
copy

At this point, our application is actually prepped to record our events.

To debug and test with cURL

We will do some testing using cURL, so the last thing we need to do is update our configuration just for that.

Update the config/initializers/ahoy.rb file to disable Ahoy.quiet and enable Ahoy.track_bots:

class Ahoy::Store < Ahoy::DatabaseStore
end

# set to true for JavaScript tracking
Ahoy.api = false

Ahoy.quiet = false
Ahoy.track_bots = true

# set to true for geocoding (and add the geocoder gem to your Gemfile)
# we recommend configuring local geocoding as well
# see https://github.com/ankane/ahoy#geocoding
Ahoy.geocode = false
copy

Setting those two values will override the defaults.

Running our code

In one terminal window, run the Rails server bin/rails s.

In another, let's run some cURL commands with different routes:

$ curl localhost:3000/home/1
{"message":"ok"}%
$ curl localhost:3000/home/2
{"message":"ok"}%
$ curl localhost:3000/home/nicer-id
{"message":"ok"}%
copy

In the Rails server logs, you will see a number of Ahoy::Visit and Ahoy::Event creations happening, which we can explore further in the Rails console.

In another terminal, fire up the Rails console bin/rails c and run the following:

$ bin/rails c
irb(main):001:0> Ahoy::Event.all.count
   (0.7ms)  SELECT sqlite_version(*)
  Ahoy::Event Count (0.1ms)  SELECT COUNT(*) FROM "ahoy_events"
=> 3
irb(main):002:0> Ahoy::Event.last
  Ahoy::Event Load (0.2ms)  SELECT "ahoy_events".* FROM "ahoy_events" ORDER BY "ahoy_events"."id" DESC LIMIT ?  [["LIMIT", 1]]
=>
#<Ahoy::Event:0x00007fe6eb124808
 id: 3,
 visit_id: 3,
 user_id: nil,
 name: "home#show",
 properties: {"search_id"=>"nicer-id"},
 time: Tue, 15 Mar 2022 04:42:47.915479000 UTC +00:00>

# Checking out the Ahoy::Visit values
irb(main):003:0> Ahoy::Visit.all.count
  Ahoy::Visit Count (0.1ms)  SELECT COUNT(*) FROM "ahoy_visits"
=> 3
irb(main):004:0> Ahoy::Visit.last
  Ahoy::Visit Load (0.2ms)  SELECT "ahoy_visits".* FROM "ahoy_visits" ORDER BY "ahoy_visits"."id" DESC LIMIT ?  [["LIMIT", 1]]
=>
#<Ahoy::Visit:0x00007fe6ec5235c0
 id: 3,
 visit_token: "[FILTERED]",
 visitor_token: "[FILTERED]",
 user_id: nil,
 ip: "::1",
 user_agent: "curl/7.64.1",
 referrer: nil,
 referring_domain: nil,
 landing_page: "http://localhost:3000/home/nicer-id",
 browser: "curl",
 os: nil,
 device_type: nil,
 country: nil,
 region: nil,
 city: nil,
 latitude: nil,
 longitude: nil,
 utm_source: nil,
 utm_medium: nil,
 utm_term: nil,
 utm_content: nil,
 utm_campaign: nil,
 app_version: nil,
 os_version: nil,
 platform: nil,
 started_at:
  Tue, 15 Mar 2022 04:42:47.904552000 UTC +00:00>
copy

In the above, we note that each request generated an Ahoy::Visit object as well as some Ahoy::Event events (as we expected).

After checking the count, we also logged out the last entry for each to get a look at what sort of information is eligible for collection.

Summary

Today's post demonstrated how to set up Ahoy analytics and also took us through some basic examples of how that might look.

Resources and further reading

Photo credit: bfilm