This post originated from an RSS feed registered with Ruby Buzz by Obie Fernandez. Original Post: counter_cache sneakiness Feed Title: Obie On Rails (Has It Been 9 Years Already?) Feed URL: http://jroller.com/obie/feed/entries/rss Feed Description: Obie Fernandez talks about life as a technologist, mostly as ramblings about software development and consulting. Nowadays it's pretty much all about Ruby and Ruby on Rails. Latest Ruby Buzz Posts Latest Ruby Buzz Posts by Obie Fernandez Latest Posts From Obie On Rails (Has It Been 9 Years Already?)

The Question

Given this pair of ActiveRecord models:

class Firm < ActiveRecord::Base
  has_many :votes
end

class Vote < ActiveRecord::Base
  belongs_to :firm, :counter_cache => :vote_total
end

And this test case:

class FirmTest < Test::Unit::TestCase
  
  fixtures :firms, :votes

  def test_vote_total_should_increment_when_new_session_reports
    firm = firms(:first)
    assert_equal 4, firm.vote_total
    firm.report('experimental','aaaaa','')
    assert_equal 5, firm.vote_total
  end
end

Why did I spend more time than I care to admit this evening trying to figure out if the :counter_cache option for belongs_to is broken?

The Answer

I'm guessing if you know the answer it's because you've been bitten by this little gotcha already. I figured it out by examining the test coverage for this feature of ActiveRecord:

  def test_counter_cache
    topic = Topic.create :title => "Zoom-zoom-zoom"
    assert_equal 0, topic[:replies_count]
    
    reply = Reply.create(:title => "re: zoom", :content => "speedy quick!")
    reply.topic = topic

    assert_equal 1, topic.reload[:replies_count]
    assert_equal 1, topic.replies.size

    topic[:replies_count] = 15
    assert_equal 15, topic.replies.size
  end

See that sneaky call to reload?

The revised test for my code, which did finally pass:

   def test_vote_total_should_increment_when_new_session_reports
    firm = firms(:first)
    assert_equal 4, firm.vote_total
    firm.report('experimental','aaaaa','')
    
    # Annoying that firm has to be reloaded here
    # but in practice doesn't hurt anything, since requests that
    # alter the total will be redirected, which reloads firm anyway.
    assert_equal 5, firm.reload.vote_total
  end

What do you think?

Is the need to invoke reload to pickup the up-to-date value for the counter cache a bug or a feature?