In my last post I was asking how I could check for the existance of a record before inserting a new one and this was the helper method that I was using to do this: def already_friend_of?(friend) begin self.friends.find(friend.id) true rescue false end end I was told that this is not very efficient, as it hits the db every time I call it. I completely agree and would like to do this in a more optimized way. I was told to create another method that reads all the friend_id's of my user into an array and work on that array instead. I think this is a good approach, since I would have to hit the db just once to get that information and then use the array from then on. My question is.. how do I go about doing that and where would my code be? In the Model? In the Controller? In a helper file? I also need some help writing the actual ruby code to define an array of friend_id's. My User model HBTM Friends. So I can call my_user.friends.find(:all) and it will return all of his/her friends. Does my_user.friends return an array of only those friends that belong to my_user? If so, can I just assign that to a variable in my User model and access it in my controllers and views? Any help would be greatly appreciated.
on 2006-01-28 21:12
on 2006-01-28 21:54
There are a variety of techniques you can use to perform optimizations when you need them. You should always wait until you have measured your application's performance before you perform these optimizations. It might turn out that this functionality isn't a big issue for your app, in which case, the code as written is all you need. Building your app in a way that allows you to isolate changes when you need to optimize is the right way to go. As you build your app, it may be that this method will be removed completely, to be replaced with something different - optimizing early is then a complete waste of time. When you do get to the optimization phase, you should run in production mode, with Active Record logging set to debug (not the default in production) so you see the SQL queries in the log file. Then you'll run your application load against the server, see what is too slow under load, and work out which queries to tackle first.