1.9.3p194 :002 > u = User.find_by_email("[email protected]")
1.9.3p194 :005 > u.addresses.size
=> 1
1.9.3p194 :006 > u.addresses.length
=> 1
1.9.3p194 :007 > u.addresses.count
There is no difference between size, length and count in Rails 3.2.3, isn't it?
length will load all your objects just to count them; something like:
select * from addresses...
and then return the results count. As you can imagine - it's bad performance
count will just issue
select count(*) from addresses...
which is better, because we are not loading all addresses just to count them
size is smarter - it'll check if the association is already loaded and if true then return the length (without issuing a call to the database).
size also checks for counter_cache if you have a field named address_count in your user model, then size will use this field for the count, so there is no need to issue a count on the addresses table.
if all fails, size will issue a select count(*)
on the database
1) In rails count is a ActiveRecord method, therefore count can be applied on model name directly as:
> User.count
(1.4ms) SELECT COUNT(*) FROM "users"
=> 1
But size is not ActiveRecord method therefore it will throw an error
> User.size
"NoMethodError".
2) Size in rails can be used with ActiveRecord array (ie. size is Array method)
> User.all.size
(1.2ms) SELECT COUNT(*) FROM "users"
=> 1
3) count will always fire ActiveRecord query. but size will not fire ActiveRecord query, only if record or ActiveRecord array is already loaded (executed) as:
> d=User.all
User Load (18.1ms) SELECT "users".* FROM "users"
=> #<ActiveRecord::Relation [#<User id: 1, email: "[email protected]", name: "mano", dob: "2017-02-16", address: "fasfafasf", created_at: "2017-02-12 08:16:12", updated_at: "2017-02-12 09:34:07", online: false>]>
2.3.3 :009 >
2.3.3 :010 > d.count
(1.3ms) SELECT COUNT(*) FROM "users"
=> 1
2.3.3 :011 > d.size
=> 1
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With