Hi,
I’m using rails 1.2.3 and mysql 5
I have a collection of models like this:
Car has_many Passengers;
Passenger has_many Pocket_Items
I want to issue a Car.find command from rails where I find all cars
where at least 1 passenger is carrying a knife (or whatever) in his/
her pocket.
The details are shown below:
Table 1: cars (id, license_plate, make, model)
Table 2: passengers (id, car_id, first_name, last_name)
Table 3: pocket_items (id, passenger_id, item_description)
The description of the Car model would be something like this:
class Car < ActiveRecord::Base
has_many :passengers
has_many :pocket_items, :through => :passengers
end
so, e.g. car # 1 may contain 2 passengers. passenger 1 may have a
wallet and a knife, whereas passenger 2 may have a knife and keys.
So, now let’s say that I want to find all cars where at least 1
passenger is carrying a knife.
in SQL, I might issue the following statement
select
c.id,
c.license_plate,
c.make,
c.model
from cars c
inner join passengers p on (p.car_id = c.id)
inner join pocket_items pi on (pi.passenger_id = p.id)
where pi.item_description = ‘Knife’
this works, except that if a car contains 2 people, each carrying a
knife, the same car will be returned on 2 separate rows.
obviously, this can be eliminated by changing the statement above to
“select distinct …”
My questions are:
-
Is there another (more elegant) way of demanding an INNER JOIN from
rails aside from issuing it directly from the :joins option within the
find method? -
Is there a rails idiom for ‘deduping’ my result set (i.e. I only
want a result set of UNIQUE cars from my find query. From the agile
development book, ed 2 (page 341), one can ‘dedupe’ child objects
using the modifier :uniq => true. However, i’m not sure if there’s a
way to do this on the parent object. -
are there any (more efficient) approaches to using rails to
accomplish the query outlined above? i.e. I want to find all cars
where at least 1 passenger carries a knife.
thanks!