2/26 Suppose I have a database of car dealers and the cars they have in
stock. I want to query this database for N cars that I'm
interested in, something like
SELECT dealerId, carId FROM cars WHERE (carId=1 OR carId=2 OR carId=42)
The problem is, I need to sort the result such that a dealer who
has all N cars I'm interested in is ranked higher than a dealer that
has only 1 or 2 such cars. Can this be done with some sort of
aggregate function? Something like
SELECT dealerId, count(dealerId) as numCars, list(carId)
FROM cars WHERE (carId=1 OR carId=2 OR carId=42)
GROUP BY dealerId
ORDER BY rank
where the output would be
dealer | numCars | list(carId)
----------------------------------------
Fred 3 1,2,42
Bob 2 1,2
Tom 2 1,42
Jerry 1 2
Thanks in advance.
\_ Please tell me this is for work.
\_ I would, but then I wouldn't get the input from the smart types
who do it because it's challenging. -op
\_ Your question was better when it was jive'd.
\_ Yo! Yo! Shiznitz! Wez be needin' ah jive update! |
http://csua.com/feed/