ten cheapest room for each city

I have 3 table Room, Hotel, City, how can i select ten cheapest room and its hotel_id for each city possible to execute the subquery for each city?

SELECT price, room_id, hotel_id, city_id
    FROM Hotel 
    JOIN  Room USING(hotel_id)
ORDER BY price
LIMIT 10


Scheme
CREATE TABLE `City` (
  `city_id` int(11) NOT NULL AUTO_INCREMENT,
  `city_name` varchar(50) NOT NULL,
  PRIMARY KEY (`city_id`)
) 

CREATE TABLE `Hotel` (
  `hotel_id` int(11) NOT NULL AUTO_INCREMENT,
  `city_id` int(11) NOT NULL,
  `hotel_name` varchar(100) NOT NULL,
   PRIMARY KEY (`hotel_id`),
   CONSTRAINT `Hotel_FK_1` FOREIGN KEY (`city_id`) REFERENCES `City` (`city_id`) ON DELETE CASCADE ON UPDATE CASCADE
)

CREATE TABLE `Room` (
  `room_id` int(11) NOT NULL AUTO_INCREMENT,
  `hotel_id` int(11) NOT NULL,
  `room_name` varchar(255) DEFAULT NULL,
  `room_price1` int(11) DEFAULT NULL,开发者_运维百科
  PRIMARY KEY (`room_id`),
  CONSTRAINT `Room_FK_1` FOREIGN KEY (`hotel_id`) REFERENCES `Hotel` (`hotel_id`) ON DELETE CASCADE ON UPDATE CASCADE
)

query with subquery

SELECT 
     r.room_price1, r.room_id, h.hotel_name, c.city_name
FROM Hotel h 
INNER JOIN Room r ON
    h.hotel_id = r.hotel_id
INNER JOIN City c USING(city_id)
WHERE (
   SELECT count(*) from Room as rt
   JOIN Hotel ht using(hotel_id)
   where c.city_id = ht.city_id and rt.room_price1 < r.room_price1 OR (rt.room_price1 = r.room_price1 AND rt.room_id < r.room_id)
) < 3 
ORDER BY c.city_name, h.hotel_id, r.room_price1

if some hotels have some room_price1 this query return more then 3 row for city and i need 3 unique hotel per city this query can return 3 room from one hotel

ill try use user vars

set @num := 0, @type := '';
select *
from (
   select r.room_price1 pr, r.room_id, h.hotel_name, c.city_name,
      @num := if(@type = city_id, @num + 1, 1) as row_number,
      @type := city_id as dummy
  from Hotel h
  JOIN Room r USING(hotel_id)
  INNER JOIN City c USING(city_id)
  order by city_name, r.room_price1 asc
) as x where x.row_number <=3;

this method return wrong sort for row_number < 6 (my init data)

---

there a ton of different blog post about this topic out there.

The most valuable for me, possibly was

Xaprb post (see section "Select the top N rows from each group" for a detailed explaination)

http://www.xaprb.com/blog/2006/12/07/how-to-select-the-firstleastmax-row-per-group-in-sql/

In your case that would mean. Modelling this SQL to your needs.

select type, variety, price
from fruits
where (
   select count(*) from fruits as f
   where f.type = fruits.type and f.price < fruits.price
) <= 2;

lets try it out

SELECT 
   r.price, r.room_id, h.hotel_id, h.city_id
FROM Hotel h INNER JOIN Room r ON
  h.hotel_id = r.hotel_id
WHERE (
   select count(*) from Room as rt
   where r.hotel_id = rt.hotel_id and r.price > rt.price
) <= 10;
ORDER BY h.city_id,h.hotel_id,r.price

as you can see you make an subquery to select the count of all rooms that have a price that is lower than the current root and see if the count is less than or equal to 10.

Observe in particular how i changed the

f.price < fruits.price

to

r.price > rt.price

because in the example of Xaprb he selects the N cheapest fruits

Hope that helps.

BTW Stackoverflow isn't for solving homework. You should learn something your self as well ;-)

---

I am assuming the table Room has a reference to the Hotel it belongs to

SELECT price, room_id, hotel_id, city_id
    FROM Hotel,Room 
    WHERE Hotel.hotel_id = Room.hotel_id
ORDER BY price GROUP BY city_id
LIMIT 10