从表中排除某些值的 SQL 查询

Question

Hi everyone I am in the process of building a room allocating system that checks if a room is available on a given date. There are about 20 unique room ids and I've just included as snippet of several above.

So for example if the user types in the date of 2012-01-01 I would want the room_id of 1 to be excluded from the results because as seen above it has been already booked for that date.

In a sense any match found between the user's in-putted date should cause the entire list of corresponding room_id's to be excluded from the list.

I used the below SQL query:

SELECT DISTINCT room_id
FROM room_booking
WHERE date_booked<>'$date'

Unfortunately it doesn't work because although it excludes the result for a single row, it includes the rest of the results for a given room_id ?

Answer 1

You could use a NOT EXISTS

SELECT DISTINCT room_id
  FROM room_booking rb1
 WHERE NOT EXISTS( SELECT 1
                     FROM room_booking rb2
                    WHERE rb2.date_booked = <<date>>
                      AND rb2.room_id     = rb1.room_id )

or a NOT IN

SELECT DISTINCT room_id
  FROM room_booking
 WHERE room_id NOT IN (SELECT room_id
                         FROM room_booking rb2
                        WHERE rb2.date_booked = <<date>>)

Answer 2

Not tried this, but here goes

SELECT DISTINCT room_id
FROM room_booking
WHERE room_id NOT IN 
    (SELECT DISTINCT room_id
     FROM room_booking
     WHERE date_booked = '$date'
    )

Answer 3

Considering you have a column called RoomID I'm going to assume that there is a Room table with a corresponding column RoomID - this is probably a much better place to look for all rooms as you can't guarantee every room has a booking, and so can't rely on the booking table alone.

Then you can look up with a sub-query which rooms are booked on a date, then return all the RoomIds which aren't in that list.

select RoomID from Room
where RoomID not in
(select RoomID from room_booking
where date = @date)

Answer 4

The above should work however may potentially become slow if a lot of rooms are added. You may want to consider a left outer join. It will typically be faster as the above will be performing sub queries ie For every room, requery the database to check if it has a value of this date which results in 40 * 40 checks each time this query is executed and will increase with the number of current bookings.

The below should be more efficient and should be valid on quite a few database platforms. Of course, this depends on how much the count of values may change if there are any string performance related items to consider...

SELECT DISTINCT room_id
           FROM room_booking rb1
LEFT OUTER JOIN room_booking datecheck
             ON rb1.room_id = datecheck.room_id
            AND datecheck.date_booked  =  @date
          WHERE datecheck.room_id is null