我希望将所有演出票数与所有演出的总收入相比较。我已设法计算预订的票数,但其完整查询无效并给我错误。
示例结果将遵循以下示例:
`tickets booked = 9
total revenue of the performance = xxx$
total revenue of all the performances = xxx$
我正在使用的查询是:
SET @performancerevenue = '50 Shades of Grey';
select distinct P.perName as Performance,
(select COUNT(ticketBooked) from bookings as B
inner join performances as P on
B.perID = p.perID where perName = @performancerevenue) as
[Number of Ticket Booked ],
CAST(AVG((T.ticketPrice*left(B.totalCost,1))) over() as decimal(6,2)) as[Total Revenue from the performance (£)],
CAST(AVG(convert(decimal(5,2),(left(PY.totalAmount,1)))) over() as decimal(5,2)) as [Total revenue]
from performances as P
inner join tickets as T
on T.perID = P.perID
inner join bookings as B
on T.bookingID = B.bookingID
inner join payments as PY
where P.perName = @performancerevenue
表格和样本数据如下:
CREATE TABLE bookings(
bookingID int IDENTITY (1,1) NOT NULL,
customerID int ,
clerkID varchar (20) ,
perID int,
ticketBooked int,
bookingDate Date,
bookingTime time,
bookingType varchar(20),
totalCost decimal(5,2),
PRIMARY KEY (bookingID),
FOREIGN KEY (customerID) REFERENCES customers,
FOREIGN Key (clerkID) REFERENCES clerks,);
insert into bookings values (16, 'clerk04', 1, 2, '2015-02-24', '12:00', 'Normal', '20');
insert into bookings values (2, 'clerk02', 20,3, '2015-02-25', '15:00', 'Advance', '34');
insert into bookings values (10, 'clerk04', 3,5, '2015-02-17', '18:00', 'Advance', '75');
insert into bookings values (5, 'clerk01', 19,2, '2015-01-24', '21:00', 'Advance', '30');
insert into bookings values (13, 'clerk01', 4,1, '2015-02-05', '12:00', 'Normal', '10');
insert into bookings values (20, 'clerk04', 5,1,'2015-01-26', '15:00', 'Normal', '12');
insert into bookings values (4, 'clerk03', 6, 3,'2015-02-09', '18:00', 'Advance', '30');
insert into bookings values (1, 'clerk03', 2, 1,'2015-01-03', '21:00', 'Advance', '15');
insert into bookings values (17, 'clerk04', 3, 1,'2015-02-09', '12:00', 'Normal', '12');
CREATE TABLE performances(
perID int IDENTITY (1,1)NOT NULL,
perName varchar (50),
perTime Time,
perDate Date,
perType varchar (20),
duration varchar (10),
screenNo int,
rating varchar (10),
location varchar (10),
PRIMARY KEY (perID),);
insert into performances values ('50 Shades of Grey', '12:00', '2015-03-08', 'Movie', '2hrs', '3', 'PG', 'Chelmsford');
insert into performances values ('50 Shades of Grey', '15:00', '2015-02-20', 'Movie', '2hrs', '2', '18', 'Chelmsford');
insert into performances values ('50 Shades of Grey', '18:00', '2015-02-26', 'Movie', '2hrs', '3', '18', 'Chelmsford');
insert into performances values ('50 Shades of Grey', '21:00', '2015-03-28', 'Movie', '2hrs', '2', 'PG', 'Chelmsford');
insert into performances values ('Paddington', '12:00', '2015-03-26', 'Movie', '2:30hrs', '2', '18', 'Chelmsford');
insert into performances values ('Paddington', '15:00', '2015-03-03', 'Movie', '2:30hrs', '3', 'PG', 'Chelmsford');
insert into performances values ('Paddington', '18:00', '2015-02-02', 'Movie', '2:3hrs', '3', 'PG', 'Chelmsford');
insert into performances values ('Paddington', '21:00', '2015-03-31', 'Movie', '2:30hrs', '1', '18', 'Chelmsford');
insert into performances values ('Shakespeare', '12:00', '2015-02-25', 'Movie', '3hrs', '2', '18', 'Chelmsford');
insert into performances values ('Shakespeare', '15:00', '2015-02-06', 'Movie', '3hrs', '3', 'PG', 'Chelmsford');
insert into performances values ('Shakespeare', '18:00', '2015-02-13', 'Movie', '3hrs', '3', '18', 'Chelmsford');
insert into performances values ('Shakespeare', '21:00', '2015-01-09', 'Movie', '3hrs', '3', 'PG', 'Chelmsford');
insert into performances values ('Jupiter Ascending', '12:00', '2015-01-11', 'Movie', '2:3hrs', '1', '18', 'Chelmsford');
insert into performances values ('Jupiter Ascending', '15:00', '2015-01-19', 'Movie', '2:30hrs', '2', '18', 'Chelmsford');
insert into performances values ('Jupiter Ascending', '18:00', '2015-02-06', 'Movie', '2:3hrs', '3', 'PG', 'Chelmsford');
insert into performances values ('Jupiter Ascending', '21:00', '2015-01-02', 'Movie', '2:3hrs', '2', 'PG', 'Chelmsford');
insert into performances values ('50 Shades of Grey', '12:00', '2015-01-06', 'Movie', '2hrs', '1', '18', 'Chelmsford');
insert into performances values ('50 Shades of Grey', '15:00', '2015-03-28', 'Movie', '2hrs', '3', 'PG', 'Chelmsford');
insert into performances values ('50 Shades of Grey', '18:00', '2015-02-09', 'Movie', '2hrs', '2', 'PG', 'Chelmsford');
insert into performances values ('50 Shades of Grey', '21:00', '2015-02-06', 'Movie', '2hrs', '2', '18', 'Chelmsford');
CREATE TABLE payments(
paymentID int IDENTITY (1,1) not null,
bookingID int,
totalAmount decimal (5,2),
paymentType varchar (20),
paymentDate Date,
PRIMARY KEY (paymentID),
FOREIGN KEY (bookingID) REFERENCES bookings,);
insert into payments values (1, 20, 'Cash', '2015-01-20');
insert into payments values (10, 20, 'Cash', '2015-01-17');
insert into payments values (12, 12, 'Credit/Debit Card', '2015-03-31');
insert into payments values (5, 10, 'Cash', '2015-01-08');
insert into payments values (18, 12, 'Cash', '2015-03-22');
insert into payments values (3, 75, 'Credit/Debit Card', '2015-02-21');
insert into payments values (2, 34, 'Credit/Debit Card', '2015-03-26');
insert into payments values (9, 12, 'Cash', '2015-03-26');
insert into payments values (13, 20, 'Credit/Debit Card', '2015-01-23');
insert into payments values (16, 20, 'Credit/Debit Card', '2015-03-08');
insert into payments values (11, 45, 'Credit/Debit Card', '2015-02-21');
insert into payments values (4, 30, 'Credit/Debit Card', '2015-02-08');
insert into payments values (7, 30, 'Cash', '2015-01-20');
insert into payments values (20, 30, 'Cash', '2015-02-26');
insert into payments values (14, 20, 'Credit/Debit Card', '2015-03-17');
insert into payments values (19, 45, 'Cash', '2015-01-10');
insert into payments values (6, 12, 'Credit/Debit Card', '2015-02-02');
insert into payments values (20, 30, 'Cash', '2015-02-01');
insert into payments values (17, 34, 'Credit/Debit Card', '2015-03-27');
insert into payments values (15, 15, 'Credit/Debit Card', '2015-02-14');
我一直在努力让结果在过去4小时内正常工作,但它只是给我错误我会非常感谢你的帮助:(
感谢
答案 0 :(得分:0)
在我看来,您可以通过以下方式获得所需的结果:
SELECT DISTINCT
P.perName as [Perf],
SUM(B.ticketBooked) OVER(PARTITION BY P.perName) as [BookedTickets],
SUM(B.totalCost) OVER(PARTITION BY P.perName) as [PerfRevenue],
SUM(B.totalCost) OVER() as [TotalRevenue]
from bookings as B
inner join performances as P on B.perID = p.perID
此查询的结果如下:
------------------------------------------------------------
Perf BookedTickets PerfRevenue TotalRevenue
------------------+---------------+------------+------------
50 Shades of Grey 15 196.00 238.00
Paddington 4 42.00 238.00
您可以为此表添加任何外部查询以进行其他过滤和计算。