錯誤代碼 1111：表示無效使用組 function

Question

我正在嘗試使用來自供應商的 select Suppliers.sid 離開加入 Catalog on Suppliers.sid=Catalog.sid where Catalog.cost>avg(Catalog.cost); 但我收到錯誤代碼 1111：不確定我的數據庫發生了什么。 我對 sql 非常好，只是在讓它工作時遇到問題。

    CREATE DATABASE CSC123Lab2;
USE CSC123Lab2;

CREATE TABLE Suppliers (
sid CHAR(20),
sname CHAR(20),
address CHAR(20),
pid CHAR(20)
);

CREATE TABLE Parts (
pid CHAR(20),
pname CHAR(20),
color CHAR(20),
sid CHAR(20)
);

CREATE TABLE Catalog (
sid CHAR(20),
pid CHAR(20),
cost REAL
);

ALTER TABLE Suppliers ADD PRIMARY KEY (sid);
ALTER TABLE Parts ADD PRIMARY KEY (pid);

INSERT INTO Suppliers (sid, sname, address) VALUES (1, "carvel", "1234 Main Street");
INSERT INTO Suppliers (sid, sname, address) VALUES (2, "coldstone", "5678 Pine Street");
INSERT INTO Suppliers (sid, sname, address) VALUES (3, "ralphs", "9876 Oak Street");
INSERT INTO Suppliers (sid, sname, address) VALUES (4, "Acme Widget", "4321 Maple Street");

INSERT INTO Parts (pid, pname, color) VALUES (1, "hammer", "red");
INSERT INTO Parts (pid, pname, color) VALUES (2, "screwdriver", "pink");
INSERT INTO Parts (pid, pname, color) VALUES (3, "nails", "green");
INSERT INTO Parts (pid, pname, color) VALUES (4, "screws", "teal");

INSERT INTO Catalog (sid, pid, cost) VALUES (1, 3, 32);
INSERT INTO Catalog (sid, pid, cost) VALUES (4, 1, 10);
INSERT INTO Catalog (sid, pid, cost) VALUES (4, 3, 48);
INSERT INTO Catalog (sid, pid, cost) VALUES (2, 4, 24);

select Suppliers.sid
from Suppliers 
left join Catalog on Suppliers.sid=Catalog.sid
where Catalog.cost>avg(Catalog.cost);


DROP TABLE Catalog;
DROP TABLE Parts;
DROP TABLE Suppliers;
DROP DATABASE CSC123Lab2;

Answer 1

使用 group by 並在下一級過濾

with cte  as 
(select Suppliers.sid,avg(Catalog.cost) as avg_val
from Suppliers 
left join Catalog on Suppliers.sid=Catalog.sid
group by Suppliers.sid
) select a.sid
from Catalog a join cte on a.sid=cte.sid
where a.cost>avg_val

Answer 2

我想你只想要一個 window function：

select s.sid
from Suppliers s join
     (select c.*, avg(c.cost) over () as avg_cost
      from Catalog c
     ) c
     on s.sid = c.sid
where c.cost > c.avg_cost;

但是請注意，您實際上並不需要left join 。 where子句將左連接變為inner join 。

然后，您根本不需要join ，因為sid在兩個表中：

select c.sid
from (select c.*, avg(c.cost) over () as avg_cost
      from Catalog c
     ) c
where c.cost > c.avg_cost;

如果您願意，可以使用子查詢來表達：

select c.sid
from Catalog c
where c.cost > (select avg(c2.cost) from catalog c2);