如何使用“Stuff and 'For Xml Path'”来合并表中的行

如何解决如何使用“Stuff and 'For Xml Path'”来合并表中的行

请帮助我在表格中获取以逗号分隔的联合行和帐户列表。我不太明白如何使用“Stuff and 'For Xml Path'”。

这是我的查询：

CREATE TABLE invoices
(
invoice VARCHAR(20) NOT NULL,quantity INT NOT NULL,price INT NOT NULL,summ INT NOT NULL,account INT NOT NULL,);
INSERT invoices(invoice,quantity,price,summ,account) 
VALUES ('ty20210110',2,100,200,1001);
INSERT invoices(invoice,3,300,1002);
INSERT invoices(invoice,1,250,120,240,4,400,account) 
VALUES ('ty20210114',5,80,1003);
INSERT invoices(invoice,500,1004);


SELECT invoices.invoice,invoices.summ,accounts = STUFF(
             (SELECT disTINCT ',' + Convert(varchar,invoices.account,60) 
              FROM invoices
              FOR XML PATH ('')),'')
FROM invoices
GROUP BY invoices.invoice,invoices.summ

这是我得到的结果：

发票	summ	账户
ty20210110	200	1001,1002,1003,1004
ty20210110	240	1001,1004
ty20210110	250	1001,1004
ty20210110	300	1001,1004
ty20210110	400	1001,1004
ty20210114	300	1001,1004
ty20210114	400	1001,1004
ty20210114	500	1001,1004

这是我需要得到的结果：

发票	summ	账户
ty20210110	1390	1001,1002
ty20210114	1200	1003,1004

所以实际上我需要获取 2 张不同发票的金额，并用逗号指定涉及这些发票的帐户。

这里的 dbfiddle 也有这些东西：https://dbfiddle.uk/?rdbms=sqlserver_2019&fiddle=7a5de9e680693b5e70ea68cecebef6cc

提前谢谢你们。

解决方法

如果要求和，请不要按 summ 分组。在其上使用 sum()。并关联子查询。否则，您将获得所有帐户。

SELECT i1.invoice,sum(i1.summ) summ,stuff((SELECT DISTINCT
                     concat(',',i2.account)
                     FROM invoices i2
                     WHERE i2.invoice = i1.invoice
                     FOR XML PATH ('')),1,'') accounts
       FROM invoices i1
       GROUP BY i1.invoice;

,

对于 SQL Server 2017 及更高版本。

利用 fgets 函数。唯一的细微差别是我们需要在子查询中选择 STRING_AGG() account 值。

SQL

DISTINCT

输出

-- DDL and sample data population,start
DECLARE @invoices TABLE 
(
    invoice VARCHAR(20) NOT NULL,quantity INT NOT NULL,price INT NOT NULL,summ INT NOT NULL,account INT NOT NULL
);
INSERT @invoices(invoice,quantity,price,summ,account) VALUES 
('ty20210110',2,100,200,1001),('ty20210110',3,300,1002),250,120,240,4,400,('ty20210114',5,80,1003),500,1004);
-- DDL and sample data population,end

SELECT i1.invoice,SUM(i1.summ) AS summ,(
       SELECT STRING_AGG(account,') FROM 
       (
        (SELECT DISTINCT account FROM @invoices AS i2 WHERE i2.invoice = i1.invoice)
       ) AS x) AS accounts
FROM @invoices AS i1
GROUP BY i1.invoice;