【数据库】使用Sql Server将分组后指定字段的行数据转为一个字段显示，并且以逗号隔开每个值，收藏不迷路

大家好，我是全栈小5，欢迎来到《小5讲堂》。

这是《Sql Server》系列文章，每篇文章将以博主理解的角度展开讲解。

温馨提示：博主能力有限，理解水平有限，若有不对之处望指正！

前言
示例
查询小技巧
- [1. 使用 `WITH TIES` 获取并列结果（分页查询时特别有用）](#1. 使用 WITH TIES 获取并列结果（分页查询时特别有用）)
- [2. 使用 `OUTPUT` 子句捕获DML操作结果（避免二次查询）](#2. 使用 OUTPUT 子句捕获DML操作结果（避免二次查询）)
- [3. 使用 `CROSS APPLY` 优化复杂查询（替代JOIN的更好选择）](#3. 使用 CROSS APPLY 优化复杂查询（替代JOIN的更好选择）)
文章推荐

前言

很久没有写报表了，今天尝试写下报表就遇到了一个常见但是太久了有点忘了的知识点。

那就是如何让指定字段的行数据转为一个字段显示并以逗号的形式隔开，一起来探讨下！

示例

STRING_AGG 是SQL Server 2017引入的新函数，更简洁高效

数据集

模拟表格数据，如下

csharp 复制代码

select '张三' as NameText,'语文' as SubjectText
union all
select '张三' as NameText,'数学' as SubjectText
union all
select '张三' as NameText,'英语' as SubjectText
union all
select '张三' as NameText,'物理' as SubjectText
union all
select '张三' as NameText,'化学' as SubjectText
union all
select '张三' as NameText,'生物' as SubjectText
union all
select '张三' as NameText,'历史' as SubjectText
union all
select '张三' as NameText,'政治' as SubjectText
union all
select '张三' as NameText,'地理' as SubjectText
union all
select '张三' as NameText,'体育' as SubjectText
union all
select '张三' as NameText,'音乐' as SubjectText
union all
select '张三' as NameText,'美术' as SubjectText
union all
select '李四' as NameText,'体育' as SubjectText
union all
select '李四' as NameText,'音乐' as SubjectText
union all
select '李四' as NameText,'美术' as SubjectText

数据分组

以姓名为分组，增加多一个科目字段，并且科目名称以逗号形式隔开，效果如下：

string_agg(cast(字段 as varchar), ',') as 自定义名称

csharp 复制代码

select t.NameText
,string_agg(cast(t.SubjectText as varchar), ',') as ALLSubject
from(
	select '张三' as NameText,'语文' as SubjectText
	union all
	select '张三' as NameText,'数学' as SubjectText
	union all
	select '张三' as NameText,'英语' as SubjectText
	union all
	select '张三' as NameText,'物理' as SubjectText
	union all
	select '张三' as NameText,'化学' as SubjectText
	union all
	select '张三' as NameText,'生物' as SubjectText
	union all
	select '张三' as NameText,'历史' as SubjectText
	union all
	select '张三' as NameText,'政治' as SubjectText
	union all
	select '张三' as NameText,'地理' as SubjectText
	union all
	select '张三' as NameText,'体育' as SubjectText
	union all
	select '张三' as NameText,'音乐' as SubjectText
	union all
	select '张三' as NameText,'美术' as SubjectText
	union all
	select '李四' as NameText,'体育' as SubjectText
	union all
	select '李四' as NameText,'音乐' as SubjectText
	union all
	select '李四' as NameText,'美术' as SubjectText
) t
group by t.NameText

增加排序

通过输出结果可能会注意到，以逗号隔开的文本不是按顺序，和表格本身排序不一致。

假设根据id进行升序排序输出，确保一致，如下：

string_agg(cast(字段 as varchar), ',') within group (order by 排序字段 asc) as 自定义名称

csharp 复制代码

select t.NameText
,string_agg(cast(t.SubjectText as varchar), ',') within group (order by t.id asc) as ALLSubject
from(
	select 1 as id,'张三' as NameText,'语文' as SubjectText
	union all
	select 2 as id,'张三' as NameText,'数学' as SubjectText
	union all
	select 3 as id,'张三' as NameText,'英语' as SubjectText
	union all
	select 4 as id,'张三' as NameText,'物理' as SubjectText
	union all
	select 5 as id,'张三' as NameText,'化学' as SubjectText
	union all
	select 6 as id,'张三' as NameText,'生物' as SubjectText
	union all
	select 7 as id,'张三' as NameText,'历史' as SubjectText
	union all
	select 8 as id,'张三' as NameText,'政治' as SubjectText
	union all
	select 9 as id,'张三' as NameText,'地理' as SubjectText
	union all
	select 10 as id,'张三' as NameText,'体育' as SubjectText
	union all
	select 11 as id,'张三' as NameText,'音乐' as SubjectText
	union all
	select 12 as id,'张三' as NameText,'美术' as SubjectText
	union all
	select 13 as id,'李四' as NameText,'体育' as SubjectText
	union all
	select 14 as id,'李四' as NameText,'音乐' as SubjectText
	union all
	select 15 as id,'李四' as NameText,'美术' as SubjectText
) t
group by t.NameText

查询小技巧

以下是三个能提高你SQL Server查询效率和便利性的实用技巧：

1. 使用 `WITH TIES` 获取并列结果（分页查询时特别有用）

sql 复制代码

-- 获取前10条记录，包括与第10条记录值相同的所有记录
SELECT TOP 10 WITH TIES column1, column2
FROM table_name
ORDER BY column1 DESC;

应用场景：当你想获取排名靠前的记录，但不想因为TOP N的限制而遗漏与第N条记录值相同的其他记录。

2. 使用 `OUTPUT` 子句捕获DML操作结果（避免二次查询）

sql 复制代码

-- 更新数据同时返回被更新的记录
UPDATE table_name
SET column1 = 'new_value'
OUTPUT inserted.*  -- 返回更新后的数据
WHERE condition;

-- 删除数据同时返回被删除的记录
DELETE FROM table_name
OUTPUT deleted.*  -- 返回被删除的数据
WHERE condition;

优势：减少数据库往返次数，提高效率，特别适合需要记录变更的场景。

3. 使用 `CROSS APPLY` 优化复杂查询（替代JOIN的更好选择）

sql 复制代码

-- 获取每个客户的最新订单
SELECT c.CustomerID, c.CustomerName, o.OrderDate, o.Amount
FROM Customers c
CROSS APPLY (
    SELECT TOP 1 OrderDate, Amount
    FROM Orders
    WHERE CustomerID = c.CustomerID
    ORDER BY OrderDate DESC
) o;

优势：