SQL Server 透视文本值

Question

我正在练习枢轴。我有一个包含姓名和职业数据的表格（生成示例的代码如下）。我需要将职业放在列中，并在每个职业、名称下方堆叠。 Null 很好。

我看过很多帖子，this 一个是我需要的，但是一旦适应了我的数据，它只拉取 1 行数据。

这里是生成示例和实际数据透视代码的代码。我该如何纠正这个错误？

declare @occupations table (ename nvarchar(10), occupation nvarchar(10) )
insert @occupations
values 
('John','Doctor'),('Mary','Doctor'),('Jack','Doctor'),('Anna','Doctor'),
('Jim','Singer'),('Kate','Singer'),('Helen','Actor'),('Paco','Singer')

SELECT [Doctor] Doctor, [Singer] Singer, [Actor] Actor
FROM @occupations
PIVOT
  (min(ename)  
    FOR occupation IN ([Doctor],[Singer], [Actor])  
  )AS pp

我得到这个结果：

 Doctor Singer Actor
 Anna   Jim    Helen    

应该有更多行来包含所有名称。 IE。例如，您可以在插入语句中看到有 4 位医生。所以应该是带有一些空值的 8 行。

Answer 1

如果您想要所有行，只需添加一个唯一键，如 row_number()

示例

SELECT [Doctor] Doctor, [Singer] Singer, [Actor] Actor
FROM (
       Select *
             ,rn=row_number() over(order by ename) 
        from @occupations
     ) src
PIVOT
  (min(ename)  
    FOR occupation IN ([Doctor],[Singer], [Actor])  
  )AS pp

退货

Doctor  Singer  Actor
Anna    NULL    NULL
NULL    NULL    Helen
Jack    NULL    NULL
NULL    Jim     NULL
John    NULL    NULL
NULL    Kate    NULL
Mary    NULL    NULL
NULL    Paco    NULL

Answer 2

您还可以使其动态自动选择所有列

create table #occupations  (ename nvarchar(10), occupation nvarchar(10) )
insert   #occupations
values 
('John','Doctor'),('Mary','Doctor'),('Jack','Doctor'),('Anna','Doctor'),
('Jim','Singer'),('Kate','Singer'),('Helen','Actor'),('Paco','Singer') ;


DECLARE @cols AS NVARCHAR(MAX),
        @query  AS NVARCHAR(MAX);

SET @cols = STUFF((SELECT distinct ',' + QUOTENAME(occupation) 
            FROM #occupations 
            FOR XML PATH(''), TYPE
            ).value('.', 'NVARCHAR(MAX)') 
        ,1,1,'');

set @query = 'SELECT * FROM #occupations
                    PIVOT
                        (min(ename)  
                FOR occupation IN ('+ @cols +')  
              )AS pp'
exec (@query) 
drop table #occupations

Answer 3

SQL Fiddle Execution

MS SQL Server 2017 架构设置：

create table occupations (ename nvarchar(10), occupation nvarchar(10) )
insert into occupations(ename,occupation)values ('John','Doctor'),('Mary','Doctor'),('Jack','Doctor'),('Anna','Doctor'),
('Jim','Singer'),('Kate','Singer'),('Helen','Actor'),('Paco','Singer')

**查询**：

WITH CTE AS (
  SELECT *,
  (case when occupation like 'singer' then ename end) AS Singer,
    (case when occupation like 'doctor' then ename end) AS Doctor,
    (case when occupation like 'Actor' then ename end) AS Actor,
  ROW_Number() OVER (PARTITION BY occupation Order By ename) as rn
  from occupations
  group by ename,occupation)

  select max(Singer) AS Singer,max(Doctor) AS Doctor, max(Actor) AS Actor
  from cte
  where rn=1 

Results：

| Singer | Doctor | Actor |
|--------|--------|-------|
|    Jim |   Anna | Helen |