Question

我有一个表格，可以为节目的剧集存储一行。请注意，并非每个剧集都存储。

showID    |    viewerID   |   episodeNumber    |    episodeDescription
____________________________________________________________________
15        |    ID-123     |      1             |    Scary
____________________________________________________________________
15        |    ID-123     |      3             |    Funny

我已经构建了一个数据透视查询来为剧集编号和描述创建列。剧集数量是已知数量。

showID    |    viewerID    |    episode1    |    episode2    |    episode3 
__________________________________________________________________________
15         |   ID-123      |    Scary       |     funny       |

我的问题是，如果没有一集的记录，如本例中第一个表中第2集缺少行所示，则查询不会创建一个＆＃34;占位符＆＃34;而是将episode3的数据放入episode2列。我甚至不知道我的需要是否可能 - 但我确实需要以某种方式确保当剧集编号没有数据（没有行）时剧集编号与正确的描述相关联，而且我确实需要为表2中的每一集显示一列（所以在上面的表2中，episode2的字段应为空白）。

我的查询如下：

SELECT viewerID, showID, 
episodeDescription1, episodeDescription2, episodeDescription3
FROM
(
  SELECT viewerID, showID, col+cast(seq as varchar(10)) as col, value
  from 
  (
    SELECT viewerID, showID, episodeNumber, episodeDescription,
           row_number() over(partition by showID
                        order by episodeNumber) seq
    FROM tbl_showEpisodes
  ) d
  CROSS APPLY
  (
   SELECT 'episodeNumber', cast(episodeNumber as varchar(20)) 
   UNION ALL
   SELECT 'episodeDescription', episodeDescription
  ) c (col, value)
) src
pivot
(
  MAX(value)
  for col in (
    episodeNumber1, episodeDescription1, episodeNumber2, 
    episodeDescription2, episodeNumber3, episodeDescription3)
  ) piv;

我希望这很清楚，并提前感谢任何帮助！使用SQL Server 2012.

Answer 1

你可以这样做：

create table tbl_showEpisodes (showID int, viewerID varchar(30), episodeNumber int, episodeDescription varchar(30));
insert into tbl_showEpisodes values
(15, 'ID-123', 1, 'Scary'),
(15, 'ID-123', 3, 'Funny');
GO

SELECT showID, viewerID, [Episode 1], [Episode 2], [Episode 3]
FROM (SELECT showID, 
             viewerID, 
             CONCAT('Episode ', episodeNumber) epNumber, 
             episodeDescription 
      FROM tbl_showEpisodes) src
PIVOT (MAX(episodeDescription) 
       FOR epNumber 
       IN ([Episode 1],[Episode 2],[Episode 3])) pvt
GO

showID | viewerID | Episode 1 | Episode 2 | Episode 3
-----: | :------- | :-------- | :-------- | :--------
    15 | ID-123   | Scary     | null      | Funny

dbfiddle here

SQL：使用占位符行进行透视

1 个答案: