好的,这是我得出结论我迷路的情况之一。
我有一个看起来像这样的表:
我写了以下代码:
WITH CTE AS(
SELECT
USER_NBR
,CASE WHEN MOVIE_TYPE = 'Action' THEN MOVIE_NAME END [ACTION]
,CASE WHEN MOVIE_TYPE = 'Drama' THEN MOVIE_NAME END DRAMA
,CASE WHEN MOVIE_TYPE = 'Romance' THEN MOVIE_NAME END ROMANCE
,ROW_NUMBER() OVER (PARTITION BY USER_NBR ORDER BY USER_NBR) AS OCCURANCE
FROM dbo.MOVIE)
SELECT * FROM CTE WHERE OCCURANCE = 1
我得到了这个结果:
我不明白为什么我在查询结果中获得Null?
我做错了什么?正如你所看到的,我正在对电影进行分类。
感谢您查看此内容
答案 0 :(得分:3)
也许这种方法对你有用。通常,当您将行值转换为列(例如示例中的电影类型)时,它就是一个数据透视功能(参考http://msdn.microsoft.com/en-us/library/ms177410.aspx)
select * from
(select user_nbr, movie_name, movie_type from movies) as movies
pivot
(max(movie_name)
for movie_type in ([Action], [Drama], [Romance])
)
as PivotTable
结果:
user_nbr Action Drama Romance
101 Casino Royale Pretty Woman Love Actually
102 Casino Royale Pretty Woman Love Actually
答案 1 :(得分:0)
请注意,如果您尝试获取此类输出,则必须使用group by子句。
如果您想获得所有类别的价值,那么您可以尝试以下查询..
WITH CTE AS(
SELECT
USER_NBR
,CASE WHEN MOVIE_TYPE = 'Action' THEN MOVIE_NAME END [ACTION]
,CASE WHEN MOVIE_TYPE = 'Drama' THEN MOVIE_NAME END DRAMA
,CASE WHEN MOVIE_TYPE = 'Romance' THEN MOVIE_NAME END ROMANCE
,count(*) AS OCCURANCE
FROM dbo.MOVIE
group by USER_NBR
)
SELECT * FROM CTE
我认为这就是你要做的事情。