问题

Note: This post is slightly change with previous post.

I have the following table with the details as shown below in the example.

Example:

Table: test

create table test
(
 cola varchar(10),
 colb varchar(10),
 colc varchar(10)
);

Insertion:

insert into test values('101','1234','A1');
insert into test values('101','4321','A2');
insert into test values('201','5678','A3');
insert into test values('301','8765','A4');
insert into test values('401','9877','A1');
 insert into test values('101','9997','A6');
 insert into test values('201','2277','A1');
 insert into test values('201','1577','A5'); 

Note: Now I want to show only that records in which cola belongs to colc's values. If the user pass the colc values as a parameter to stored procedure then it has to match the exact value of colc belongs to which cola value.

Expected Result:

If the user pass A1,A2,A6 then the result should be:

cola   A1   A2   A6
--------------------
101    1      1     1

Note: In the above result the record 101 appears because it belongs to A1,A2,A6 not other values. 201 not appear because that belongs to A1,A3 and A5 also.

If the user pass A1 then the result should be:

cola   A1 
--------
401    1  

Note: In the above result the record 401 appears because it belongs to A1.

My Attempt:

Pivot Query:

DECLARE @Stuff varchar(max) = 'A1'
DECLARE @Sql varchar(max)

SET @Sql = 'SELECT cola,' +@Stuff+ '
            from
            (
                select cola,colc
                from test 
                where colc in(''A1'')
                group by cola,colc
                having count(distinct colc) = 1

            )p
            PIVOT
            (
                COUNT(colc)
                FOR colc IN ('+@Stuff+')
            )AS pvt'

PRINT(@Sql)
EXEC(@Sql)      

Getting Result:

cola    A1  
----------
101     1
401     1

You can go here for practicle:

SQL Fiddle:

http://sqlfiddle.com/#!3/9b1fd/4

回答1:

You could use this :

DECLARE @Stuff varchar(max) = 'A1'
        DECLARE @Sql varchar(max)
    DECLARE @totalparam int = 1
        SET @Sql = 'SELECT cola,' +@Stuff+ '
                    from
                    (
                        select cola,colc
                        from test a
                        where colc in(''A1'')
                        group by cola,colc
                        having (select count(colc) from test b where b.cola = a.cola )  = @totalparam

                    )p
                PIVOT
                (
                    COUNT(colc)
                    FOR colc IN ('+@Stuff+')
                )AS pvt'

    PRINT(@Sql)
    EXEC(@Sql)  

回答2:

Try this.

DECLARE @Stuff VARCHAR(max) = 'A1,A2'--'A1'
DECLARE @Sql NVARCHAR(max),
        @cnt INT

SELECT @cnt = Count(1)
FROM   (SELECT Cast ('<M>' + Replace(@Stuff, ',', '</M><M>') + '</M>' AS XML) AS Data) AS A
       CROSS APPLY Data.nodes ('/M') AS Split(a)

SET @Sql = 'SELECT cola,' + @Stuff+ ' from
                (SELECT a.cola, a.colc
                 FROM   test a
                 JOIN (SELECT cola
                 FROM   test
                 GROUP  BY cola
                 HAVING Count(DISTINCT colc) = '
           + CONVERT(VARCHAR(30), @cnt)+ ') B
                 ON a.cola = b.cola
                 where colc in (select rtrim(ltrim(Split.a.value(''.'', ''VARCHAR(100)'')))
FROM   (SELECT Cast (''<M>'' + Replace('''
           + @Stuff + ''', '','', ''</M><M>'') + ''</M>'' AS XML) AS Data) AS A
       CROSS APPLY Data.nodes (''/M'') AS Split(a) )
                )p
                PIVOT
                (COUNT(colc) FOR colc IN (' + @Stuff+ ')
                )AS pvt'

EXEC Sp_executesql @Sql 

OUTPUT :

(For Input string 'A1,A2')

+-----+----+----+
|cola | A1 | A2 |
+-----+----+----+
|101  | 1  | 1  |
+-----+----+----+

(For Input string 'A1')

+-----+----+
|cola | A1 |
+-----+----+
|401  | 1  |
+-----+----+

来源：https://stackoverflow.com/questions/27077468/sql-server-2008-r2-dynamic-query-for-pivot-table-with-where-and-having-clause