For this scenario,does a single self-join exist?

Question

I have a table ALPHA with 2 fields GroupId,Member:

GroupId | Member;
A1----------A;
A1----------B; 
A1----------C;
A2----------A;
A2----------B;
A3----------A;
A3----------D;
A3----------E;

Objective: Given the input of - A,B,C - I have to query the table to find if a GroupId exists for this exact set of members. So, this is what I plan to do:

1. Query the table for all GroupIds whose count is 3 (since my inpt is A,B,C ..I knw its 3)
2. This will give me A1,A3. Now, query this set for exact matching Member values..which will give me A1.

I plan to write a Stored Procedure and would achieve the objective somehow. But, my question can this be achieved in a single query...a single self-join perhaps.

Clarification: The set of (A,B,C) is unique to A1. And if give an input of (A,B,C,D) the query should NOT return A1.

Answer 1

Answers given so far assume that the Member field is unique for any given GroupID. In work I have done this isn't the case. And also if the group has what you're looking for, plus some extra, you need to exlucde that group.

SELECT
   [Alpha].GroupID
FROM
   [Alpha]
GROUP BY
   [Alpha].GroupID
HAVING
       SUM(CASE WHEN [alpha].Member IN ('A','B','C') THEN 1 ELSE 0 END) = 3
   AND MIN(CASE WHEN [alpha].Member IN ('A','B','C') THEN 1 ELSE 0 END) = 1

You can also replace the IN clause with a join on to a table holding the members you are searching for...

SELECT
   [Alpha].GroupID
FROM
   [Alpha]
LEFT JOIN
   [Search]
       ON [Search].Member
GROUP BY
   [Alpha].GroupID
HAVING
       SUM(CASE WHEN [alpha].Member = [search].Member THEN 1 ELSE 0 END) = (SELECT COUNT(*) FROM [search])
   AND MIN(CASE WHEN [alpha].Member = [search].Member THEN 1 ELSE 0 END) = 1

Answer 2

SELECT GroupID
  FROM ALPHA
 WHERE Member IN ('A', 'B', 'C')
 GROUP BY GroupID
HAVING COUNT(*) = 3

This relies on you writing out the list of members in the IN clause and setting the number of (distinct) entries in the member list in the HAVING clause. If you can't generate the SQL thusly, then you have to work harder.

---

As noted in an early comment, this also relies on the interpretation that you want the groups where all three of A, B, C (and possibly some others) are members of the group. One way, not necessarily the best way, of getting 'where the group contains exactly three people, namely A, B, C', is to use:

SELECT GroupID
  FROM ALPHA A1
 WHERE Member IN ('A', 'B', 'C')
   AND 3 = (SELECT COUNT(*) FROM ALPHA A2 WHERE A2.GroupID = A1.GroupID)
 GROUP BY GroupID
HAVING COUNT(*) = 3

This explicitly checks that the total number of people in the group is 3 and that the members are A, B, and C (assuming that there is a unique constraint on Alpha(GroupID, Member) so that a member can't be listed twice as belonging to the same group).

Answer 3

SELECT DISTINCT aa.GroupId
FROM Alpha aa
JOIN Alpha ab ON (aa.GroupId = ab.GroupId)
JOIN Alpha ac ON (aa.GroupId = ac.GroupId)
LEFT OUTER JOIN Alpha ax ON (aa.GroupId = ax.GroupId AND ax.Member NOT IN ('A', 'B', 'C')
WHERE aa.Member = 'A' AND ab.Member = 'B' AND ac.Member = 'C'
 AND ax.GroupId IS NULL;

There's also solutions involving GROUP BY but I find the JOIN solution often has better performance. I usually work in MySQL, and I understand MS SQL Server is better at grouping queries. So try both solutions and see what works best for the brand of RDBMS you use.

Answer 4

try this:

declare @YourTable table (GroupID char(2),Member char(1))

insert into @YourTable values ('A1','A')
insert into @YourTable values ('A1','B')
insert into @YourTable values ('A1','C')
insert into @YourTable values ('A2','A')
insert into @YourTable values ('A2','B')
insert into @YourTable values ('A3','A')
insert into @YourTable values ('A3','D')
insert into @YourTable values ('A3','E')
insert into @YourTable values ('A5','A')
insert into @YourTable values ('A5','B')
insert into @YourTable values ('A5','C')
insert into @YourTable values ('A5','D')
SELECT t1.GroupID
    FROM @YourTable t1
        LEFT OUTER JOIN @YourTable t2 ON t1.GroupID=t2.GroupID AND t2.Member NOT IN ('A', 'B', 'C') 
    WHERE t1.Member IN ('A', 'B', 'C') 
        AND t2.GroupID IS NULL
    GROUP BY t1.GroupID
    HAVING COUNT(*) = 3

OUTPUT:

GroupID
-------
A1

(1 row(s) affected)

Here is a complete solution:

Before you use my function, you need to set up a "helper" table, you only need to do this one time per database:

CREATE TABLE Numbers
(Number int  NOT NULL,
    CONSTRAINT PK_Numbers PRIMARY KEY CLUSTERED (Number ASC)WITH (PAD_INDEX  = OFF, STATISTICS_NORECOMPUTE  = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS  = ON, ALLOW_PAGE_LOCKS  = ON) ON [PRIMARY]
) ON [PRIMARY]
DECLARE @x int
SET @x=0
WHILE @x<8000
BEGIN
    SET @x=@x+1
    INSERT INTO Numbers VALUES (@x)
END

use this function to split your string, which does not loop and is very fast:

CREATE FUNCTION [dbo].[FN_ListToTable]
(
     @SplitOn              char(1)              --REQUIRED, the character to split the @List string on
    ,@List                 varchar(8000)        --REQUIRED, the list to split apart
)
RETURNS
@ParsedList table
(
    ListValue varchar(500)
)
AS
BEGIN

/**
Takes the given @List string and splits it apart based on the given @SplitOn character.
A table is returned, one row per split item, with a column name "ListValue".
This function workes for fixed or variable lenght items.
Empty and null items will not be included in the results set.


Returns a table, one row per item in the list, with a column name "ListValue"

EXAMPLE:
----------
SELECT * FROM dbo.FN_ListToTable(',','1,12,123,1234,54321,6,A,*,|||,,,,B')

    returns:
        ListValue  
        -----------
        1
        12
        123
        1234
        54321
        6
        A
        *
        |||
        B

        (10 row(s) affected)

**/



----------------
--SINGLE QUERY-- --this will not return empty rows
----------------
INSERT INTO @ParsedList
        (ListValue)
    SELECT
        ListValue
        FROM (SELECT
                  LTRIM(RTRIM(SUBSTRING(List2, number+1, CHARINDEX(@SplitOn, List2, number+1)-number - 1))) AS ListValue
                  FROM (
                           SELECT @SplitOn + @List + @SplitOn AS List2
                       ) AS dt
                      INNER JOIN Numbers n ON n.Number < LEN(dt.List2)
                  WHERE SUBSTRING(List2, number, 1) = @SplitOn
             ) dt2
        WHERE ListValue IS NOT NULL AND ListValue!=''



RETURN

END --Function FN_ListToTable

you can now use that function like this to query for any list:

DECLARE @List varchar(100)
SET @List='A,B,C'
declare @YourTable table (GroupID char(2),Member char(1))

insert into @YourTable values ('A1','A')
insert into @YourTable values ('A1','B')
insert into @YourTable values ('A1','C')
insert into @YourTable values ('A2','A')
insert into @YourTable values ('A2','B')
insert into @YourTable values ('A3','A')
insert into @YourTable values ('A3','D')
insert into @YourTable values ('A3','E')
insert into @YourTable values ('A5','A')
insert into @YourTable values ('A5','B')
insert into @YourTable values ('A5','C')
insert into @YourTable values ('A5','D')

SELECT t1.GroupID
    FROM @YourTable t1
        LEFT OUTER JOIN @YourTable t2 ON t1.GroupID=t2.GroupID AND t2.Member NOT IN (SELECT ListValue FROM dbo.FN_ListToTable(',',@List))
    WHERE t1.Member IN (SELECT ListValue FROM dbo.FN_ListToTable(',',@List))
        AND t2.GroupID IS NULL
    GROUP BY t1.GroupID
    HAVING COUNT(*) = (SELECT COUNT(*) FROM dbo.FN_ListToTable(',',@List)) 

OUTPUT:

GroupID
-------
A1

Answer 5

select * from ALPHA where Member in( select Member from ALPHA group by Member having COUNT(*) = 3)

Answer 6

Try this one:

SELECT GroupId
  FROM ALPHA
 GROUP BY GroupId
HAVING SUM(CASE WHEN Member='A' THEN 1.0
                WHEN Member='B' THEN 2.0
                WHEN Member='C' THEN 4.0
                ELSE 7.31415
          END) = 7.0

Answer 7

My suggestion is to parse that delimited string into a temp table, and then try something like this.

create table #temp(member varchar(10))

create table #groups
(
groupID varchar(2),
member char(1)
)

--#temp holds the members from your delimited string.
--#groups holds your relationships.



select distinct groupID
from #groups
where 
 (select count(*) from #groups i, #temp t
  where i.member = t.member and i.groupID = #groups.groupID) = 
(select count(*) from #temp)