mezamorphic mezamorphic - 5 months ago 12
SQL Question

Selecting first row per group

I have a table of 10 rows (apologies but i cannot seem to format examples of tables on SO) and there are three columns-

A,B,C
.

The values are:

{1,-10,5},
{1,0,5},
{1,10,5},
{2,10,12},
{2,0,12},
{3,-10,14},
{4,0,8},
{4,10,8},
{5,0,6},
{5,1,6},
{5,-5,6}


If I first order the data by Col 1, then ColB, then ColC (although i did make all Col C values the same per Col A value) how would I select the "first row" per Col A?

So this should result in:

{1,-10,5},
{2,0,12},
{3-10,14},
{4,0,8},
{5,-5,6}

Answer
SELECT  a, b, c
FROM    (
        SELECT  *, ROW_NUMBER() OVER (PARTITION BY a ORDER BY b, c) rn
        FROM    mytable
        ) q
WHERE   rn = 1
ORDER BY
        a

or

SELECT  mi.*
FROM    (
        SELECT  DISTINCT  a
        FROM    mytable
        ) md
CROSS APPLY
        (
        SELECT  TOP 1 *
        FROM    mytable mi
        WHERE   mi.a = md.a
        ORDER BY
                b, c
        ) mi
ORDER BY
        a

Create a composite index on (a, b, c) for the queries to work faster.

Which one is more efficient depends on your data distribution.

If you have few distinct values of a but lots of records within each a, the second query would be better.

You could improve it even more by creating an indexed view:

CREATE VIEW v_mytable_da
WITH   SCHEMABINDING
AS
       SELECT  a, COUNT_BIG(*) cnt
       FROM    dbo.mytable
       GROUP BY
               a

GO

CREATE UNIQUE CLUSTERED INDEX
       pk_vmytableda_a
ON     v_mytable_da (a)

GO

SELECT  mi.*
FROM    v_mytable_da md
CROSS APPLY
        (
        SELECT  TOP 1 *
        FROM    mytable mi
        WHERE   mi.a = md.a
        ORDER BY
                b, c
        ) mi
ORDER BY
        a