Select count() max() Date

I have a table with shifts history along with emp ids.

I'm using this query to retrieve a list of employe开发者_如何学Goes and their total shifts by specifying the range to count from:

SELECT ope_id, count(ope_id)
FROM operator_shift
WHERE ope_shift_date >=to_date( '01-MAR-10','dd-mon-yy') and ope_shift_date
<= to_date('31-MAR-10','dd-mon-yy')
GROUP BY OPE_ID 

which gives

   OPE_ID      COUNT(OPE_ID)
     1            14
     2             7
     3             6
     4             6
     5             2
     6             5
     7             2
     8             1
     9             2
    10             4

10 rows selected.

How do I choose the employee with the highest number of shifts under the specified range date?

---

Assuming your version of Oracle is new enough to support common table expressions:

With ShiftCounts As
    (
    SELECT ope_id, count(ope_id) ShiftCount
        , ROW_NUMBER() OVER( ORDER BY Count(ope_id) Desc ) ShiftRank
    FROM operator_shift
    WHERE ope_shift_date >=to_date( '01-MAR-10','dd-mon-yy') 
        and ope_shift_date <= to_date('31-MAR-10','dd-mon-yy')
    GROUP BY OPE_ID 
    )
Select ope_id, ShiftCount
From ShiftCounts
Where ShiftRank = 1

---

something like this maybe:

SELECT TOP 1 ope_id, Count(ope_id)
FROM operator_shift
WHERE ope_shift_date >=to_date( '01-MAR-10','dd-mon-yy') and ope_shift_date
<= to_date('31-MAR-10','dd-mon-yy')
GROUP BY OPE_ID 
ORDER BY Count(ope_id) DESC

---

Use:

SELECT t.ope_id,
       t.num
  FROM (SELECT os.ope_id, 
               COUNT(os.ope_id) AS num
          FROM OPERATOR_SHIFT os
         WHERE os.ope_shift_date BETWEEN TO_DATE('01-MAR-10','dd-mon-yy') 
                                     AND TO_DATE('31-MAR-10','dd-mon-yy')
      GROUP BY os.ope_id
      ORDER BY num DESC) t
 WHERE ROWNUM = 1

Reference:

• Ask Tom: On ROWNUM and Limiting Results