sql >> Database >  >> RDS >> Sqlserver

Meerdere PIVOTS? Moet per uur per maand tellen

Een manier om het te doen

WITH hour_list AS (
  SELECT 0 hour
  UNION ALL
  SELECT hour + 1 FROM hour_list WHERE hour < 23
)
SELECT h.hour,
       COALESCE(jan, 0) jan,
       COALESCE(feb, 0) feb,
       COALESCE(mar, 0) mar,
       COALESCE(apr, 0) apr,
       COALESCE(may, 0) may,
       COALESCE(jun, 0) jun,
       COALESCE(jul, 0) jul,
       COALESCE(aug, 0) aug,
       COALESCE(sep, 0) sep,
       COALESCE(oct, 0) oct,
       COALESCE(nov, 0) nov,
       COALESCE(dec, 0) dec
  FROM hour_list h LEFT JOIN
(
  SELECT DATEPART(HOUR, logtime) hour, 
         SUM(CASE WHEN DATEPART(MONTH, logtime) =  1 THEN 1 END) jan,
         SUM(CASE WHEN DATEPART(MONTH, logtime) =  2 THEN 1 END) feb,
         SUM(CASE WHEN DATEPART(MONTH, logtime) =  3 THEN 1 END) mar,
         SUM(CASE WHEN DATEPART(MONTH, logtime) =  4 THEN 1 END) apr,
         SUM(CASE WHEN DATEPART(MONTH, logtime) =  5 THEN 1 END) may,
         SUM(CASE WHEN DATEPART(MONTH, logtime) =  6 THEN 1 END) jun,
         SUM(CASE WHEN DATEPART(MONTH, logtime) =  7 THEN 1 END) jul,
         SUM(CASE WHEN DATEPART(MONTH, logtime) =  8 THEN 1 END) aug,
         SUM(CASE WHEN DATEPART(MONTH, logtime) =  9 THEN 1 END) sep,
         SUM(CASE WHEN DATEPART(MONTH, logtime) = 10 THEN 1 END) oct,
         SUM(CASE WHEN DATEPART(MONTH, logtime) = 11 THEN 1 END) nov,
         SUM(CASE WHEN DATEPART(MONTH, logtime) = 12 THEN 1 END) dec
    FROM somelog (NOLOCK)
   GROUP BY DATEPART(HOUR, logtime)
) l 
    ON h.hour = l.hour

of met PIVOT

WITH hour_list AS (
  SELECT 0 hour
  UNION ALL
  SELECT hour + 1 FROM hour_list WHERE hour < 23
)
SELECT h.hour,
       COALESCE([1],  0) jan,
       COALESCE([2],  0) feb,
       COALESCE([3],  0) mar,
       COALESCE([4],  0) apr,
       COALESCE([5],  0) may,
       COALESCE([6],  0) jun,
       COALESCE([7],  0) jul,
       COALESCE([8],  0) aug,
       COALESCE([9],  0) sep,
       COALESCE([10], 0) oct,
       COALESCE([11], 0) nov,
       COALESCE([12], 0) dec
  FROM hour_list h LEFT JOIN 
(
  SELECT DATEPART(MONTH, logtime) month,
         DATEPART(HOUR, logtime) hour,
         COUNT(*) log_count
    FROM somelog (NOLOCK)
   GROUP BY DATEPART(MONTH, logtime), DATEPART(HOUR, logtime)
) s
PIVOT
(
  SUM(log_count) FOR month IN([1],[2],[3],[4],[5],[6],[7],[8],[9],[10],[11],[12])
) p
    ON h.hour = p.hour

Voorbeelduitvoer voor beide zoekopdrachten:

| HOUR | JAN | FEB | MAR | APR | MAY | JUN | JUL | AUG | SEP | OCT | NOV | DEC |
|------|-----|-----|-----|-----|-----|-----|-----|-----|-----|-----|-----|-----|
|    0 |   2 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |
|    1 |   1 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |
|    2 |   0 |   1 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |
|    3 |   0 |   0 |   1 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |
|    4 |   0 |   0 |   0 |   1 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |   0 |
...

Hier is SQLFiddle demo (met behulp van CASE )
Hier is SQLFiddle demo (met behulp van PIVOT )




  1. CASE gebruiken in PostgreSQL om meerdere kolommen tegelijk te beïnvloeden

  2. Prestaties van mysql verbeteren LOAD DATA / mysqlimport?

  3. Update MySQL-tabel in brokken

  4. Selecteer max(datum) werkt niet als de rijen verschillende waarden hebben, ik wil alleen de rij met de hoogste datum ophalen