使用 SQL Server 2008 统计数据

Question

I'm using SQL Server 2008 and I have a big database, I need to count the data for every manzana but every rows is 0 so I don't know how to do that

SELECT  
    Manzana, 
    SUM(CASE WHEN Manzana   IN( 'CBrc','CBrc1', 'CBrc2') THEN 1 ELSE 0 END )AS CBrc,
    SUM(CASE WHEN  Manzana    IN ('W1','W11','W12','W13','W14','W15','W16') THEN 1 ELSE 0 END) AS W1,
    SUM(CASE WHEN  Manzana   IN('PC1','PC11') THEN 1 ELSE 0 END) AS PC1,
    SUM(CASE WHEN  Manzana  IN ('CLu','CLU1') THEN 1 ELSE 0 END) AS CLu,
    SUM(CASE WHEN  Manzana   IN('S3','S31') THEN 1 ELSE 0 END) AS S3,
    SUM(CASE WHEN  Manzana  IN ('C2L') THEN 1 ELSE 0 END) AS C2L,
    SUM(CASE WHEN  Manzana    IN ('PCA') THEN 1 ELSE 0 END) AS PCA,
    SUM(CASE WHEN  Manzana    IN ('ADO') THEN 1 ELSE 0 END) AS AD,
    SUM(CASE WHEN  Manzana   IN ('PRC') THEN 1 ELSE 0 END) AS CBu,
    SUM(CASE WHEN  Manzana    IN('SIN') THEN 1 ELSE 0 END) AS SIN_ACCESO_A_INFORMACION   
FROM  
    dbo.Hoja3$
GROUP BY 
    Manzana

I need to obtain something like this

Manzana CBrc W1 PC1 CLU S3 C2L 
5445     8   11  0   0   0  0 
4545     15   0  30  0   0  0
5455     21   0  5   3   2  0
7893     0    0  3   1   0  0
1566     100  4  1   0   0  80

Answer 1

After reading your comment "I try to count how many Cbrc,Pc1, Clu etc I have in every Manzana" I see the problem.

You're using the IN operator incorrectly.

To test fo the occurrence of a substring use the CHARINDEX function:

SELECT
    Manzana,
    SUM( CASE WHEN CHARINDEX( 'SIN', Manzana ) > 0 THEN 1 ELSE 0 END ) AS SIN,
    SUM( CASE WHEN CHARINDEX( 'PCA', Manzana ) > 0 THEN 1 ELSE 0 END ) AS PCA,
    ...
FROM
    Hoja3$
GROUP BY
    Manzana

Answer 2

You can't group by Manzana and not group by Manzana . Also, from the sample data, I can tell that you haven't provided enough information. Just from the first row:

Manzana CBrc W1 PC1 CLU S3 C2L 
5445     8   11  0   0   0  0 

If Manzana is 5445 , then what column has the value CBrc ? It can't be the Manzana column, but has to be some other column. In this vacuum of information, I'm going to call the mystery column Code .

Then try this (of course, you'll need to put all your code groups in):

WITH CodeGroupings AS (
   SELECT * FROM (VALUES -- Do this in a permanent table instead!
      ('CBrc%', 'CBrc'), -- can use wildcards
      ('W1_', 'W1'), -- or single letter wildcards
      ('PC1', 'PC1'), -- or just put individual values
      ('PC11', 'PC1'), -- mapping to the same group as the previous line
      ('SIN', 'SIN_ACCESO_A_INFORMACION')
   ) M (CodePattern, CodeGroup)
), ManzanaCodeGroups AS (
   SELECT
      H.Manzana,
      CG.CodeGroup
   FROM
      dbo.Hoja3$ H
      CROSS APPLY (
         SELECT TOP 1 *
         FROM CodeGroupings CG
         WHERE H.Code LIKE CG.CodePattern
      ) CG
)
SELECT
   *
FROM
   ManzanaCodeGroups
   PIVOT (Count(*) FOR (CodeGroup IN CBrc, W1, PC1, SIN_ACCESO_A_INFORMACION)) P
;