where子句中的多個列值

Question

我需要顯示每周的單用戶時間表，如時間表，

場景：一個教師被分配到一個星期的多個批次（例如：BBA，數學和上午的第1小時和第2小時）和（MBA，數學，前三個小時和第4小時）在同一個日期說（30-06-2015）我的gridview行將存儲為1和2行作為存儲為2，依此類推.........

我的表定義：

CREATE TABLE [dbo].[test] (
    [datedif]     NVARCHAR (50)  NOT NULL,
    [hour]         INT              NULL,
    [subject]    NVARCHAR (MAX) NULL,
    [faculty]    NVARCHAR (MAX) NULL,
    [attendence] BIT            NULL,
    [dayweek]     NVARCHAR (50)  NULL,
    [weekmonth]   NVARCHAR (MAX) NULL,
    [batch]       NVARCHAR (MAX) NULL,
    [section]     NVARCHAR (MAX) NULL,
    [session]     NVARCHAR (MAX) NULL
);

表看起來像：

Datefdiff | hour | subject | faulty| batch
----------+-------+----------+---------+-------+----------+--------+-------+----------+---------+-------
30-06-2015| 1| Maths    | Kevin   | BBA
30-06-2015| 2| Science  | Amal    | MBA
30-06-2015|3 | chemistry|Jaya     |BBA
30-06-2015|4 | chemistry|Jaya     |BBA 
30-06-2015|5 | chemistry|Jaya     |BBA
31-06-2015 |1| science  | Amal    |BBA
31-06-2015 |2| Maths    | kevin   |BBA 
31-06-2015 |3| Science  | Amal    |BBA 
31-06-2015 |4 | chemistry|Jaya     |BBA
31-06-2015 |5| science  | Amal    |BBA

只為教師提供預期產出：Amal

Datefdiff |hour|subject| batch |hour|subject | batch |faculty|hour | subject | batch | hour | subject | batch| hour | subject | batch | 
----------+-------+----------+-------+-------+----------+-------+-------+----------+
30-06-2015| 1    | Maths| BBA| 2| Science  | MBA   | 3| Science  |   BBA| 4| chemistry| BBA | 5 |Physics |MBA
31-06-2015| 1    | Maths| BBA| 2| Science  | MBA   | 3| Science  |   BBA| 4| chemistry| BBA | 5 |Physics |MBA

Answer 1

這可能是您的表設計的起點：

declare @tbFaculty table (
      FacultyID int --identity(1,1) primary key
    , Name varchar(50)
)

insert into @tbFaculty ( FacultyID, Name )
values    ( 1, 'Kevin' )
        , ( 2, 'Amal' )

declare @tbBatch table(
      BatchID int --identity(1,1) primary key
    , Name char(3)
)

insert into @tbBatch ( BatchID, Name )
values    ( 1, 'BBA' )
        , ( 2, 'MBA' )

declare @tbClass table (
      [Hour] tinyint
    , [Subject] nvarchar (128)
    , [FacultyID] int
    , [Attendence] bit
    , [BatchID] char(3)
    , [ClassDate] date
)

insert into @tbClass ( [Hour], [Subject], FacultyID, Attendence, BatchID, ClassDate )
values    ( 1, 'Maths', 1, 1, 1, '2015-06-30' )
        , ( 2, 'Maths', 1, 1, 1, '2015-06-30' )
        , ( 3, 'Science', 2, 1, 1, '2015-06-30' )
        , ( 1, 'Science', 2, 1, 2, '2015-06-30' )
        , ( 2, 'Science', 2, 1, 2, '2015-06-30' )
        , ( 3, 'Maths', 1, 1, 2, '2015-06-30' )

select
    cl.ClassDate
    , cl.[Hour]
    , cl.[Subject]
    , ba.Name as BatchName
    , fa.Name as FacultyName
from
    @tbClass cl
    inner join @tbBatch ba on ba.BatchID = cl.BatchID
    inner join @tbFaculty fa on fa.FacultyID = cl.FacultyID
where
    fa.Name = 'Amal'

您也可以規范化主題。

Answer 2

我只想將此作為單獨的答案添加。 試試這個：

--INSERT INTO dbo.test(datedif,[hour],[subject],faculty,batch)
--VALUES
--('30-06-2015',1,'Maths','Kevin','BBA'),
--('30-06-2015',2,'Science','Amal','MBA'),
--('30-06-2015',3,'chemistry','Jaya','BBA'),
--('30-06-2015',4,'chemistry','Jaya','BBA'),
--('30-06-2015',5,'chemistry','Jaya','BBA'),
--('31-06-2015',1,'science','Amal','BBA'),
--('31-06-2015',2,'Maths','kevin','BBA'),
--('31-06-2015',3,'Science','Amal','BBA'),
--('31-06-2015',4,'chemistry','Jaya','BBA'),
--('31-06-2015',5,'science','Amal','BBA');

WITH CTE_Hours
AS
(
    SELECT
            1 AS hour1,
            MAX(CASE WHEN [Hour] = 1 THEN [subject] END)    AS subject1,
            MAX(CASE WHEN [Hour] = 1 THEN [batch] END)      AS batch1 ,

            2 AS hour2,
            MAX(CASE WHEN [Hour] = 2 THEN [subject] END)    AS subject2,
            MAX(CASE WHEN [Hour] = 2 THEN [batch] END)      AS batch2

            --etc...
    FROM dbo.test
    WHERE faculty = 'Amal'
)

SELECT *
FROM    (
            SELECT DISTINCT datedif
            FROM dbo.test
        ) A
CROSS JOIN CTE_Hours

Answer 3

以下是如何規范化您的數據，這將使查詢更加簡單。

使用表定義，我插入了您的數據

INSERT INTO test(Datedif,hour1,subject1,faculty1,hour2,subject2,faculty2,hour3,subject3,faculty3,batch)
VALUES  ('30-06-2015',1,'Maths','Kevin',1,'Maths','Kevin',1,'Science','Amal','BBA'),
        ('30-06-2015',1,'Science','Amal',1,'Science','Amal',1,'Maths','Kevin','MBA');

然后我使用較窄的數據類型為您的數據創建了一個新的規范化結構。 它有助於保持數據清潔以及不浪費存儲空間，這意味着查詢速度更快，因為它們不必處理如此多的數據。

CREATE TABLE new_test
(
    dt          DATE            NULL,
    hr          TINYINT         NULL, --holds values between 0 to 255
    subj        VARCHAR(100)    NULL, --plenty big enough. No need for NVARCHAR unless you are using Unicode characters
    faculty     VARCHAR(100)    NULL,
    attendance  BIT             NULL,
    dayweek     TINYINT         NULL,
    weekmonth   TINYINT         NULL,
    section     VARCHAR(100)    NULL, --not sure what this is
    sess        VARCHAR(100)    NULL, --not sure what this is
    batch       CHAR(3)         NULL  --looks like there are three character codes
);

這是我規范化數據的地方。 我不確定你是否有出勤2,3,4等...如果你在實際的表中，那么你應該修復我的代碼。

注意：我計算了從日期開始的每周和每周。 我做了最好的猜測，如果它們不正確，那么隨意調整它們！

WITH norm_data
AS
(
SELECT datedif,hour1,subject1,faculty1,attendence1,dayweek,weekmonth,section,[session],batch FROM test
UNION ALL
SELECT datedif,hour2,subject2,faculty2,attendence1,dayweek,weekmonth,section,[session],batch FROM test
UNION ALL
SELECT datedif,hour3,subject3,faculty3,attendence1,dayweek,weekmonth,section,[session],batch FROM test
UNION ALL
SELECT datedif,hour4,subject4,faculty4,attendence1,dayweek,weekmonth,section,[session],batch FROM test
UNION ALL
SELECT datedif,hour5,subject5,faculty5,attendence1,dayweek,weekmonth,section,[session],batch FROM test
)

INSERT INTO new_test
SELECT  PARSE(datedif AS DATE USING 'de-DE') AS [datedif],
        hour1,
        subject1,
        faculty1,
        attendence1,
        DATEPART(WEEKDAY,PARSE(datedif AS DATE USING 'de-DE')) AS dayweek,
        datepart(day, datediff(day, 0, PARSE(datedif AS DATE USING 'de-DE'))/7 * 7)/7 + 1 AS weekmonth,
        section,
        [session],
        batch
FROM norm_data

現在讓我們來看看你的新表

SELECT *
FROM new_test

如果它是正確的，那么這里是如何重命名表

EXEC SP_rename  @objname = 'test', --if you don't want to drop the old table
                @newname = 'test_old'

EXEC SP_rename  @objname = 'new_test', --now give the new table the actual name
                @newname = 'test'


SELECT *
FROM test