SqlServer和Oracle中一些常用的sql语句10 特殊应用

--482, ORACLE / SQL SERVER
--订购数量超过平均值的书籍
WITH Orders_Book
AS
  (
  SELECT Book_Name, SUM(Qty) Book_Qty
  FROM Orders
  GROUP BY Book_Name
  )
SELECT *
FROM Orders_Book
WHERE Book_Qty >
      (
      SELECT AVG(Book_Qty)
      FROM Orders_Book
      )
 
--递归 产生连续数列1至10000
WITH Tally(N)
AS
  (
  SELECT 1 N
  --FROM DAUL       -- ORACLE
  UNION ALL
  --2.递归区块
  SELECT N+1
  FROM Tally
  WHERE N<=10000
  )
  SELECT N
FROM TALLY
OPTION (MAXRECURSION 10000)  --SQL SERVER设定深度    
 
--490, SQL SERVER
--随机抽出3笔员工数据
SELECT TOP 3
    E.Emp_Id
    , E.Emp_Name
    , E.Dept_Id
FROM Employees E
ORDER BY NEWID()       
 
--491, SQL SERVER
--在I100和I200部门中随机抽出一名员工(子分组中各抽出N笔)
SELECT E.Emp_Id
       , E.Emp_Name
       , E.Dept_Id
FROM
  (
  SELECT Emp_Id, Emp_Name, Dept_Id
         , ROW_NUMBER() OVER (PARTITION BY Dept_Id
                                      ORDER BY NEWID()) RowNo
  FROM Employees
  WHERE Dept_Id IN ('I100', 'I200')
  ) E
WHERE E.RowNo <=1 
 
--492, ORACLE
--随机抽出3笔员工数据
SELECT Emp_Id
    , Emp_Name
    , Dept_Id
FROM
  (
    SELECT *
    FROM Employees
    ORDER BY DBMS_RANDOM.VALUE()
  )
WHERE ROWNUM<=3 
 
--493, ORACLE
--在I100和I200部门中随机抽出一名员工(子分组中各抽出N笔)
SELECT E.Emp_Id
       , E.Emp_Name
       , E.Dept_Id
FROM
(
SELECT Emp_Id, Emp_Name, Dept_Id
   , ROW_NUMBER()
         OVER (PARTITION BY Dept_Id
               ORDER BY DBMS_RANDOM.VALUE()) RowNo
   FROM Employees
   WHERE Dept_Id IN ('I100', 'I200')
   ) E
WHERE E.RowNo <=1 
 
--495, SQL SERVER
--以符号分割的字符串 分拆成table返回,含一字段 Column_Value
create function [dbo].[m_split](@c varchar(2000),@split varchar(2))
    returns @t table(col varchar(200))
as
begin
      while(charindex(@split,@c)<>0)
        begin
          insert @t(col) values (substring(@c,1,charindex(@split,@c)-1))
          set @c = stuff(@c,1,charindex(@split,@c),'')
        end
      insert @t(col) values (@c)
      return
end
 
--测试
select * from [dbo].[m_split]('1,2,3', ',')
 
--496, ORACLE
--以符号分割的字符串 分拆成table返回,含一字段 Column_Value
CREATE OR REPLACE TYPE split_tbl AS TABLE OF VARCHAR2(32767);
 
--测试1
SELECT Column_Value
FROM TABLE(Split_Tbl(1,2,3))
 
--测试2
SELECT Column_Value
FROM TABLE(Split_Tbl('A','B','C'))