oracle數據庫去重，只保留一條數據

一、使用 Row_Number() Over(Partition By ... Order By ...) 為每個分組生成一個內部序號

第一步，根據需求查詢重復數據，並為每個分組生成一個內部行號

Select t.*,
       Row_Number() Over(Partition By Visit_Id, Presc_Detail_Id, Rule_Code Order By t.Visit_Id) Su
  From Test t;

第二步，篩選出行號大於1的，即重復的多余的數據

Select *
  From (Select t.*,
               Row_Number() Over(Partition By Visit_Id, Presc_Detail_Id, Rule_Code Order By t.Visit_Id) Su
          From Test t)
 Where Su > 1;

第三步，刪除重復數據，只保留1條

Delete From Test
 Where Id In (Select Id
                From (Select t.*,
                             Row_Number() Over(Partition By Visit_Id, Presc_Detail_Id, Rule_Code Order By t.Visit_Id) Su
                        From Test t)
               Where Su > 1);

這種方式有個弊端，如果數據本身並沒有每行的唯一標識id，刪除重復數據還是非常困難的。

 

二、使用 ROWID

第一步，查找重復的記錄

Select *
  From Test t
 Where (Visit_Id, Presc_Detail_Id, Rule_Code) In
       (Select Visit_Id, Presc_Detail_Id, Rule_Code
          From Test
         Group By Visit_Id, Presc_Detail_Id, Rule_Code
        Having Count(*) > 1)
 Order By Visit_Id, Presc_Detail_Id, Rule_Code;

第二步，查找表中多余的重復記錄，不包含rowid最小的記錄

Select *
  From Test t
 Where (Visit_Id, Presc_Detail_Id, Rule_Code) In
       (Select Visit_Id, Presc_Detail_Id, Rule_Code
          From Test
         Group By Visit_Id, Presc_Detail_Id, Rule_Code
        Having Count(*) > 1)
   And Rowid Not In (Select Min(Rowid)
                       From Test
                      Group By Visit_Id, Presc_Detail_Id, Rule_Code
                     Having Count(*) > 1);

第三步，刪除表中多余的重復記錄，只保留rowid最小的記錄

Delete From Test t
 Where (Visit_Id, Presc_Detail_Id, Rule_Code) In
       (Select Visit_Id, Presc_Detail_Id, Rule_Code
          From Test
         Group By Visit_Id, Presc_Detail_Id, Rule_Code
        Having Count(*) > 1)
   And Rowid Not In (Select Min(Rowid)
                       From Test
                      Group By Visit_Id, Presc_Detail_Id, Rule_Code
                     Having Count(*) > 1);