Question

我尝试进行一次查询，删除Access 2010表中的所有重复行。我的问题是查询正在删除所有重复记录（甚至是原始记录）。

所以，让我说我有1条原始记录和2条重复记录。它不删除2个重复项，而是删除所有3个。

这是我做的：

DELETE *
    FROM Table
    WHERE [field1] IN (SELECT [field1] FROM [table] As Tmp 
    GROUP BY [field1],[field2],[field3],[field4],[field5] HAVING Count(*)>1);

主键字段为（ID）

Answer 1

访问永远不会允许您在DELETE查询中GROUP BY。所以你需要一种根本不同的方法。

创建一个查询，该查询仅为每个重复的字段组返回一行：

SELECT [field1], [field2], [field3], [field4], [field5], Min([ID]) AS keep_id
FROM [table]
GROUP BY [field1], [field2], [field3], [field4], [field5];

将其另存为 qryKeepIDs ，然后使用DELETE查询：

DELETE FROM [Table]
WHERE [Table].ID Not In (SELECT keep_id FROM qryKeepIDs);

Answer 2

以下代码将生成与@HansUp提到的相同的查询。如果您想检查许多不同的表以查找重复项，这应该会带来优势。除了删除＆＃34; all＆＃34;你甚至有机会删除一些重复项。

提出改进意见会很棒：

Public Sub CleanUpDuplicates(strTable As String, strQuery, strID As String, strField1 As String, _
                Optional strField2 As String, _
                Optional strField3 As String, _
                Optional strField4 As String, _
                Optional strField5 As String)

    strQueryDUP = strQuery & "_DUP"
    strQueryCleanUp_Single = strQuery & "_CleanUp_Single"
    strQueryCleanUp_All = strQuery & "_CleanUp_All"

    'SELECT [field1], [field2], [field3], [field4], [field5], Min([ID]) AS keyID
    strSQLDUP = "SELECT [" & strField1
    If strField2 <> "" Then strSQLDUP = strSQLDUP + "], [" & strField2
    If strField3 <> "" Then strSQLDUP = strSQLDUP + "], [" & strField3
    If strField4 <> "" Then strSQLDUP = strSQLDUP + "], [" & strField4
    If strField5 <> "" Then strSQLDUP = strSQLDUP + "], [" & strField5
    strSQLDUP = strSQLDUP & "], Min(" & strID & ") AS keyID "

    'FROM [Table]
    strSQLDUP = strSQLDUP & "FROM " & strTable & " "

    'GROUP BY [field1], [field2], [field3], [field4], [field5];
    strSQLDUP = strSQLDUP & "GROUP BY [" & strField1
    If strField2 <> "" Then strSQLDUP = strSQLDUP + "], [" & strField2
    If strField3 <> "" Then strSQLDUP = strSQLDUP + "], [" & strField3
    If strField4 <> "" Then strSQLDUP = strSQLDUP + "], [" & strField4
    If strField5 <> "" Then strSQLDUP = strSQLDUP + "], [" & strField5
    strSQLDUP = strSQLDUP & "];"


    'FROM [Table]
    strSQLCleanUp = "FROM " & strTable & " "

    'WHERE [Table].ID Not In (SELECT keep_id FROM [Query]);
    strSQLCleanUp = strSQLCleanUp & "WHERE " & strTable & "." & strID & " Not In (SELECT keyID FROM "
    strSQLCleanUp = strSQLCleanUp & strQueryDUP & ");"

    'SELECT [Table].*
    strSQLCleanUp_Single = "SELECT " & strTable & ".* " & strSQLCleanUp
    'DELETE [Table].*
    strSQLCleanUp_All = "DELETE " & strTable & ".* " & strSQLCleanUp



    '###### Create Queries ##########
    Dim db As DAO.Database
    Set db = CurrentDb
    Dim qdfDUP As DAO.QueryDef
    Dim qdfCleanUp_Single As DAO.QueryDef
    Dim qdfCleanUp_All As DAO.QueryDef

    On Error Resume Next
        DoCmd.DeleteObject acQuery, strQueryDUP
        Set qdfDUP = db.CreateQueryDef(strQueryDUP, strSQLDUP)

        DoCmd.DeleteObject acQuery, strQueryCleanUp_Single
        Set qdfCleanUp_Single = db.CreateQueryDef(strQueryCleanUp_Single, strSQLCleanUp_Single)

        DoCmd.DeleteObject acQuery, strQueryCleanUp_All
        Set qdfCleanUp_All = db.CreateQueryDef(strQueryCleanUp_All, strSQLCleanUp_All)
    On Error GoTo 0

    DoCmd.OpenQuery strQueryCleanUp_Single, acViewNormal

    response = MsgBox("Do you like to clean-up all duplicates?", vbOKCancel + vbCritical, "Attention")
    If response = vbOK Then
            DoCmd.Close acQuery, strQueryCleanUp_Single
            DoCmd.OpenQuery strQueryCleanUp_All, acViewNormal
        End If
End Sub

Answer 3

我们可以使用此查询删除重复项并保留唯一的查询，例如。找到3个重复项，它保留1并删除其他2个。

Delete * From myTable Where UniqueID Not IN (select UniqueID From myTable Where UniqueID IN (select Min(UniqueID) From myTable Group by [Duplicates]))

谢谢。

访问使用查询删除重复项

3 个答案: