Removing duplicates from multiple columns

I have a whole bunch of columns, 100+, and each one has many rows of data in it. (Numbers or text or maybe a combination of the two.) I want to delete the duplicate data, but I DON'T want to delete the rows. If you delete an entire row, that would delete data from other columns.
I need a macro that would look at each column independently. I want it to looks at the data in the first cell and then compare it to all of the data in that column in the rows below it. If it finds a match, it deletes that match. It then continues with the next item in that same column. It does this for every item I have in that column.
Then, once it is done with the last row in that column, a bunch of empty cells are left . The macro would then continue and sort the data, putting the empty cells below. The end result is that all of the duplicates are removed (deleted)
The macro then continues until each of the 100+ columns has been checked.
then transposes the data on a separate sheet
A generic code will be better so i can use it for few or many columns (totally new at writing code).
Is this possible?

I'm going to have to assume that the data in adjacent columns are completely unrelated to each other.

Code:

---

Sub NoDupColumnData()
Dim oSheet As Worksheet
Dim oRng As Range
Dim varVals As Variant
  ThisWorkbook.Sheets(1).UsedRange.Copy 'Whatever sheet has your 400 columns
  Set oSheet = ThisWorkbook.Sheets.Add
  oSheet.Name = "Filtered"
  oSheet.Paste
  Application.ScreenUpdating = False
  For Each oRng In oSheet.UsedRange.Columns
    varVals = fcnUnique(oRng)
    oRng.Clear
    oRng.Resize(UBound(varVals), 1).Value = WorksheetFunction.Transpose(varVals)
  Next oRng
  Application.ScreenUpdating = True
lbl_Exit:
  Exit Sub
End Sub
Public Function fcnUnique(oColRng As Range) As Variant
Dim colUniques As New Collection
Dim oCell As Range
Dim lngIndex As Long
Dim vUnique As Variant
  On Error Resume Next
  For Each oCell In oColRng.Cells
    If Len(CStr(oCell)) > 0 Then
      colUniques.Add oCell, CStr(oCell)
    End If
    If Err.Number <> 0 Then
      'colUniques.Add "", CStr(lngIndex)
      'lngIndex = lngIndex + 1
    End If
  Next oCell
  On Error GoTo 0
  ReDim vUnique(1 To colUniques.Count)
  For lngIndex = LBound(vUnique) To UBound(vUnique)
    vUnique(lngIndex) = colUniques(lngIndex)
  Next lngIndex
  fcnUnique = vUnique
lbl_Exit:
  Exit Function
End Function

---

SEE SHEET 2.

How about this

Code:

---

Sub abc()
 Dim arr, iRow As Long, iCol As Long, i As Long
 
 arr = Range("a1").CurrentRegion
 
 Worksheets.Add: i = 1
 With CreateObject("scripting.dictionary")
    For iCol = 1 To UBound(arr, 2)
        For iRow = 1 To UBound(arr)
            If Not .exists(arr(iRow, iCol)) And Trim$(arr(iRow, iCol)) <> "" Then
                .Item(arr(iRow, iCol)) = Empty
            End If
        Next
        Cells(i, "a").Resize(, .Count) = .keys
        .RemoveAll: i = i + 1
    Next
 End With
 Cells(1).CurrentRegion.Borders.LineStyle = xlContinuous
End Sub

---

SORRY, EDITED MY PREVIOUS REPLY INSTEAD OF POSTING A NEW REPLY. SEE ABOVE ATTACHMENT. Visited and read your website. You have a very impressive background.

Does my solution in Post#4 not do what you want?

Perhaps this will do what you want

Code:

---

Sub test()
    Dim arrInput As Variant
    Dim colNum As Long, rowNum As Long
    Dim destinationRange As Range
   
    Set destinationRange = Sheet3.Range("A1")
    arrInput = Sheet1.Range("A1").CurrentRegion
   
    colNum = 1
    For colNum = 1 To UBound(arrInput, 2)
    For rowNum = 1 To UBound(arrInput, 1)
        If arrInput(rowNum, colNum) <> vbNullString Then
            If rowNum <> Application.Match(arrInput(rowNum, colNum), Application.Index(arrInput, 0, colNum), 0) Then
                arrInput(rowNum, colNum) = vbNullString
            End If
        End If
    Next rowNum
    Next colNum
   
    Application.ScreenUpdating = False
    With destinationRange
        With .Resize(UBound(arrInput, 2), UBound(arrInput, 1))
            .Value = Application.Transpose(arrInput)
            .SpecialCells(xlCellTypeBlanks).Delete shift:=xlToLeft
        End With
    End With
    Application.ScreenUpdating = True
End Sub

---

Maybe something like this

Some cells had a leading space so I had to TRIM them also to make certain that ZH9 and spaceZH9 would be de-dupped

Code:

---

Option Explicit

Sub ColumnDeDup()
    Dim rCol As Range
    Dim wsInput As Worksheet, wsOutput As Worksheet

    Application.ScreenUpdating = False

    Set wsInput = Worksheets("Input")

    With wsInput.Cells(1, 1).CurrentRegion
       
        'trim
        .Value = Application.Evaluate("IF(" & .Address & "<>"""",TRIM(" & .Address & "),"""")")
       
        'remove empty cells
        On Error Resume Next
        .Cells.SpecialCells(xlCellTypeBlanks).Delete Shift:=xlUp
        On Error GoTo 0

        'remove dups in each col
        For Each rCol In .Columns
            Call rCol.RemoveDuplicates(1, xlYes)
        Next
    End With
   
   
    'add new worksheet named 'Output'
    Application.DisplayAlerts = False
    On Error Resume Next
    Worksheets("Output").Delete
    On Error GoTo 0
    Application.DisplayAlerts = True
    Worksheets.Add.Name = "Output"
   
    Set wsOutput = Worksheets("Output")
   
    'copy and transpose to Output
    wsInput.Cells(1, 1).CurrentRegion.Copy
    wsOutput.Select
    wsOutput.Cells(1, 1).Select
    Selection.PasteSpecial Paste:=xlPasteAll, Operation:=xlNone, SkipBlanks:=False, Transpose:=True
   
   
    'make Output pretty
    wsOutput.Cells(1, 1).CurrentRegion.EntireColumn.AutoFit
   
   
    Application.ScreenUpdating = True
End Sub

---

All you need:

Code:

---

Sub M_snb()
    Application.ScreenUpdating = False
    Sheet3.UsedRange.ClearContents
   
    For j = 1 To Sheet1.Cells(1).CurrentRegion.Columns.Count
      Sheet1.Cells(1).CurrentRegion.Columns(j).AdvancedFilter 2, , Sheet3.Cells(1, j), True
    Next
    Sheet3.UsedRange.SpecialCells(4).Delete
    sn = Application.Transpose(Sheet3.UsedRange)
   
    Sheet3.UsedRange.ClearContents
   
    Sheet3.Cells(1).Resize(UBound(sn), UBound(sn, 2)) = sn
End Sub

---

gives me:run time error 424

hi Paul.
i need the duplicate values to be gone, only the unique value to remain.

Sorry Mike7952,
new at this so thought i was still replying to gmaxey. i want the duplicate values to go and only teh unique value to remain. see my attachment sheet one above.

In the attached workbook click the Run button on Sheet1 to create a list of required unique values in Sheet2
The code:

Code:

---

Sub ExtractUnique()
'ZVI:2017-01-16
 
  '--> Settings, change to suit
  Const Src = "Sheet1"  ' Name of the source sheet
  Const Dst = "Sheet2"  ' Name of the destination sheet
  ' <-- End of settingw
 
  Dim a
  Dim Col As Range, Rng As Range
  Dim i As Long, j As Long
  Dim s As String
 
  ' Define source data range taking into the account possible empty rows among data rows
  With Worksheets(Src)
    Set Rng = Intersect(.UsedRange, .UsedRange.Cells(1).CurrentRegion.EntireColumn)
  End With
 
  ' Clear the destination sheet
  Sheets("Sheet2").UsedRange.ClearContents
 
  ' Main
  With CreateObject("Scripting.Dictionary")
    .CompareMode = 1
    ' We need to go column by column to prevent overflow of amount values in a()
    For Each Col In Rng.Columns
      a = Col.Cells.Value
      For i = 1 To UBound(a)
        s = Trim(a(i, 1))
        If Len(s) Then .Item(s) = vbNullString
      Next
      If .Count Then
        j = j + 1
        a = .Keys
        Sheets(Dst).Cells(j, 1).Resize(, UBound(a)).Value = a
      End If
      .RemoveAll
    Next
  End With
 
  ' Fit widths and activate the destination sheet
  Sheets(Dst).UsedRange.Columns.AutoFit
  Sheets(Dst).Select
 
End Sub

---

Quote:

---

Originally Posted by grose456

gives me:run time error 424

---

Do not use VBA-code you don't understand.

Of course you have to add a sheet3 first (see attachment)

or use

Code:

---

Sub M_snb()
  sn = Sheet1.Cells(1).CurrentRegion
 
  With CreateObject("scripting.dictionary")
    For jj = 1 To UBound(sn, 2)
      For j = 1 To UBound(sn)
        If .exists(sn(j, jj)) Or sn(j, jj) = "" Then
          sn(j, jj) = ""
        Else
          x0 = .Item(sn(j, jj))
          sn(.Count, jj) = sn(j, jj)
        End If
      Next
      .RemoveAll
    Next
    Sheet2.Cells(1).Resize(UBound(sn, 2), UBound(sn)) = Application.Transpose(sn)
  End With
End Sub

---

Quote:

---

Originally Posted by grose456

hi Paul.
i need the duplicate values to be gone, only the unique value to remain.

---

I thought they were

Can you provide an example?

In post #8 with the macro and the attachment, 'Original' is your data, 'Input' is a copy I made of 'Original' for testing

When I ran the macro on 'Input' it appeared to remove the duplicates on a column-by-column basis

Attachment 18045

'Output' is the transpose you asked for using 'input' as an input

Paul
I believe I have a solution in post#4 also based on a column to column basis.

Code:

---

Sub abc()
    Dim arr, iRow As Long, iCol As Long, i As Long
   
    arr = Range("a1").CurrentRegion
   
    Worksheets.Add: i = 1
    With CreateObject("scripting.dictionary")
        For iCol = 1 To UBound(arr, 2)
            For iRow = 1 To UBound(arr)
                If Not .exists(arr(iRow, iCol)) And Trim$(arr(iRow, iCol)) <> "" Then
                    .Item(arr(iRow, iCol)) = Empty
                End If
            Next
            Cells(i, UBound(arr, 2) + 3).Resize(, .Count) = .keys
            Cells(1, iCol).Resize(.Count) = Application.Transpose(.keys)
            .RemoveAll: i = i + 1
        Next
    End With
    Cells(1).CurrentRegion.Borders.LineStyle = xlContinuous
    Cells(1, UBound(arr, 2) + 5).CurrentRegion.Borders.LineStyle = xlContinuous
End Sub

---

Well, since Excel provides a built in .RemoveDuplicates method, I figure it's easier and faster to just use it instead of rolling my own

i went back and checked the file. when i say i want to delete duplicates i want only the unique value to remailn. for example in the attachment "original" the pink colored cell are duplicates which i need to disppear and only the green value to remain.

i would have done that except i need it for one column at a time. done independently of each other. i tried doing it one column at a time but after 30 columns i figured there had to be an easier way.

see post #3