Grouping two columns to shrink row count by comparing | code optimization - excel

Question

I try to find a vba solution for the following problem:
I have two columns and try to group column1 in a comma separate way to have less rows.
e.g.
example:
I tried this, and it worked - but It take too long (about 300.000 Rows). Is there any better solution that task?
*Its just one part of my macro
For Each r In fr
If st = "" Then
st = Application.WorksheetFunction.Clean(Trim(ws.Cells(r.row, "L").Value))
Else
If Not IsInArray(Split(st, ","), ws.Cells(r.row, "L").Value) Then
st = st & ", " & Application.WorksheetFunction.Clean(Trim(ws.Cells(r.row, "L").Value))
End If
End If
If usrCheck = True Then
If str = "" Then
str = Application.WorksheetFunction.Clean(Trim(ws.Cells(r.row, "A").Value))
Else
If Not IsInArray(Split(str, ","), ws.Cells(r.row, "A").Value) Then
str = str & ", " & Application.WorksheetFunction.Clean(Trim(ws.Cells(r.row, "A").Value))
End If
End If
End If
Next

Answer 1

Maybe using Dictionary would be fast. What about:
Sub Test()
Dim x As Long, lr As Long, arr As Variant
Dim dict1 As Object: Set dict1 = CreateObject("Scripting.Dictionary")
Dim dict2 As Object: Set dict2 = CreateObject("Scripting.Dictionary")
With Sheet1 'Change accordingly
'Return your last row from column A
lr = .Cells(.Rows.Count, 1).End(xlUp).Row
'Get array and loop through it
arr = .Range("A2:B" & lr).Value
For x = LBound(arr) To UBound(arr)
dict1(arr(x, 2)) = arr(x, 2)
Next
'Loop through dictionary filling a second one
For Each Key In dict1.keys
For x = LBound(arr) To UBound(arr)
If arr(x, 2) = Key Then dict2(arr(x, 1)) = arr(x, 1)
Next x
.Range("F" & .Cells(.Rows.Count, 6).End(xlUp).Row + 1) = Key
.Range("G" & .Cells(.Rows.Count, 7).End(xlUp).Row + 1) = Join(dict2.Items, ", ")
dict2.RemoveAll
Next
End With
End Sub
This will get you all unique items from column A though, so if there can be duplicates and you want to keep them, this is not for you =)

Answer 2

Try also this, please. It works only in memory and on my computer takes less then 3 seconds for 300000 rows. The range must be filtered, like in your picture. If not, the filtering can also be easily automated.
Private Sub CondensData()
Dim sh As Worksheet, arrInit As Variant, arrIn As Variant, i As Long
Dim arrFinal() As Variant, lastRow As Long, Nr As Long, El As Variant
Dim strTemp As String, k As Long
Set sh = ActiveSheet
lastRow = sh.Cells(sh.Rows.count, "A").End(xlUp).Row
arrIn = sh.Range("B2:B" & lastRow + 1).Value
'Determine the number of the same accurrences:
For Each El In arrIn
i = i + 1
If i >= 2 Then
If arrIn(i, 1) <> arrIn(i - 1, 1) Then Nr = Nr + 1
End If
Next
ReDim arrFinal(Nr, 1)
arrInit = sh.Range("A2:B" & lastRow).Value
For i = 2 To UBound(arrInit, 1)
If i = 1 Then
strTemp = arrInit(1, 1)
Else
If arrInit(i, 2) = arrInit(i - 1, 2) Then
If strTemp = "" Then
strTemp = arrInit(i, 1)
Else
strTemp = strTemp & ", " & arrInit(i, 1)
End If
Else
arrFinal(k, 0) = arrInit(i - 1, 2)
arrFinal(k, 1) = strTemp
k = k + 1: strTemp = ""
End If
End If
Next i
sh.Range("C2:D" & lastRow).Clear
sh.Range("C2:D" & k - 1).Value = arrFinal
sh.Range("C:D").EntireColumn.AutoFit
MsgBox "Solved..."
End Sub
It will return the result in columns C:D