I need help with the script below. I need to highlight all cells in range(r) which contain values from another range(dictionary), but currently it only highlights first occurrences of each cell in the dictionary range.
Sub SearchAndFormat_Click()
Dim Dictionary As Variant
Dictionary = Range("L5:L9")
Dim r As Range, cell As Variant
Set r = Application.InputBox("Select range", "Selection Window", Type:=8)
r.ClearFormats
r.NumberFormat = "General"
For Each subj In Dictionary
For Each cell In r
Set target_cell = r.Find(subj, LookIn:=xlValues, SearchOrder:=xlByRows, SearchDirection:=xlNext)
If Not target_cell Is Nothing Then
target_cell.Interior.ColorIndex = 4
End If
Next
Next
End Sub
I used to have a version of code without nested loop, but it would only highlight the first occurrence of the first value in the dictionary range:
For Each cell In r
Set target_cell = r.Find(Dictionary, LookIn:=xlValues, SearchOrder:=xlByRows, SearchDirection:=xlNext)
If Not target_cell Is Nothing Then
target_cell.Interior.ColorIndex = 4
End If
Next
Any help is greatly appreciated!
If you use Find in order find all values equal to the target, then it should be sort of a Do ... Loop with active use of a parameter After:=... For example:
Set x = MyRange.Find(target)
If Not x Is Nothing Then
Start = x.Address
Do
Debug.Print x.Address, x.Value
Set x = MyRange.FindNext(After:=x)
Loop While x.Address <> Start
End If
But there's no need to use Find when you loop through each value in a dictionary and in a range of interest:
Sub test_colorizing()
Dim d As Range
Dim r As Range
Dim x, y
Set d = Range(...) ' Dictionary
Set r = Range(...) ' Range of interest
For Each x In r
For Each y In d
If x = y Then
x.Interior.ColorIndex = 4
Exit For ' go to the next word in r
End If
Next
Next
End Sub
How it can look with Find:
Sub test_colorizing_with_find()
Dim dict As Range ' Dictionary
Dim rng As Range ' Range of interest
Dim cell, word, start
Set dict = Range(...)
Set rng = Range(...)
For Each word In dict
Set cell = rng.Find(word)
If Not cell Is Nothing Then
start = cell.Address
Do
cell.Interior.ColorIndex = 4
Set cell = rng.FindNext(cell)
Loop While cell.Address <> start
End If
Next
End Sub
Related
As the title states, I need to find all cells in a worksheet with a specific string in them and replace the cell 2 columns to the right with a specific value
Edit: this is what I have so far but no idea where to go from here
Sub t()
Dim searchCell As Range
Dim replaceCell As Range
With Sheets("Chainwire")
Set searchCell = .Cells.Find(what:="UFFT50")
Set replaceCell = searchCell.Offset(0, 2)
End With
End Sub
Any help is much appreciated
Try this:
Sub ReplaceItems()
Dim col, c, dest As Range
Set col = FindAll(Sheets("Chainwire").UsedRange, "UFFT50") 'get all matches
'loop matches and replace value 2 cells over
For Each c In col
c.Offset(0, 2).Value = "a specific value"
Next
End Sub
'search a Range for `val` and return all matches as a Collection
' of cells
Public Function FindAll(rng As Range, val As String) As Collection
Dim rv As New Collection, f As Range
Dim addr As String
'adjust Find() arguments to suit your need...
Set f = rng.Find(what:=val, after:=rng.Cells(rng.Cells.CountLarge), _
LookIn:=xlValues, LookAt:=xlPart, SearchOrder:=xlByRows, _
SearchDirection:=xlNext, MatchCase:=True)
If Not f Is Nothing Then addr = f.Address() 'remember first cell found
Do Until f Is Nothing
rv.Add f
Set f = rng.FindNext(after:=f)
If f.Address() = addr Then Exit Do 'exit if we looped back to the first hit
Loop
Set FindAll = rv
End Function
Focusing on:
rows = w2.Columns("A:A").Find(What:=aCell, LookAt:=xlWhole).Row
Instead of looking at the whole cell to match (xlwhole), I'm trying to only match the first 6 characters of the cell. I've looked into xlpart and a few other options but have been unsuccessful.
Sub test()
Dim aCell
Dim A, B As Long, rows As Long
Dim w1, w2 As Worksheet
Dim cell As Range
Set w1 = Workbooks("Book1").Sheets("Sheet1")
Set w2 = Workbooks("Book2").Sheets("Sheet1")
A = w1.Cells(w1.Rows.Count, 1).End(xlUp).Row
B = w2.Cells(w2.Rows.Count, 1).End(xlUp).Row
For Each aCell In w1.Range("A2:A" & A)
On Error Resume Next
rows = w2.Columns("A:A").Find(What:=aCell, LookAt:=xlWhole).Row
On Error GoTo 0
If rows = 0 Then
Else
w2.Range("B" & rows).Value = aCell.Offset(0, 1).Value
End If
rows = 0
Next
End Sub
Find supports wildcards *. Use Left$ to get the first 6 characters of the cell value.
For Each aCell In w1.Range("A2:A" & A)
Dim rng As Range
Set rng = w2.Columns("A:A").Find(What:=Left$(aCell.Value, 6) & "*", LookAt:=xlWhole)
' First test if the find succeeded
If Not rng Is Nothing Then
rng.Offset(,1).Value = aCell.Offset(,1).Value
End If
Next
My question is in the title. I have searched up everywhere and this one feels like the only answer that is working:
https://stackoverflow.com/a/15180079/17038705
I have created a sample Excel file and validated that his VBA code works, the sample he shows looks like it is working too. However, when I ran it with the Excel file I am working on, I got Error 91, Object variable or With block variable not set.
After some digging, it is probably because of his Find() function that returns Nothing.
My question is why this is the case for my file and not for others. The values there are based on formulas and values of other cells, would that be a problem?
Other approaches are appreciated as well. Thanks!
Since your data contains formulas, you need to set the LookIn parameter to xlValues in the Find method. I updated the original code with these changes, take a look:
Sub Highlight_Duplicate_Entry()
Dim ws As Worksheet
Dim cell As Range
Dim myrng As Range
Dim clr As Long
Dim lastCell As Range
Set ws = ThisWorkbook.Sheets("Sheet1")
Set myrng = ws.Range("A2:D" & Range("A" & ws.Rows.Count).End(xlUp).Row)
With myrng
Set lastCell = .Cells(.Cells.Count)
End With
myrng.Interior.ColorIndex = xlNone
clr = 3
For Each cell In myrng
If Application.WorksheetFunction.CountIf(myrng, cell) > 1 Then
' addresses will match for first instance of value in range
'[================]
If myrng.Find(what:=cell, LookIn:=xlValues, lookat:=xlWhole, MatchCase:=False, after:=lastCell).Address = cell.Address Then
' set the color for this value (will be used throughout the range)
cell.Interior.ColorIndex = clr
clr = clr + 1
Else
' if not the first instance, set color to match the first instance
'[================]
cell.Interior.ColorIndex = myrng.Find(what:=cell, LookIn:=xlValues, lookat:=xlWhole, MatchCase:=False, after:=lastCell).Interior.ColorIndex
End If
End If
Next
End Sub
A slightly different approach using a Dictionary to track values vs. colors:
Sub Tester()
ColorDups Range("A1").CurrentRegion
End Sub
Sub ColorDups(rng As Range)
Dim c As Range, dict As Object, i As Long, v
Set dict = CreateObject("scripting.dictionary")
i = 0
Application.ScreenUpdating = False
For Each c In rng.Cells
v = CStr(c.Value)
If Len(v) > 0 Then
If Not dict.exists(v) Then
dict.Add v, c 'store the first cell with this value
Else
If TypeOf dict(v) Is Range Then 'second cell with this value?
i = i + 1 'next index
dict(v).Interior.ColorIndex = i 'color the first cell
dict(v) = i 'store the index
End If
c.Interior.ColorIndex = dict(v) 'color this duplicate
End If
End If
Next c
End Sub
I am trying to delete duplicate values in a temporary list based on a value in a combobox. The code below loops through individual rows to check whether a value matches. It is slow.
Dim ws As Worksheet
Dim i As Long
Set ws = Sheets("TempList3")
On Error Resume Next
For i = Cells(Rows.Count, 2).End(xlUp).Row To 1 Step -1
If Cells(i, 2) <> Sheets("Sheet1").ComboBox2.Value Then
ws.Rows(i).EntireRow.Delete
End If
Next
Is there a way to evaluate the entire column's values against the combobox's value once and then delete all rows on a worksheet. Or perhaps there is a better way?
I used a looping Find function, it deletes the row where the value was found and then it searches again and deletes the next row it finds until it can no longer find the Combo value on the sheet:
Sub find_cell()
Dim find_cell As Range
Set ws = Sheets("TempList3")
stop_loop = False
Do Until stop_loop = True
Set find_cell = ws.Cells.Find(What:=Sheets("Sheet1").ComboBox2.Value, LookAt:=xlWhole)
If Not find_cell Is Nothing Then
ws.Rows(find_cell.Row).EntireRow.Delete
Else
stop_loop = True
End If
Loop
End Sub
Not knowing how many rows you are talking about, I used 10 thousand for my example codes. here are two examples, try the both and see what works best for you.
You can run through the column and unionize the range found, then delete the rows, for example.
See here for example workbook
Sub UnIonRng()
Dim FrstRng As Range
Dim UnIonRng As Range
Dim c As Range, s As String
s = Sheets("Sheet1").ComboBox2
Set FrstRng = Range("B:B").SpecialCells(xlCellTypeConstants, 23)
For Each c In FrstRng.Cells
If c = s Then
If Not UnIonRng Is Nothing Then
Set UnIonRng = Union(UnIonRng, c) 'adds to the range
'MsgBox UnionRng.Address 'remove later
Else
Set UnIonRng = c
End If
End If
Next c
UnIonRng.EntireRow.Delete
End Sub
Or you can try to filter the column B and delete the rows that way:
Sub FilterDeleteRow()
Dim ws As Worksheet
Dim LstRw As Long, Rng As Range, s As String, x
Set ws = Sheets("TempList3")
s = Sheets("Sheet1").ComboBox2
Application.ScreenUpdating = 0
With ws
LstRw = .Cells(.Rows.Count, "B").End(xlUp).Row
x = Application.WorksheetFunction.CountIf(.Range("B:B"), s)
If x > 0 Then
Columns("B:B").AutoFilter Field:=1, Criteria1:=s
Set Rng = .Range("B2:B" & LstRw).SpecialCells(xlCellTypeVisible)
Rng.EntireRow.Delete
.AutoFilterMode = 0
Else: MsgBox "Not Found"
End If
End With
End Sub
I have two spreadsheets, vda.xlsx and main.xlsm. At the moment I'm comparing the values in:
main.xlsm column J
with
vda.xlsx column A
To see if there is a match. If a match is found then the value in column gets highlighted in red.
However the format of the data in vda.xlsx column A has changed .
It used to look like this
1234
Now it looks like this
Test\1234 or Best\1234 or Jest\1234 - it could be anything...
Sp I need to split Test\1234 by the "\" and extract 1234 for comparison.
Any idea how I can accomplish this. This is my code so far:
Sub VDA_Update()
Dim wshT As Worksheet
Dim wbk As Workbook
Dim wshS As Worksheet
Dim r As Long
Dim m As Long
Dim cel As Range
Application.ScreenUpdating = False
Set wshT = ThisWorkbook.Worksheets("Master")
On Error Resume Next
' Check whether vda.xlsx is already open
Set wbk = Workbooks("vda.xlsx")
On Error GoTo 0
If wbk Is Nothing Then
' If not, open it
Set wbk = Workbooks.Open("C:\Working\vda_test.xlsx")
End If
' Set worksheet on vda.xlsx
Set wshS = wbk.Worksheets("imac01")
m = wshT.Cells(wshT.Rows.Count, 1).End(xlUp).Row
' Loop though cells in column J on main.xlsm
For r = 1 To m
' Can we find the value in column C of vda.xlsx?
Set cel = wshS.Columns(1).Find(What:=wshT.Cells(r, 10).Value, _
LookAt:=xlWhole, MatchCase:=False)
If Not cel Is Nothing Then
' If we find a match, then change the text to red
wshT.Cells(r, 10).Font.ColorIndex = 3
End If
Next r
Application.ScreenUpdating = True
End Sub
Use Split(CellValue, "\") to get an array and then retrieve the last item in the array.
Change:
' Loop though cells in column J on main.xlsm
For r = 1 To m
' Can we find the value in column C of vda.xlsx?
Set cel = wshS.Columns(1).Find(What:=wshT.Cells(r, 10).Value, _
LookAt:=xlWhole, MatchCase:=False)
If Not cel Is Nothing Then
' If we find a match, then change the text to red
wshT.Cells(r, 10).Font.ColorIndex = 3
End If
Next r
To something like:
' Loop though cells in column A on vda.xlsx
For r = 1 To m
' Can we find the value in column J of main.xlsm?
cellSplit = Split(wshS.Cells(r, 1).Value, "\")
Set cel = wshT.Columns(10).Find(cellSplit(UBound(cellSplit)), _
LookAt:=xlWhole, MatchCase:=False)
If Not cel Is Nothing Then
' If we find a match, then change the text to red
cel.Cells(1, 1).Font.ColorIndex = 3
End If
Next r