7
\$\begingroup\$

I have this Excel/VBA code and here is what it does:

  1. I have 3 sheets, where first sheet has two columns that will be used (A,C).
  2. 2nd Sheet is just the raw data file that will be imported from external source.
  3. 3rd sheet is the final output.
  4. Column A in the first sheet is "Raw Column Headings" where the column headers from raw data table are copied, transposed, and pasted. Third column in the first sheet is the final headers that I need the raw ones to be changed to. Then, I will run a macro which will copy the final headers into the first row of second sheet and then copy/paste the entire columns with the certain headers specified in the code to the final output.

The problem is that this process takes about 20~30 seconds per file and I have so many files to be processed. Can anyone take a look at the code and tell me how it can be done faster?

Option Explicit
Private Function GetHeaders() As Collection

    Dim result As New Collection
    With result
        .Add "Account_ID"
        .Add "Claim_ID"
        .Add "Account_Name"
        .Add "Claim_Type"
        .Add "Coverage"
        .Add "Claim_Level"
        .Add "Claim_Count"
        .Add "File_Date"
        .Add "File_Year"
        .Add "Resolution_Date"
        .Add "Resolution_Year"
        .Add "Claim_Status"
        .Add "Indemnity_Paid"
        .Add "Disease_Category"
        .Add "State_Filed"
        .Add "First_Exposure_Date"
        .Add "Last_Exposure_Date"
        .Add "Claimant_Employee"
        .Add "Claimant_DOB"
        .Add "Claimant_Deceased"
        .Add "Claimant_Name"
        .Add "Claimant_DOD"
        .Add "Claimant_Diagnosis_Date"
        .Add "Product_Type"
        .Add "Product_Line"
        .Add "Company/Entity/PC"
        .Add "Plaintiff_Law_Firm"
        .Add "Asbestos_Type"
        .Add "Evaluation_Date"
        .Add "Tier"
        .Add "Data_Source"
        .Add "Data_Source_Category"
        .Add "Jurisdiction/County"
        .Add "Settlement_Demand"
        .Add "Jury_Verdict"
        .Add "Exposure_Site"
        .Add "National_Defendant_Firm"
        .Add "Local_Defendant_Firm"
        .Add "Expense_Amount"
        .Add "NCC_Expense_Amount"
        .Add "Non_NCC_Expense_Amount"
    End With

    Set GetHeaders = result
End Function

Private Function FindHeaderRange(ByVal ws As Worksheet, ByVal header As String) As Range
    Set FindHeaderRange = ws.Cells.Find(header, , , xlWhole)
End Function

Private Function BuildMessage(ByVal currentMessage As String, ByVal ws As Worksheet, ByVal header As String) As String
    BuildMessage = currentMessage & vbLf & header & Space(1) & ws.Name
End Function


Public Sub ProjectionTemplateFormat()
On Error GoTo ExitSub

    Sheets(1).Range("C2", Cells(Rows.Count, "C").End(xlUp)).Copy
    Sheets(2).Range("A1").PasteSpecial transpose:=True
    Range("A1").ClearOutline

    Dim headers As Collection
    Set headers = GetHeaders

    Dim msg As String

    Dim wsImport As Worksheet, wsMain As Worksheet

    Set wsImport = ThisWorkbook.Sheets(2)
    Set wsMain = ThisWorkbook.Sheets(3)

    Application.ScreenUpdating = False

    Dim header As Variant
    Dim source As Range
    Dim dest As Range
    For Each header In headers

        Set source = FindHeaderRange(wsImport, header)
        If source Is Nothing Then
            msg = BuildMessage(msg, wsImport, header)
        Else

            Set dest = FindHeaderRange(wsMain, header)
            If dest Is Nothing Then
                msg = BuildMessage(msg, wsMain, header)
            Else
                wsImport.Range(source.Offset(1), wsImport.Cells(Rows.Count, source.Column).End(xlUp)).Copy _
                    wsMain.Cells(Rows.Count, dest.Column).End(xlUp)(2)
            End If
        End If
    Next



With wsMain
    .Columns("A:AO").AutoFit
    .Cells.ClearFormats
    .Rows(1).Font.Bold = True
    .Cells.Font.Name = "Georgia"
    .Cells.Font.Color = RGB(0, 0, 225)
    .Cells.Resize(wsMain.Rows.Count - 1).Offset(1).Interior.Color = RGB(216, 228, 188)


End With

'Apply Style

Dim ws As Worksheet
    For Each ws In Worksheets
        ws.Select
        ActiveWindow.Zoom = 85
    Next ws

Dim rng As Range
Set rng = wsMain.Cells
With rng.Borders
    .LineStyle = xlDot
    .Weight = xlThin
End With

Dim cell As Range
    For Each cell In Range("$A$2:" & Range("$A$2").SpecialCells(xlLastCell).Address)
        If Len(cell) > 0 Then cell = UCase(cell)
    Next cell

ExitSub:
    Application.ScreenUpdating = True

End Sub
\$\endgroup\$
  • \$\begingroup\$ How big is the data set? \$\endgroup\$ – RubberDuck Jun 4 '15 at 17:07
2
\$\begingroup\$

Previous answer is getting a bit long, so here is the rest of the code and the flow I'd suggest:

.

Option Explicit

Public Sub mainProcess()

    ProjectionTemplateFormat

End Sub

Private Function GetHeaders() As Collection
    '...
End Function

Private Sub ProjectionTemplateFormat()

    GetHeaders

    '...

    applyFormat ThisWorkbook.Worksheets(1)

End Sub

Private Sub applyFormat(ByRef sh As Worksheet)

    With sh.UsedRange

        .Columns.AutoFit
        .ClearFormats

        With .Font
            .Name = "Georgia"
            .Color = RGB(0, 0, 225)
        End With
        With .Borders
            .LineStyle = xlDot
            .Weight = xlThin
        End With
        .Interior.Color = RGB(216, 228, 188)

        With .Rows(1)
            .Font.Bold = True
            .ColorIndex = xlAutomatic
        End With

    End With

    Dim ws As Worksheet
    For Each ws In Worksheets
        ws.Zoom = 85
    Next

    allUpper1 sh   'or allUpper2 sh

End Sub

Private Sub allUpper1(ByRef sh As Worksheet) 'FOR loop, with VBA UCase()

    Dim arr As Variant, i As Long, j As Long

    If WorksheetFunction.CountA(sh.UsedRange) > 0 Then
        arr = sh.UsedRange  'one interaction with the sheet
        For i = 2 To UBound(arr, 1)         'each "row"
            For j = 1 To UBound(arr, 2)     'each "col"
                arr(i, j) = UCase(arr(i, j))
            Next
        Next
        sh.UsedRange = arr  'second interaction with the sheet
    End If
End Sub

Private Sub allUpper2(ByRef sh As Worksheet) 'No loop, with Excel UPPER()

    Const FIRST_ROW As Long = 2

    Dim lRow    As Long
    Dim lCol    As Long
    Dim usedRng As Range
    Dim tempRng As Range

    If WorksheetFunction.CountA(sh.UsedRange) > 0 Then
        Set usedRng = sh.UsedRange
        With usedRng
            lRow = .Rows.Count
            lCol = .Columns.Count
        End With
        'remove header row from working range
        Set usedRng = usedRng.Offset(1, 0).Resize(lRow - 1, lCol)
        'offset cell: 2 collumns to the right of 1st cell in used range
        Set tempRng = sh.Cells(FIRST_ROW, lCol + 2)
        With tempRng

            'apply formula to offset cell: UPPER(A2)
            .Formula = "=Upper(" & sh.Cells(FIRST_ROW, 1).Address(0, 0) & ")"

            'fill down the column
            .AutoFill Destination:=sh.Range(tempRng, .Offset(lRow, 0)), Type:=xlFillDefault
            'fill right all rows
            sh.Range(tempRng, .Offset(lRow, 0)).AutoFill Destination:=sh.Range(tempRng, .Offset(lRow, lCol)), Type:=xlFillDefault

            With sh.Range(tempRng, .Offset(lRow, lCol))
                usedRng.Value2 = .Value2    'copy upper case values back
                .EntireColumn.Delete        'remove temp range
            End With
        End With
    End If
End Sub

.

As a performance reference - measurement between UCase(), and UPPER() methods:

FOR loop method:       6.61 seconds
Excel formula method: 15.29 seconds

With 100,000 rows, and 26 columns

Most text cells contain "Test Cell 1", "Test Cell 2", "Test Cell 3",..., and 11 cells with numbers, 5 in the first few rows and 6 in the last rows

\$\endgroup\$
  • 1
    \$\begingroup\$ This really would have been better in the other answer... :-/ Too late now: this one got accepted, and the other one has the votes. It's a mess that could have been avoided. Next time, please use just one answer if they are really about the same line of logic. \$\endgroup\$ – janos Sep 23 '15 at 14:00
  • \$\begingroup\$ @janos - Noted and thanks (this was my first post on CR, and I won't make this mistake again) \$\endgroup\$ – paul bica Sep 23 '15 at 15:01
8
\$\begingroup\$

Your code has tremendously improved since the first time I saw it - good job!


This particular line is hard to parse:

wsImport.Range(source.Offset(1), wsImport.Cells(Rows.Count, source.Column).End(xlUp)).Copy _
                                 wsMain.Cells(Rows.Count, dest.Column).End(xlUp)(2)

Literally: it's crashing the parser!

You could introduce a local variable here:

Dim target As Range
Set target = wsMain.Cells(Rows.Count, dest.Column).End(xlUp)

wsImport.Range(source.Offset(1), wsImport.Cells(Rows.Count, source.Column).End(xlUp)) _
        .Copy target(RowIndex:=2)

Notice how the line continuation is placed so that no instruction is split; doesn't care about line continuations, but it's much easier for the human eye to see what function calls return the arguments for which procedure if you don't split an instruction between the name of a procedure and its arguments - vertically lining up .Range and .Copy also make it clearer that .Copy operates on the result of .Range.


Dim wsImport As Worksheet, wsMain As Worksheet

Is this really buying you anything? Multiple declarations on a single line make it harder to locate declarations at a glance. Compare to:

Dim wsImport As Worksheet
Dim wsMain As Worksheet

My eye sees Dim, my brain sees "variable declaration here" - two Dims, two variables. And I read the variable name at pretty much the exact same millisecond as the one I notice the Dim statement, because I don't need to mentally scroll horizontally and locate the comma. Two variables isn't too bad, but more than that would be problematic. Better avoid multiple declarations on a single line.


There's a redundant reference to wsMain in this With block:

With wsMain
    .Columns("A:AO").AutoFit
    .Cells.ClearFormats
    .Rows(1).Font.Bold = True
    .Cells.Font.Name = "Georgia"
    .Cells.Font.Color = RGB(0, 0, 225)
    .Cells.Resize(wsMain.Rows.Count - 1).Offset(1).Interior.Color = RGB(216, 228, 188)


End With

See it? Right here:

.Cells.Resize(wsMain.Rows.Count - 1).Offset(1).Interior.Color = RGB(216, 228, 188)

Could be

.Cells.Resize(.Rows.Count - 1).Offset(1).Interior.Color = RGB(216, 228, 188)

The extraneous empty lines before End With should be removed, too.


Your indentation isn't consistent.

Dim ws As Worksheet
    For Each ws In Worksheets
        ws.Select
        ActiveWindow.Zoom = 85
    Next ws

The only thing that should be at the same indentation level as Public Sub/End Sub, is line labels (which the VBE forces to start at column 1 anyway).

\$\endgroup\$
4
\$\begingroup\$

@Mat's Mug made good suggestions (all), but in particular: redundant or implicit references:

Sheets(1).Range("C2", Cells(Rows.Count, "C")

Try to always be explicit as possible - the above line has one (semi) explicit reference Sheets(1) and 3 implicit ones:

  • Rows.Count implies "ActiveSheet.Cells.Rows.Count"
  • Cells(Rows.Count, "C") = ActiveSheet.Cells(Rows.Count, "C")
  • Sheets(1) should be ThisWorkbook.Worksheets(1)

All explicit:

Sheets(1).Range("C2", Sheets(1).Cells(Sheets(1).Cells.Rows.Count, "C")

Or

With Sheets(1)
    .Range("C2", .Cells(.Cells.Rows.Count, "C")
End With

As Mat pointed out: Sheets(1) is as also an implicit reference to ThisWorkbook.Worksheets(1)

So the complete reference is

With ThisWorkbook.Worksheets(1)
    .Range("C2", .Cells(.Cells.Rows.Count, "C")
End With

A few suggestions that might improve performance:

  1. In function FindHeaderRange(), replace the Find method with Match.

    From

    Set FindHeaderRange = ws.Cells.Find(header, , , xlWhole)
    
    ...
    
    Set source = FindHeaderRange(wsImport, header)
    If source Is Nothing Then...
    

    To

    FindHeaderRange = Application.Match(header, ws.UsedRange.Rows(1), 0)
    

    FindHeaderRange()'s return type will change from Range to Variant - Match returns a Double if it finds something and error if not, so checking the result changes to:

    If Not IsError(FindHeaderRange) Then ...
    

    (if ok, FindHeaderRange will be 3, 4, 99, etc because we're searching only Row 1)

    Match is significantly faster than Find.

  2. In function BuildMessage(): replace Space(1) with " ".

    Space(1) is a function call not worth calling for just one space.

  3. Collections (and dictionaries) are great in reducing the amount of code, and retrieving items, but are quite slow at loading data.

    When using the "headers" collection you're not reducing code too much, compared to "headers" as an array; arrays are very fast in both loading and retrieving; more complex code can get convoluted and hard to follow with multidimensional arrays, but in your case a 1-dimensional array would not change the structure: For Each element in Collection = For Each Element in array.

  4. Your code is very well organized, modularized, and easy to read - great work! However, there is a price to pay for over-modularizing.

    Function are meant to isolate blocks of code for easier maintenance and to make logical distinction between different functionalities, but calls to functions can get expensive, especially when they're very frequent and have a small number of lines; in-line code is much faster to execute than the extra memory navigation to the function called.

    Your 2 small (and neat) functions FindHeaderRange() and BuildMessage() can be executed directly, without negatively impacting readability.

Another important performance improvement:

The most time consuming line in your For loop:

wsImport.Range(source.Offset(1), wsImport.Cells(Rows.Count, source.Column).End(xlUp)).Copy _
                    wsMain.Cells(Rows.Count, dest.Column).End(xlUp)(2)

This is meant to copy and paste the used range part of the current column.

Finding current column used range can be optimized (and explicit):

With wsImport
    Set fromCol = .Range( source.Offset(1), .Cells(.UsedRange.Rows.Count + 1, source.Column).End(xlUp))
End With

With wsMain
    Set toCol = .Cells(.UsedRange.Rows.Count + 1, dest.Column).End(xlUp)(2)
End With

fromCol.Copy toCol

Do you need to dynamically find the used range in wsMain in order to remove previous data?

If so, it would be faster to delete the contents of the entire column:

wsMain.UsedRange.Column(dest.Column).Value2 = vbNullString

Then, if all columns in wsImport are the same size you can get the last used row before the For loop (just once) and use that value inside the loop.

Edit: tested version of the code:

.

Option Explicit

Public Sub projectionTemplateFormat()
  Dim t1 As Double, t2 As Double

  xlSpeed True
    t1 = Timer
    mainProcess
    t2 = Timer
  xlSpeed False
  MsgBox "Duration: " & t2 - t1 & " seconds"
End Sub

Private Sub mainProcess()
  Const SPACE_DELIM     As String = " "
  Dim wsIndex           As Worksheet
  Dim wsImport          As Worksheet  'Raw
  Dim wsFinal           As Worksheet
  Dim importHeaderRng   As Range
  Dim importColRng      As Range
  Dim importHeaderFound As Variant
  Dim importLastRow     As Long
  Dim finalHeaderRng    As Range
  Dim finalColRng       As Range
  Dim finalHeaderRow    As Variant
  Dim finalHeaderFound  As Variant
  Dim indexHeaderCol    As Range
  Dim header            As Variant  'Each item in the FOR loop
  Dim msg               As String

  Set wsIndex = aIndex    'This is the Code Name; top-left pane: aIndex (Index)
  Set wsImport = bImport  'Direct reference to Code Name: bImport.Range("A1")
  Set wsFinal = cFinal    'Reference using Sheets collection: ThisWorkbook.Worksheets("Final")

  With wsImport.UsedRange
    Set importHeaderRng = .Rows(1)  'Import - Headers
    importLastRow = .Rows.Count     'Import - Total Rows
  End With
  With wsFinal.UsedRange
    finalHeaderRow = .Rows(1)       'Final - Headers (as Array)
    Set finalHeaderRng = .Rows(1)   'Final - Headers (as Range)
  End With
  With wsIndex.UsedRange            'Transpose col 3 from Index (without the header), as column names in Import
    Set indexHeaderCol = .Columns(3).Offset(1, 0).Resize(.Rows.Count - 1, 1)
    wsImport.Range(wsImport.Cells(1, 1), wsImport.Cells(1, .Rows.Count - 1)).Value2 = Application.Transpose(indexHeaderCol)
  End With
  If Len(bImport.Cells(2, 2).Value2) > 0 Then 'if Import sheet is not empty (excluding header row)
    With Application
      For Each header In finalHeaderRow   'Loop through all headers in Final
        If Len(Trim(header)) > 0 Then     'If the Final heade is not empty
          importHeaderFound = .Match(header, importHeaderRng, 0)    'Find header in Import sheet
          If IsError(importHeaderFound) Then
            msg = msg & vbLf & header & SPACE_DELIM & wsImport.Name 'Import doesn't have current header
          Else
            finalHeaderFound = .Match(header, finalHeaderRng, 0)    'Find header in Final sheet
            With wsImport
              Set importColRng = .UsedRange.Columns(importHeaderFound).Offset(1, 0).Resize(.UsedRange.Rows.Count - 1, 1)
            End With
            With wsFinal
              Set finalColRng = .Range(.Cells(2, finalHeaderFound), .Cells(importLastRow, finalHeaderFound))
              finalColRng.Value2 = vbNullString       'Delete previous values (entire column)
            End With
            finalColRng.Value2 = importColRng.Value2  'Copy Import data in Final columns
          End If
        End If
      Next header
    End With
    allUpper wsFinal
    'wsFinal.UsedRange.AutoFilter
    applyFormat wsFinal.Range(wsFinal.Cells(1, 1), wsFinal.Cells(importLastRow, wsFinal.UsedRange.Columns.Count))
    Dim ws As Worksheet
    For Each ws In Worksheets
      ws.Activate
      ActiveWindow.Zoom = 85
      ws.Cells(2, 2).Activate
      ActiveWindow.FreezePanes = True
      ws.Cells(1, 1).Activate
    Next
  Else
    MsgBox "Missing raw data (Sheet 2 - 'Import')", vbInformation, "   Missing Raw Data"
  End If
End Sub

Private Sub applyFormat(ByRef rng As Range)
  With rng
    .ClearFormats
    With .Font
      .Name = "Georgia"
      .Color = RGB(0, 0, 225)
    End With
    .Interior.Color = RGB(216, 228, 188)
    With .Rows(1)
      .Font.Bold = True
      .Interior.ColorIndex = xlAutomatic
    End With
    With .Borders
      .LineStyle = xlDot  'xlContinuous
      .ColorIndex = xlAutomatic
      .Weight = xlThin
    End With
  End With
  refit rng
End Sub

Private Sub allUpper(ByRef sh As Worksheet)
  Dim arr As Variant, i As Long, j As Long

  If WorksheetFunction.CountA(sh.UsedRange) > 0 Then
    arr = sh.UsedRange
    For i = 2 To UBound(arr, 1)    'each "row"
      For j = 1 To UBound(arr, 2)  'each "col"
        arr(i, j) = UCase(RTrim(Replace(arr(i, j), Chr(10), vbNullString)))
      Next
    Next
    sh.UsedRange = arr
  End If
End Sub

Public Sub allImportTrim()
  Dim arr As Variant, i As Long, j As Long, sh As Worksheet

  Set sh = bImport
  If WorksheetFunction.CountA(sh.UsedRange) > 0 Then
    arr = sh.UsedRange
    For i = 2 To UBound(arr, 1)    'each "row"
      For j = 1 To UBound(arr, 2)  'each "col"
        arr(i, j) = RTrim(Replace(arr(i, j), Chr(10), vbNullString))
      Next
    Next
    sh.UsedRange = arr
  End If
  refit sh.UsedRange
End Sub

Private Sub refit(ByRef rng As Range)
  With rng
    .WrapText = False
    .HorizontalAlignment = xlGeneral
    .VerticalAlignment = xlCenter
    .Columns.EntireColumn.AutoFit
    .Rows.EntireRow.AutoFit
  End With
End Sub
\$\endgroup\$
  • \$\begingroup\$ Nice answer. Welcome to Code Review! We'll have to disagree about the call to Space though. I think it offers readability over the literal. \$\endgroup\$ – RubberDuck Jun 6 '15 at 22:55
  • \$\begingroup\$ Welcome to CR! That is a beautiful answer, I hope you stick around! \$\endgroup\$ – Mathieu Guindon Jun 6 '15 at 23:03
  • \$\begingroup\$ Ahh. I just noticed that you used A constant. Nice choice! \$\endgroup\$ – RubberDuck Jun 6 '15 at 23:29
  • \$\begingroup\$ @RubberDuck and Mat - thank you both (or all 3 - don't know who else upvoted) - I've already noticed your work and it's very impressive ! \$\endgroup\$ – paul bica Jun 7 '15 at 0:12
  • 2
    \$\begingroup\$ I count Sheet1(1) as an implicit reference to ThisWorkbook.Sheets, which is probably intended to be ThisWorkbook.Worksheets. \$\endgroup\$ – Mathieu Guindon Jun 7 '15 at 20:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.