[vba] What does the Excel range.Rows property really do?

OK, I am finishing up an add-on project for a legacy Excel-VBA application, and I have once again run up against the conundrum of the mysterious range.Rows (?) and worksheet.Rows properties.

Does anyone know what these properties really do and what they are supposed to provide to me? (Note: all of this probably applies to the corresponding *.Columns properties also).

What I would really like to be able to use it for is to return a range of rows, like this:

   SET rng = wks.Rows(iStartRow, iEndRow)

But I have never been able to get it to do that, even though the Intellisense shows two arguments for it. Instead I have to use one of the two or three other (very kludgy) techniques.

The help is very unhelpful (typically so for Office VBA), and googling for "Rows" is not very useful, no matter how many other terms I add to it.

The only things that I have been able to use it for are 1) return a single row as a range ( rng.Rows(i) ) and 2) return a count of the rows in a range ( rng.Rows.Count ). Is that it? Is there really nothing else that it's good for?

Clarification: I know that it returns a range and that there are other ways to get a range of rows. What I am asking for is specifically what do we get from .Rows() that we do not already get from .Cells() and .Range()? The two things that I know are 1) an easier way to return a range of a single row and 2) a way to count the number of rows in a range.

Is there anything else?

This question is related to vba excel ole-automation

The answer is


I'm not sure, but I think the second parameter is a red herring.

Both .Rows and .Columns take two optional parameters: RowIndex and ColumnIndex. Try to use ColumnIndex, e.g. Rows(ColumnIndex:=2), generates an error for both .Rows and .Columns.

My feeling it's inherited in some sense from the Cells(RowIndex,ColumnIndex) Property but only the first parameter is appropriate.


Since the .Rows result is marked as consisting of rows, you can "For Each" it to deal with each row individually, like this:

Function Attendance(rng As Range) As Long
Attendance = 0
For Each rRow In rng.Rows
    If WorksheetFunction.Sum(rRow) > 0 Then
        Attendance = Attendance + 1
    End If
Next
End Function

I use this to check attendance in any of a few categories (different columns) for a list of people (different rows).

(And of course you could use .Columns to do a "For Each" over the columns in the range.)


Your two examples are the only things I have ever used the Rows and Columns properties for, but in theory you could do anything with them that can be done with a Range object.

The return type of those properties is itself a Range, so you can do things like:

Dim myRange as Range
Set myRange = Sheet1.Range(Cells(2,2),Cells(8,8))
myRange.Rows(3).Select

Which will select the third row in myRange (Cells B4:H4 in Sheet1).

update: To do what you want to do, you could use:

Dim interestingRows as Range
Set interestingRows = Sheet1.Range(startRow & ":" & endRow)

update #2: Or, to get a subset of rows from within a another range:

Dim someRange As Range
Dim interestingRows As Range

Set myRange = Sheet1.Range(Cells(2, 2), Cells(8, 8))

startRow = 3
endRow = 6

Set interestingRows = Range(myRange.Rows(startRow), myRange.Rows(endRow))

Range.Rows, Range.Columns and Range.Cells are Excel.Range objects, according to the VBA Type() functions:

?TypeName(Selection.rows)
Range
However, that's not the whole story: those returned objects are extended types that inherit every property and method from Excel::Range - but .Columns and .Rows have a special For... Each iterator, and a special .Count property that aren't quite the same as the parent Range object's iterator and count.

So .Cells is iterated and counted as a collection of single-cell ranges, just like the default iterator of the parent range.

But .Columns is iterated and counted as a collection of vertical subranges, each of them a single column wide;

...And .Rows is iterated and counted as a collection of horizontal subranges, each of them a single row high.

The easiest way to understand this is to step through this code and watch what's selected:

Public Sub Test() 
Dim SubRange As Range Dim ParentRange As Range
Set ParentRange = ActiveSheet.Range("B2:E5")

For Each SubRange In ParentRange.Cells SubRange.Select Next
For Each SubRange In ParentRange.Rows SubRange.Select Next
For Each SubRange In ParentRange.Columns SubRange.Select Next
For Each SubRange In ParentRange SubRange.Select Next
End Sub
Enjoy. And try it with a couple of merged cells in there, just to see how odd merged ranges can be.


There is another way, take this as example

Dim sr As String    
sr = "6:10"
Rows(sr).Select

All you need to do is to convert your variables iStartRow, iEndRow to a string.


It's perhaps a bit of a kludge, but the following code does what you seem to want to do:

Set rng = wks.Range(wks.Rows(iStartRow), wks.Rows(iEndRow)).Rows

I've found myself using range.Rows for its effects in the Copy method. It copies the height of the rows from the origin to the destination, which is the behaviour I want.

rngLastRecord.Rows.Copy Destination:=Sheets("Availability").Range("a" & insertRow)

If I had used rngLastRecord.Copy instead of rngLastRecord.Rows.Copy, the row heights would be whatever was there before the copy.


I've found this works:

Rows(CStr(iVar1) & ":" & CStr(iVar2)).Select