asked on

searching a bi-dim array

I have a bi-dim array say with 3 column and n rows.

I need to conduct a search in that array to see if there is a row which values match to 3 values I have say 'cc', 'ff', 'ddd' x each column

Thanks

ASKER CERTIFIED SOLUTION

Guy Hengel [angelIII / a3]

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

TimCottee

So something like:

Dim intRow As Integer
For intRow = LBound(aryMyArray(0)) To UBound(aryMyArray(0))
If aryMyArray(0,intRow) = "cc" And aryMyArray(1,intRow) = "dd" And aryMyArray(2,intRow) = "ddd" Then
Msgbox "Found a match at row " & CStr(intRow)
Exit For
End If
Next

jhsd

ASKER

thanx a lot also to TimCottee .
Kind angelll, I allow myself to profite of your knowledge to try to grab from u another quick answer :)
The thing is the following:
I load a di-dim array.
then when I go looping say:

Dim i As Integer
Dim h As Integer
'1 le righe

Dim str As String

For i = LBound(e, 1) To UBound(e, 1)
For y = LBound(e, 2) To UBound(e, 2)

str = str & "mystr" & (e(i, y)) & vbCrLf

Next y
Next i
MsgBox (str)

when I msgbox or debug.Print, str =

mystr*
mystr*
mystr*
mystr*
mystr*DE
mystr*01
mystr*cdr1
mystr*
mystr*DE
mystr*02
mystr*cdr1
mystr*
mystr*DE
mystr*01
mystr*cdr2
mystr*
mystr*DE
mystr*02
mystr*cdr2
mystr*

that is u can see there some void slots.

But if I do

For i = LBound(e, 1) To UBound(e, 1) -1
For y = LBound(e, 2) To UBound(e, 2) -1

str = str & "mystr" & (e(i, y)) & vbCrLf

Next y
Next i

It seems I get right:

mystr*DE
mystr*01
mystr*cdr1
mystr*DE
mystr*02
mystr*cdr1
mystr*DE
mystr*01
mystr*cdr2
mystr*DE
mystr*02
mystr*cdr2

and this seems that what I wanted because I don't see 'bugs' in the slots.

My question is: I am doing ok, or by adding - 1 and -1
I am going to force something it is actually not right:
i.e.: there r some bug in the charging operation of the array.

Thanks x your precipous advive! Best Regards

Guy Hengel [angelIII / a3]

I guess that the charging of the array has a bug in the loop.
As you start sizing the arrays with lower bound of 0, you must take care of this when increasing the size. I did the same error also in the beginning, now my loops to charge an array look like this:

dim lngArraySize as long
dim arrData()

lngArraySize = 0

'assume i read a recordset
while not rs.eof
lngArraySize = lngArraySize + 1
redim preserve arrData(1 to lngArraySize)
arrData(lngArraySize) = rs.Fields("data").value
rs.movenext
wend

What is important here is that the array is resized 1-based, and lngArraySize has the actual size of the array. So later on, when I want to loop the array, i do use lngArraySize instead of ubound(arrData), which is also much faster.

Now, to be honest, for tuning reasons (redim preserve is SLOW), i do even different:

dim lngArraySize as long
dim lngArrayMaxSize as long
dim arrData()

lngArraySize = 0
lngArrayMaxSize = 0

'assume i read a recordset
while not rs.eof
lngArraySize = lngArraySize + 1
if lngArraySize > lngArrayMaxSize
lngArrayMaxSize = lngArrayMaxSize + 10
redim preserve arrData(1 to lngArrayMaxSize)
end if

arrData(lngArraySize) = rs.Fields("data").value
rs.movenext
wend

This means that the array is larger than it has values, but the loop is faster. The more important it is to keep the number of actual items in a variable for later processing.

CHeers