Reputation: 877
I am trying to make a Bill of Material list from my CAD software(Creo), this is exported out as a text file and its almost working perferct. The problem is that the text file spits out all parts in model, and I would like to count up duplicates instead of list them out after each other i.e quanties larger then 1.
My code does this nicely if the duplicate is listed in the row above, however if its nested(sub assy) this does not work. In picture below 00151564.asm(level 2) should be listed as quantity=3 but the parts on level 3 should be listed as quantity=1
Below is the original textfile creo spitts out, groups and patterns makes extra unwanteded indents, and material is just sometimes given..
Here is my code:
Sub simen2(Optional myFile As String = "Z:\Prosjekt\33907\Equipment and
materials\Structure\treetool2.txt")
Dim text As String
Dim textline As String
Dim textlineTemp As String
Dim foo As String
Dim output As String
Dim parent As String
Dim grandma As String
Dim greatgrandma As String
Dim greatgreatgrandma As String
Dim partNumber As String
Dim quantity As Integer
Dim material As String
Dim wsOut As Worksheet
Dim i, k As Long
Dim level, levelOld, levelTemp, levelTempOld, subtractLevel As Integer
Dim duplicate As Boolean
Dim levelDictionary As Object
'Init variables
Set wsOut = ThisWorkbook.Worksheets("Output")
subtractLevel = 0
quantity = 1
duplicate = True
partNumberOld = ""
commonNameOld = ""
levelOld = 0
levelTemp = 0
levelTempOld = 0
materialOld = ""
textlineOld = ""
material = "NA"
materialOld = "NA"
text = wsOut.Cells(1, 1).Value
wsOut.Cells.ClearContents
wsOut.Cells(1, 1).Value = text
wsOut.Cells(1, 2).Value = Now
wsOut.Cells(1, 3).Value = myFile
Call write2ExcelHeader(wsOut)
Set levelDictionary = CreateObject("Scripting.Dictionary")
i = 0
k = 1
FileNum = FreeFile()
Open myFile For Input As #FileNum
Line Input #FileNum, foo
Line Input #FileNum, foo
Do Until EOF(FileNum)
k = k + 1
' read in
Line Input #FileNum, textline
' Get level, however group and pattern fuck things up
If InStr(10, textline, "<HTML>") > 0 Or InStr(textline, "Pattern") > 0 Or InStr(textline, "Group") > 0 Then
levelTemp = getLevel(textline)
If levelTemp < levelTempOld Then
subtractLevel = 0
End If
If InStr(textline, "Pattern") > 0 Or InStr(textline, "Group") > 0 Then
subtractLevel = subtractLevel + 1
k = 1
End If
End If
' Grab material
If InStr(textline, "Materials") > 0 Then
Line Input #FileNum, textline
material = Trim(Replace(textline, "<curr>", ""))
End If
'we need to find out if the line has number as first item, i.e trim away spaces, it prints out previous item here...
If InStr(10, textline, "<HTML>") > 0 Then
'textlineTemp = RemoveHTML(textline)
'textlineTemp = Replace(textlineTemp, "Ã", "Ø")
partNumber = getPartNumber(textline)
commonName = getCommonName(textline)
partType = getType(partNumber)
material = "NA"
' add part to dictionary, this is unique parts
If levelDictionary.exists(partNumber) Then
levelDictionary(partNumber) = levelDictionary(partNumber) + 1
Else
levelDictionary.Add partNumber, 1
End If
' Remove duplicates ......
If partNumberOld = partNumber And levelTempOld = levelTemp Then
duplicate = True
quantity = quantity + 1
Else
quantity = 1
duplicate = False
End If
' get family history
level = levelTemp - subtractLevel
LevelArray(level) = partNumber
'lets present result
If Not duplicate Then
If level > 1 Then
parent = LevelArray(level - 1)
Else
parent = "NA"
End If
If level > 2 Then
grandma = LevelArray(level - 2)
Else
grandma = "NA"
End If
If level > 3 Then
greatgrandma = LevelArray(level - 3)
Else
greatgrandma = "NA"
End If
If level > 4 Then
greatgreatgrandma = LevelArray(level - 4)
Else
greatgreatgrandma = "NA"
End If
If i > 0 Then
Call write2Excel(wsOut, i + 2, partNumberOld, commonNameOld, quantityOld, materialOld, levelOld, partTypeOld, parentOld, grandmaOld, greatgrandmaOld, greatgreatgrandmaOld)
End If
i = i + 1
End If
End If
'we always uses previous values for print out
partNumberOld = partNumber
commonNameOld = commonName
levelOld = level
levelTempOld = levelTemp
partTypeOld = partType
quantityOld = quantity
materialOld = material
textlineOld = textline
parentOld = parent
grandmaOld = grandma
greatgrandmaOld = greatgrandma
greatgreatgrandmaOld = greatgreatgrandma
Loop
Close #FileNum
Debug.Print "How many parts " & i
Call DeList(wsOut)
Call CreateList(wsOut, "FilterOutput")
Call totalBOM(levelDictionary)
End Sub
Upvotes: 0
Views: 193
Reputation: 16392
An alternative OO approach using a class module. Output is to a sheet named "Output2".
Update 1 - Added debugging log, creo.log in same folder as workbook.
Option Explicit
Sub ProcessTextFile()
Const TXTFILE = "treetool_Rextroth.txt" '"treetool20210503.txt"
Const MAX_LEVEL = 10
Dim tree() As clsItem, item As clsItem
Dim ruler() As Integer, level As Integer, rs
Dim FileNum As Integer, textline As String, text As String
Dim start_name As Integer, width_name As Integer, n As Long
Dim t0 As Single: t0 = Timer
ReDim tree(MAX_LEVEL)
ReDim ruler(MAX_LEVEL)
FileNum = FreeFile()
Open ThisWorkbook.Path & "\" & TXTFILE For Input As #FileNum
' use first header line to get common name column position
Line Input #FileNum, textline
start_name = InStr(1, textline, "PTC_COMMON_NAME")
width_name = InStr(1, textline, "PRO_MP_") - start_name
' skip
Line Input #FileNum, textline
' set start level and indent
Set tree(0) = New clsItem
tree(0).level = 0
tree(0).id = "NA"
level = 1
ruler(1) = 1
' open log file
Dim fso, ts
Set fso = CreateObject("Scripting.Filesystemobject")
Set ts = fso.createtextfile("creo.log")
' scan text file
n = 2
Do Until EOF(FileNum)
n = n + 1
Line Input #FileNum, textline
rs = ParseLine(textline, n, ts)
If rs(0) = "ASM" Or rs(0) = "PRT" Then
' determine level from indent
level = GetLevel(ruler, rs(2), level, ts)
' create new item
Set item = New clsItem
With item
.itemtype = rs(0)
.id = rs(1)
.name = Mid(textline, start_name, width_name)
.qu = 1
.level = level
.parent = tree(level - 1).id
.creo = n
End With
' build tree
Set tree(level) = item
tree(level - 1).addItem item
' groups or patterns
ElseIf rs(0) = "GRP" Or rs(0) = "PTN" Then
' increase ruler for current level by 2
ruler(level) = ruler(level) + 2
ts.writeline n & " " & rs(0) & " change ruler(" & level & ")=" & ruler(level)
' materials
ElseIf rs(0) = "MTL" Then
If item.itemtype = "PRT" Then
' get material from next line
n = n + 1
Line Input #FileNum, textline
item.material = Trim(Replace(textline, "<curr>", ""))
End If
End If
Loop
' output tree
Application.ScreenUpdating = False
With Sheets("Output2")
text = .Range("A1")
.Cells.ClearContents
.Cells.Clear
.Range("A1") = text
.Range("B1") = Now
.Range("C1") = TXTFILE
With .Range("A2:H2")
.Value2 = Array("Part No", "Common Name", _
"Qu.", "Material", "Level", "Type", "Parent", "Creo Lines")
.Interior.Color = RGB(255, 200, 0)
.Font.Bold = True
End With
' save objects to sheet
tree(0).SaveToWorksheet .Range("A3")
' prettify
.Range("C:C,E:E").HorizontalAlignment = xlCenter
.Columns("H:H").HorizontalAlignment = xlRight
.Columns("A:H").AutoFit
.ListObjects.add(xlSrcRange, .UsedRange.Offset(1), , xlYes).name = "Table2"
.ListObjects("Table2").TableStyle = "TableStyleLight1"
.Activate
.Range("A1").Select
End With
Application.ScreenUpdating = True
MsgBox Format(n, "#,###") & " lines parsed in " & _
Format(Timer - t0, "0.00") & " seconds"
End Sub
' determine level from indent using ruler
Function GetLevel(ByRef ruler, indent, level, ts) As Integer
Dim n As Integer
n = level ' current level
' is this an increase on previous
If indent > ruler(n) Then
ts.writeline "GetLevel before ruler(" & n & ")=" & ruler(n)
n = n + 1
Else
' find previous level
n = 0
Do
n = n + 1
Loop While indent > ruler(n)
End If
ruler(n) = indent ' update
ts.writeline "Level now " & n & " ruler(" & n & ")=" & indent
GetLevel = n
End Function
' determine linetype, partno, indent
Function ParseLine(s As String, n, ts) As Variant
Dim indent As Integer, partno As String
Dim tmp As String, linetype As String
indent = Len(s) - Len(LTrim(s)) ' no of spaces
If InStr(1, s, "<HTML>") Then
tmp = WorksheetFunction.Trim(s)
partno = Split(tmp, " ")(0)
linetype = Right(partno, 3)
ts.writeline vbCrLf & n & " INDENT=" & indent & " '" & s
ElseIf InStr(1, s, "Materials") Then
linetype = "MTL"
ElseIf InStr(1, s, "Group") Then
linetype = "GRP"
ElseIf InStr(1, s, "Pattern") Then
linetype = "PTN"
End If
ParseLine = Array(linetype, partno, indent)
End Function
class module clsItem
========================
Option Explicit
' this class represent a part or assembly
Public id As String ' partno
Public name As String ' common name
Public itemtype As String ' PRT or ASM
Public parent As String
Public level As Integer
Public qu As Integer
Public material As String
Public creo As String ' source line nos in Creo file
Public items As New Collection
' add items
Public Function addItem(obj As clsItem)
' check if exists, if so increment quantity
Dim item As clsItem, bExists As Boolean
For Each item In items
If item.id = obj.id Then
item.qu = item.qu + obj.qu
item.creo = item.creo & " " & obj.creo
bExists = True
Exit For
End If
Next
' does not exist so add new
If Not bExists Then items.add obj, obj.id
End Function
' save object and all children
Public Sub SaveToWorksheet(rng As Range)
Const SP = 5 ' no of spaces to indent at each level
Dim item As clsItem
If level > 0 Then
rng = Space(level * SP) & id
rng.Offset(0, 1) = name
rng.Offset(0, 2) = qu
rng.Offset(0, 3) = material
rng.Offset(0, 4) = level
rng.Offset(0, 5) = itemtype
rng.Offset(0, 6) = parent
rng.Offset(0, 7) = creo
Set rng = rng.Offset(1)
End If
' recurse
For Each item In Me.items
item.SaveToWorksheet rng
Next
End Sub
Upvotes: 1
Reputation: 16392
This scans the Output sheet and creates a new sheet called "No Duplicates". However, with the bugs in your data I suspect this won't work for more complex cases e.g. where an assembly appears in other assemblies at different levels.
Update 1 - dictionary at each level
Sub RemoveDuplicates()
Const START_ROW = 3 ' skip headers
Const COL_PARTNO = 1 ' A
Const COL_QU = 3 ' C
Const COL_LEVEL = 5 ' E
Const COL_TYPE = 6 ' F
Dim wb As Workbook, ws As Worksheet, wsNew As Worksheet
Dim iLastRow As Long, i As Long, r As Long
Dim PartNo As String, isASM As Boolean
Dim dictLevel As Integer, level As Integer
' dictionary at each level
Dim dictASM(10) As Object, arQu As Variant
Set wb = ThisWorkbook
Set ws = wb.Sheets("Output")
iLastRow = ws.Cells(Rows.Count, "A").End(xlUp).Row
ReDim arQu(iLastRow)
For i = 0 To 10
Set dictASM(i) = CreateObject("Scripting.Dictionary")
Next
r = START_ROW
dictLevel = 0
Do
PartNo = Trim(ws.Cells(r, COL_PARTNO))
isASM = Trim(ws.Cells(r, COL_TYPE) = "ASM")
arQu(r) = ws.Cells(r, COL_QU)
level = ws.Cells(r, COL_LEVEL)
If isASM Then
' is this a new assembly clear lower level dictionaries
If level < dictLevel Then
For i = dictLevel To UBound(dictASM)
Set dictASM(i) = CreateObject("Scripting.Dictionary")
Next
dictLevel = level
Else
If dictASM(dictLevel).exists(PartNo) Then
' duplicate assembly
' increment first occurence
i = dictASM(dictLevel)(PartNo)
arQu(i) = arQu(i) + arQu(r)
arQu(r) = 0
level = ws.Cells(r, COL_LEVEL)
' delete lower level items
Do While ws.Cells(r + 1, COL_LEVEL) > level
r = r + 1
arQu(r) = 0 ' delete later
Loop
Else
' first occurence
dictASM(dictLevel).add PartNo, r
End If
End If
End If
r = r + 1
Loop While r < iLastRow
' create new sheet without duplicates
Set wsNew = Sheets.add
wsNew.Name = "No Duplicates"
ws.Range("A2:G2").Copy wsNew.Range("A2") ' header
i = START_ROW
For r = START_ROW To iLastRow
If arQu(r) > 0 Then
ws.Cells(r, 1).Resize(1, 8).Copy wsNew.Cells(i, 1)
wsNew.Cells(i, COL_QU) = arQu(r)
i = i + 1
End If
Next
wsNew.Columns("A:G").AutoFit
MsgBox Format(iLastRow - 1, "#,###") & " rows scanned", vbInformation
End Sub
Upvotes: 0