asked on
SSIS Converting Flat File with Header and Detail lines
Hi,
I have a flat file that has two row types (OH = Order header, OL = Order line). One OH row can have multiple OL lines. I would like to combine the OH data rows and the OL data rows into one output row.
So for example the source data may look like this:
OH, Ted, #12345, M, United Kingdom
OL, 1, 2, 3, 4
OL, 2, 2, 4, 5
OH, Bob, #12346, M, United Kingdom
OL, 2, 4, 4, 6
OL, 2, 2, 4, 7
I would like the target file to be:
Ted, #12345, M, United Kingdom, 1, 2, 3, 4
Ted, #12345, M, United Kingdom, 2, 2, 4, 5
Bob, #12346, M, United Kingdom, 2, 4, 4, 6
Bob, #12346, M, United Kingdom, 2, 2, 4, 7
I know I could split the data rows into two outputs but I do not know how I would then link the data back together as there is no link in the source file between the OH and OL rows.
Thanks, Greg
I have a flat file that has two row types (OH = Order header, OL = Order line). One OH row can have multiple OL lines. I would like to combine the OH data rows and the OL data rows into one output row.
So for example the source data may look like this:
OH, Ted, #12345, M, United Kingdom
OL, 1, 2, 3, 4
OL, 2, 2, 4, 5
OH, Bob, #12346, M, United Kingdom
OL, 2, 4, 4, 6
OL, 2, 2, 4, 7
I would like the target file to be:
Ted, #12345, M, United Kingdom, 1, 2, 3, 4
Ted, #12345, M, United Kingdom, 2, 2, 4, 5
Bob, #12346, M, United Kingdom, 2, 4, 4, 6
Bob, #12346, M, United Kingdom, 2, 2, 4, 7
I know I could split the data rows into two outputs but I do not know how I would then link the data back together as there is no link in the source file between the OH and OL rows.
Thanks, Greg
Microsoft SQL Server 2008Microsoft SQL ServerSSIS
Last Comment
spen_lang
I think you'll need to iterate over the source using a script task so that you're able to detect when you're moving to a new order. I can't think of another way to ensure you don't lose the relationship between header and lines, unless the source has structural entities like XML.
ASKER
This is the route I have gone down. The CSV file is delimited using commas and the Order header and order detail rows have different amount of columns (something I forgot to mention in the post).
So in my script I am splitting the row in the file using the delimiter of a comma. However, my concern is that someone may put a comma in the text of the one of the columns, which would cause problems (e.g. in an address line). Is there any way to overcome this?
So in my script I am splitting the row in the file using the delimiter of a comma. However, my concern is that someone may put a comma in the text of the one of the columns, which would cause problems (e.g. in an address line). Is there any way to overcome this?
ASKER
Code being used in the script component:
Public Class ScriptMain
Inherits UserComponent
Private sDeliveryName As String = ""
Private sDeliveryAddress1 As String = ""
Private sDeliveryAddress2 As String = ""
Private sDeliveryAddress3 As String = ""
Private sDeliveryAddress4 As String = ""
Private sDeliveryAddress5 As String = ""
Private sCustomerOrderNumber As String = ""
Private sDespatchDate As String = ""
Private sDeliveryDate As String = ""
Private sOrderNotes1 As String = ""
Private sOrderNotes2 As String = ""
Private sOrderNotes3 As String = ""
Private sErrorLog As String = ""
Private sRow As String()
Const sDelimiter As String = ","
Public Overrides Sub OrderRows_ProcessInputRow(ByVal Row As OrderRowsBuffer)
'
' Add your code here
''
sRow = Row.OrderRow.Split(sDelimiter.ToCharArray())
If sRow(0).ToString.ToUpper = "OH" Then
processHeaderRow()
Row.DirectRowToHeaderRow()
ElseIf sRow(0).ToString.ToUpper = "OL" Then
Row.Address1 = sDeliveryName
Row.Address2 = sDeliveryAddress1
Row.Address3 = sDeliveryAddress2
Row.Address4 = sDeliveryAddress3
Row.Address5 = sDeliveryAddress4
Row.BarCode = ""
Row.BatchNo = ""
Row.BBE = ""
Row.CustomerCode = "TEST"
Row.CustomerGoodsRef = ""
Row.CustOrderNo = sCustomerOrderNumber
Row.CustRef = ""
Row.DeliveryTime = "08:00"
Row.DeliveryDateTime = sDeliveryDate
Row.DeliveryInstructions = sOrderNotes1
Row.Depot = ""
Row.DespatchDate = sDespatchDate
Row.DestLocation = ""
Row.Location = ""
Row.OrderPriority = ""
Row.OrderDate = Date.Today.ToString("dd/MM/yyyy")
Row.OrderField1 = ""
Row.OrderField2 = ""
Row.OrderField3 = ""
Row.OrderField4 = ""
Row.Postcode = sDeliveryAddress5
Row.ProductCode = sRow(1).ToString
Row.Quantity = sRow(7).ToString
Row.TransportOut = ""
Row.UnitsOfOrder = "2"
Row.UserField3 = ""
Row.UserField4 = ""
Row.Warehouse = ""
'Row.ProductDesc = sRow(2).ToString
Row.DirectRowToDetailRows()
End If
End Sub
Private Sub processHeaderRow()
sDeliveryName = sRow(13).ToString
sDeliveryAddress1 = sRow(14).ToString
sDeliveryAddress2 = sRow(15).ToString
sDeliveryAddress3 = sRow(16).ToString
sDeliveryAddress4 = sRow(17).ToString
sDeliveryAddress5 = sRow(18).ToString
sCustomerOrderNumber = sRow(23).ToString
sDespatchDate = sRow(28).ToString
sDeliveryDate = sRow(29).ToString
sOrderNotes1 = sRow(25).ToString
sOrderNotes2 = sRow(26).ToString
sOrderNotes3 = sRow(27).ToString
End Sub
Is it possible to delimit by a less common character, such as pipe ("|") ? There isn't a simple way to detect the difference between a delimiter and a comma in an address.
ASKER
It doesn't look like the provider is able to change the delimiter...
ASKER CERTIFIED SOLUTION
THIS SOLUTION ONLY AVAILABLE TO MEMBERS.
View this solution by signing up for a free trial.
Members can start a 7-Day free trial and enjoy unlimited access to the platform.
ASKER
I think throwing an error if there is an incorrect amount of elements is the best solution for this as the provider will not make any changes. Thank you for your help.