We are using WM 6.1.

He have a CSV in the following format:

dateTime,fileType,sender,receiver
lineNo,name,description
lineNo,name,description
lineNo,name,description

As you can see, the first line is a header, and the rest are detail lines. We dont have record identifiers. I cannot seem to model this in a flat file schema. Does anyone know how to do this?

Cheers,

Paul,

Create your record in the dictionary.
Then in the schema set it as the default record.
See Marks’ response in this post [url=“wmusers.com”]wmusers.com

Our schema is a little different Chris. We have a header record followed by detail records. by assigning a default record, the schema assumes every record is of the type defined in the default record.

At the moment we have worked around it by assigning the detail line in the data dictionary as the default record. he then add the header record to the schema. The problem with this method is that the header line must have a record identifier.

This isn’t a bad solution, but we want to try and have it so the first record is assumed to be a header, and the rest as detail lines.

I hope this makes sense

Paul,

You are correct in that a recordWithNoId types assume that all the records are of the same structure.
interesting problem.
Hmm??

Paul,

As chris mentioned to use dictionary and default record option (recordWithNoID)make this unlimited and as you said first line will always be header record then take recordWithNoID first iteration and map the header values and for lines it should not be any problem.right?

HTH,
RMG.