Mapping header fields to items with context change

Hi,

1. Consider the input flat file (csv):

field1,field2,field3,field4

field1,field2,field3,field4

2. The fields are read into the following input structure:

data[]

field1

field2

field3

field4

Consider the output:

data[]

header

field1

items[]

field2

field3

field4

Mapping: For every unique field1 string, I map field2, field3 and field4 to items. However, if field1 (in row 1 and row 2) is the same value, I need to suppress creating the second data[] node and instead add field2, field3 and field4 to a second item[] under the original field1 entry (see below).

data[]

header

field1 (from row 1 of flat file)

items[] (from row1 of flat file)

field2

field3

field4

items[] (from row 2 of flat file)

field2

field3

field4

I would really appreciate some assistance with this. Would I need to create a UDF?

Thanks,

Dave