Hi
I think you are facing the situation where the text file is actually a multipage report with headers and footers along with it. If this is the case, then I suggest you first extract all the data into a temporary table. Then in a subsequent dataflow use this temporary table as the source to a query transform and apply a condition in "Where" filter to eliminate the pattern of the header and footer of the document. IF you need more info please do let me know.
Raghu
