This looks like a recipe for disaster. You need to think of a better(more unique) way to identify each column. Counting white space just ain't gonna cut it.
It's not counting whitespace, it's relying on a fixed width column format. There's no way to identify empty fields in the sample that dineshsachdeva gave. But if the first field always starts at column 11, the second field always starts at column 21, the third field always starts at column 31, and the fourth field always starts at column 41, that's a completely unique way of identifying each column. There's no recipe for disaster unless the file format isn't really fixed width. ;)
In fact, this is exactly what I did to parse and reformat some reports for a local wholesaler. The column widths were always the same and the company is still using the report filters to this day without any problems. If dineshsachdeva's file isn't fixed width, this won't work, but neither will anything else that I can think of because there's no way to tell if there's an empty field in between one field and the next.