That fails because parsing CSV-style data is more difficult than it appears. The first hurdle is how to handle when the separator appears in a quoted field ("joe _blow" _ 2 _ "hourly"). The second hurdle is more fiendish: newlines embedded in quoted fields. It's less likely to occur, but if it does, iterating over records becomes painful.

I hadn't considered newlines but the separator sounded like it would be unique enough in this instance, the original poster described it as "space underscore space with _UNSC being the escape string"... which I admit isn't entirely clear to me and why I suggested the OP post some sample data, I just took the liberty of simplifying it to "_ " for my example, my bad! :-)

--
Murray Barton
Do not seek to follow in the footsteps of the wise. Seek what they sought. -Basho