On Aug 27, 1:50 pm, mich...@hooperprinting.com wrote:

Well, I can tell you how to find the space before the first integer,
but that may not help.

What about John Smith PO Box 123 Washington DC 08610 or Tom Jones One
Hammerskjold Plaza New York NY 10128?  If you have those cases, an $8/
hr data entry temp is the way to go.

With those caveats in mind, to get names, create a calc field with
this as the calculation, where field is the name and address field:

Let([
num = Filter(field; "1234567890");
num = left(num;1);
pos = Position(field; num; 1;1);
pos = pos -1;
names = Trim(left(field; pos))
];
names)

Look at the Substitute, Position, RightWords, Right, Left and Middle
text functions.

The easy way to parse out that kind of address is to extract what you
know, and then remove it from the string, until the unknown part (the
name) is left.

So extract the zip code with RightWords, then Substitute out the string
in the Zip field in the original field. The remaining RightWord is the
State. Put that in the state field, then Substitute it out in the
original field.

The Address will almost always start with a number and except for names
like Joe Smith the 3rd won't conflict with the address. This should be
rare enough for manual correction. So test on the string for
"1234567890" and start there and bang everything to the right into the
Address field. Substitute out what's in the address field, and what's
left, once you use Trim, is the name.

Of course, PO Boxes cause some confusion. :/  Darn! I guess you could
test on "PO box" or "P.O. Box" and start the address there.  
Patterncount is your friend.

This kind of text parsing is a pain even for experienced developers,
and it takes a lot of testing to get it right. A good hint is to build
"test" calc fields that SHOW what result string is returned for a
specific calculation.