Parse Amounts at end of Text String, But Not the Value following the word "line"

[hellyars]

Hi,

Need help with a parse / regex question.   My fields are trapped in a single field called RAW_TXT.   In this case, I am trying to isolate the values (amounts) found at the end of RAW_TXT.  I then parse the values (if more than one) into their respective fields in a later step.

I am using the expression (.*\u\S*) (.*).  This works for 99% of the records.   The exception is when the comment ends with ...line \d {1,3}.  See #2 below.  In these instances, I do NOT want to capture the digits immediately following "line". 

How can I modify my expression to account for this?

RECORD	RAW_TXT	LINE_NMBR and TITLE	VALUES
1	58 E-3 172141 118068	58 E-53	172141 118068
2	NATO AWACS -Air Force requested transfer to line 87 -36401		-36401
3	IPEC B-kit NRE unjustified growth -6593		-6593

[Greg_Murray]

@hellyars 

I might use the tokenize option for this. You can set use a negative look around to avoid capturing the numbers that come directly after line, but you may have to fiddle with it a bit if there are other scenarios like that. 

here is the explanation of the regex too: regexr.com/5j2fi 

Hope that helps,

Greg

[hroderick-thr]

Regex makes my eyes cross. For this I would formulate

reversestring(left(reversestring(RAW_TXT),findstring(reversestring(RAW_TXT),' ')))