Regex Parse logic not working correctly

[AnandKumar1]

I've written a parse to split a record into multiple column.

Below is my input file on which some record has 4 different values , some has only 3 and some has 5

:25:A1BC 0055555 USD

:25:X1Y2 USD

:25:HK13 49000000 USD NY

And to achieve this i have written this into parse= (:25:)(\w+)(\s)(\d+)(\s)([A-Z]*[0-9]*)(\s)*([A-Z]*[0-9]*)

But this parse result is getting correct for 1st record line. but Null is coming for 2nd record line and 3rd is again right.

:25:	A1BC	0055555	USD
:25:	HK13	49000000	USD	NY

This is required output

:25:	A1BC	0055555	USD
:25:	X1Y2		USD
:25:	HK13	49000000	USD	NY

Can someone tell me what's going wrong here? Is the logic in parse wrong?

[jdunkerley79 · Accepted solution]

I think something like:

(:25:)(\w+)\s*(\d+)?\s([A-Z]*[0-9]*)\s*([A-Z]*[0-9]*)?

What I have done is marked a few blocks as optional (with a trailing ?) and made the spaces 0 or more.

Seems like it works with your example.

[mceleavey]

Hi @AnandKumar1 ,

I've attached the regex workflow for you.

M.

[Qiu]

@AnandKumar1 
In oder to make it work, we need more data to test it I think 😁