I have several tables from SAP in TXT format. For example MARD, MARC, RESB, etc. I want to do some data validation in Lava BRE.

The problem is all the text documents from SAP start with date, dashes, and useless information on the first two rows. Even after I clean it using filter nodes I still have the problem of spitting different columns correctly.

The text documents goes like this:

10/05/2017 date - useless text
l column 1 l column 2
l row 1 l row 1
l row 2 l row 2

I want to know the most efficient way (because I have several tables) to clean all useless information and dashes (-), then spit the important information into separate columns by delimiter (l). So basically I want to get rid of everything except the column 1, 2 and the rows

Thank you in advance,