Announcing New Text and Array Functions

Henn Sarv 

I see from your proposed solution that I mis-read the initial data, taking the spaces as string separators rather than separate columns.  That modifies my solution since I cannot pass a 2D table to REDUCE; it either has to be a thunk array or an index.  Using an index, I get

WorksheetFormula
= REDUCE(Header, {1;2}, Unpackλ)

Header = {"Name","Year","Value"};

Unpackλ
= LAMBDA(acc,r,
     LET(
        terms,          CHOOSEROWS(Table1,r),
        initialTerms,   DROP(terms,,-1),
        finalTerm,      TAKE(terms,,-1),
        values,         VALUE(TEXTSPLIT(@finalTerm, ,",")),
        newRows,        HSTACK(IF(values,initialTerms),values),
        VSTACK(acc, newRows)
     )
  );

Since that leaves both the table name and the row index hard-wired into the function I would normally wrap it in a further Lambda in order to pass the name as a parameter.

That could be done with

WorksheetFormula
= Normaliseλ(Table1)

Normaliseλ =  LAMBDA(table,
    REDUCE(Header, SEQUENCE(ROWS(table)), Unpackλ))