Splitting output field strings after stats2 -a corr

[NikosAlexandris]

The following command

mlr --csv --c2x stats2 -a corr -f $(combine_variables "$FIELDS_A" "$FIELDS_B") input.csv

will create a new CSV file with two fields. Each record's first field will contain the concatenation of the input field names (each time a single pair) followed by the suffix _corr.

The combine_variables function does:

function combine_variables() {
    echo $(parallel echo ::: $1 ::: $2 |sed 's/\ /,/g' |paste -s -d',') ;
 }

to combine in all possible ways strings from $FIELDS_A and $FIELDS_B.

I wonder if there is a way to fine control this string because it is not easy to split the combined string and create a new field/s that possibly help/s grouping the data in other categories.

So, given an input field pair field_a and field_b, the first field in the output record will be field_a_field_b_corr.
It would be nice, if for example the field name could be field_a_vs_field_b or else, using some unique sequence of characters that can be used to split the columns.

Does this make sense? How can this be done?

[NikosAlexandris]

As a work-around, I can think of adding a sequence of characters (say _vs) at the end of all $FIELDS_A field names of the input.csv . This suffix would be then in the middle between two combined field names (from $FIELDS_A and $FIELDS_B). Still figuring how to do this. Any better ways to handle this case?