How to use the formula row by row to find the dynamic change of the total number of columns.

[#1]

How to use the formula row by row to find the dynamic change of the total number of columns.
Such as setting:
The number a is the number of starting column positions that need to be summed
The number b is the total number of columns that need to be summed

dt = Open( "$SAMPLE_DATA/Diabetes.jmp" );
dt << New Column( "test", formula( As Column( 6 ) + As Column( 7 ) + As Column( 8 ) + As Column( 9 ) ) );
dt << run formulas;
Column( "test" ) << deleteFormula;

Such as:
A = 6;
B = 4;
Find the sum of columns 6, 7, 8, and 9 in the new "text" column (starting with column a, total column b).

Thanks!

[#2]

• I tried to write it this way, but it didn't work.

a=6;b=4;
dt = Current Data Table();
dt<<New Column("text",formula(
g=0;
for(i=6,i<=a+b-1,i++,
g=g+As Column(i+1)	
)));dt<<run formulas;
Column("text")<<deleteFormula;Wait(0);

Looks like it's going to be in the brute force form of a line by line loop, right?

[#3 · Accepted solution]

This situation is weird. This script builds the formula:

Names Default To Here( 1 );

dt = Open( "$SAMPLE_DATA/Diabetes.jmp" );

col = dt << New Column( "test" );

lo = 6;
hi = 9;
f = Expr( Add() );

For( c = lo, c <= hi, c++,
	r = (dt << Get Column Reference( Eval List( { Column( dt, c ) << Get Name } ) ))[1];
	Insert Into( f, r );
);

Eval(
	Substitute(
		Expr( col << Set Formula( fff ) ),
		Expr( fff ),
		Name Expr( f )
	)
);

But it does not evaluate. Yet it works if I examine the preview in the formula editor.

Anyway, why use a formula when you don't want to?

Names Default To Here( 1 );

dt = Open( "$SAMPLE_DATA/Diabetes.jmp" );

col = dt << New Column( "test" );

lo = 6;
hi = 9;
sum = J( N Row( dt ), 1, 0 );

For( c = lo, c <= hi, c++,
	sum += Column( dt, c ) << Get As Matrix;
);

col << Set Values( sum );

[#4]

Thank markbailey!

https://community.jmp.com/t5/Discussions/How-to-SUM-rows-with-variable-column-names/m-p/9817 

Seeing this post, I also tried this method, but to determine the number of text format columns.

dt = Open( "$SAMPLE_DATA/Diabetes.jmp" );
a = 4;//has 2 text format columns.
b = 4;
m = (dt << get as matrix)[0, Index( a, a + b - 1 )];
s = (V Sum( m` ))`;
dt << New Column( "test", Numeric, Continuous, Values( s ) );