topic Try statement with Group Columns in Discussions

Try statement with Group Columns

SpannerHead — Fri, 20 Feb 2026 21:21:16 GMT

I have a list of columns I want to group but the script fails if any of the columns are absent. A try statement does not seem to fix the problem.

dt = Current Data Table();

Try( dt <<  Group Columns ("TM", {:Param_1, :Param_2, :Param_3}));

Re: Try statement with Group Columns

mmarchandFSLR — Fri, 20 Feb 2026 21:35:49 GMT

If you want to group all columns present in that list (even if, say, :Param_2 is missing), you could try something like this.

Names Default To Here( 1 );
cols = {:a, :b, :c, :d, :e};
dt << Group Columns( "Some Columns", Filter Each( {v, i}, cols, dt << Has Column( Char( Name Expr( v ) ) ) ) );

Re: Try statement with Group Columns

SpannerHead — Fri, 20 Feb 2026 21:47:26 GMT

Had to tweak this a bit to identify dt.

Names Default To Here( 1 );
dt = Current Data Table();
cols = {:a, :b, :c, :d, :e};
dt << Group Columns( "Some Columns", Filter Each( {v, i}, cols, dt << Has Column( Char( Name Expr( v ) ) ) ) );

Re: Try statement with Group Columns

jthi — Fri, 20 Feb 2026 22:13:42 GMT

My suggestion would be to avoid :col and use strings. With JMP19 you can use Set Intersection

Names Default To Here(1); 

dt = open("$SAMPLE_DATA/Big Class.jmp");

g = {"age", "sex", "height", "test", "test2"};

mycols = Set Intersection(dt << Get Column Names("String"), g);

dt << Group Columns("MyGroup", mycols);

In earlier versions Associative Arrays (or Filter Each) can replace Set Intersection