Accepted Solutions
HI,
Watching this thread, I'm thinking that this script could be challenging to maintain / proofread later. I'll post an alternate approach in case anyone finds it useful... hopefully someone will.
1) Stack the original table, so it looks like the one below. I call this "MainTable"
2) Store the conditions in a data table just for this purpose. In your case, anything that is not in buckets 1 or 2 is in bucket 3, so those conditions need not be listed. I've only done this for a couple of cases, just for illustration. I call this "ConditionTable". Note that the dt1 column name :Label is hardcoded here... you can work around this of course, if needed or desired.
3) Now, the script becomes a lot shorter and easier to follow, assuming familiarity with expression handling, which it appears from your code that you possess. The sums can be easily obtained by using a summary table, as we now have a "bucket" column for grouping. Overall, the project should be easier to maintain, as the condition table is much easier to read than code. (Be sure to use the == operator in the condition table.)
I've attached sample tables in case you want to try this out.
Cheers,
Brady
Names Default To Here( 1 );
dt1 = Data Table( "MainTable" );
dtC = Data Table( "ConditionTable" );
//set all buckets to 3, and change as needed
dt1 << New Column( "Bucket", <<set each value( 3 ) );
//this will assign buckets 1 and 2 to appropriate rows in dt1 based on conditions
//in the condition table
condExpr = Expr(
dt1:Bucket[dt1 << get rows where( _COND_ )] = _BUCKET_
);
//loop through to apply the conditions
For( i = 1, i <= N Row( dtC ), i++,
Eval(
Substitute( Name Expr( condExpr ),
Expr( _COND_ ), Parse( dtC:Condition[i] ),
Expr( _BUCKET_ ), dtc:Bucket[i]
)
)
);
//produce a table of results
dtSum = dt1 << summary ( sum(:Data), group(:Item, :Bucket));