abap - Get total without LOOP

Question

I refer to this question and answers To get total and subtotal without loop in new abap. But my requirements are much simpler than that. I only need the categories with their corresponding totals.

I have internal table like:

Category Amount
AAA     10
AAA     20
BBB     30
CCC     40
CCC     50
CCC     60

I need to build the internal table as this:

Category Amount
AAA       30
BBB       30
CCC       150

Can anyone help with this?

Answer 1

From release 7.40 and onward, this can be accomplished by combining grouping in internal tables (FOR GROUPS) with REDUCE for calculating the sum of each group.

First a couple declarations:

TYPES:
   BEGIN OF data_struct,
    category TYPE c LENGTH 3,
    amount TYPE i,
   END OF data_struct,
   data_tab TYPE STANDARD TABLE OF data_struct WITH EMPTY KEY.

DATA(input) = VALUE data_tab(
    ( Category = 'AAA' Amount = 10 )
    ( Category = 'AAA' Amount = 20 )
    ( Category = 'BBB' Amount = 30 )
    ( Category = 'CCC' Amount = 40 )
    ( Category = 'CCC' Amount = 50 )
    ( Category = 'CCC' Amount = 60 )
).

And here the solution:

DATA(output) = VALUE data_tab(
    FOR GROUPS group OF input_line IN input 
        GROUP BY ( category = input_line-category )
        (   category = group-category
            amount = REDUCE #(
                INIT subtotal = 0
                FOR group_line IN GROUP group
                NEXT subtotal  = subtotal  + group_line-amount 
            ) 
        ) 
).

Now what exactly happens here?

The data for the internal table output is generated by FOR GROUPS. The FOR GROUPS table expression consists of three parts: The declaration of the group, the grouping criteria and the results.

First about the declarative part: GROUPS group OF input_line IN input. This defines that there is a group called "group", that its source is the internal table "input" and that we refer to each line of that source table as "input_line".

Next the grouping criteria: GROUP BY ( category = input_line-category ). This means that each group is identified by a key with a single field "category" and that we want to group all lines where the field "category" is identical.

And now for the result part:

( category = group-category
  amount = REDUCE #(
    INIT subtotal = 0
    FOR group_line IN GROUP group
    NEXT subtotal = subtotal + group_line-amount
  )
)

This defines how each group is supposed to be represented in the output set. The first field "category" is simple - just put in the value of "category" of the group called "group".

The second field "amount" is a bit more complicated, because its value is calculated by a REDUCE. The purpose of the REDUCE construction expression is to take many values and turn them into one value. It does this by doing the following things:

1. Declare and initialize a variable: INIT subtotal = 0
2. Do a loop over an internal table or the lines of a group in an outer FOR GROUPS. In this case the latter: FOR group_line IN GROUP group
3. Do the same code snippet with each line of that input, changing the variable. In this case by adding its amount to the subtotal: NEXT subtotal = subtotal + group_line-amount
4. Return the final value of the variable declared in step 1.

Result from debugger: