Concatenating strings within REDUCE

[BTP_Architect]

Hello,

I'm trying to use new syntax provided by 7.4 and later versions

My requirement is to concatenate the fields of one internal table into a string.

DATA(lv_text) = REDUCE string( INIT lv_where_clause TYPE string
                           FOR ls_werks  IN lt_werks
                           NEXT lv_where_clause = lv_where_clause && ls_werks-low && ',' ).

This will replace the loop on the internal table with concatenation at each step.

My problem with reduce is how to get rid of the last comma?

Many Thanks.

Moez.

[Jean_Sagi]

What about:

NEXT lv_where_clause =  ',' && lv_where_clause && ls_werks-low 

That way you'll end with a first comma instead of a last comma. So get rid of the first character instead of the last one.

Last but not least the lv_where_clause will be a string of the form 1000,2000,3000... (Assuming your logistic centers are 1000,2000,3000, etc...) this seems to me not a proper way to a where clause. Shouldn't it be something like "1000","2000","3000",... ?

J.

[BTP_Architect]

Thanks Jesus for your feedback

Your proposal will give something like: ,,,100020003000 (all comma in the beginning)

They should be separated by comma.

For the lv_where_clause, yes I need it without double quotation marks for further processing...

[Jean_Sagi]

Like Home says : Ouch !!
You know the copy paste syndrome... 😉

By now you have already guessed it, but what about:

NEXT lv_where_clause = lv_where_clause && ',' && ls_werks-low 

J.

[BTP_Architect]

Yes you are right, by this way I will have the comma as first character and remove it using the shift

SHIFT lv_text BY 1 PLACES LEFT.

In my first coding it was also possible to remove the last comma using SHIFT.

But my question was more about the REDUCE and if there is a way to manage such concatenation within REDUCE (without adding any other instruction like SHIFT).

[Jean_Sagi]

Yep. That's though !.

[DoanManhQuynh]

i suggest using condition to check if its last record, like this:

DATA(lv_text) = REDUCE string( INIT lv_where_clause TYPE string
  FOR i = 1 UNTIL i = lines( lt_werks )
  NEXT lv_where_clause = COND #( WHEN i < lines( lt_werks ) THEN lv_where_clause && lt_werks[ i ]-low && ',' 
  ELSE lv_where_clause && lt_werks[ i ]-low ) ).

[matt]

When I needed to do this, I used the example in the documentation. https://help.sap.com/doc/abapdocu_751_index_htm/7.51/en-US/abenreduce_text_abexa.htm

The simplest thing is to use an extra variable. Excuse the renaming of your variable, but I detest prefexing to indicate variable type.

DATA(text) = REDUCE string( INIT where_clause = '' sep = ''
                           FOR werk_range_line IN werk_selection
                           NEXT where_clause = where_clause && sep && werk_range-low
                                sep = ',' ).

[BTP_Architect]

Thanks Matthew

This is what I was searching for

I just changed the initialization of the where_clause (I exchanged where_clause = ' ' by where_clause type string)

Otherwise the text will be on one character...

Final solution:

DATA(text) = REDUCE string( INIT where_clause TYPE string sep = ''
                           FOR werk_range_line IN werk_selection
                           NEXT where_clause = where_clause && sep && werk_range-low
                                sep = ',' ).

[matt]

Try using | as in the SAP example.