Need to Process Large Internal Tables Efficiently

[Former Member]

I have two internal tables, T_EKKO(for PO Header data) and T_EKPO(for PO Item data). These tables carry data for 10 PO's with at least 1 line item in each one.

The requirement is to write the following fields on the result screen:

Line 1: PO Number.-

Line 2:PO Item, Article Number.

What is the most efficient way to do this?

[Former Member]

I have two internal tables, T_EKKO(for PO Header data) and T_EKPO(for PO Item data). These tables carry data for 10 PO's with at least 1 line item in each one.

The requirement is to write the following fields on the result screen:

Line 1: PO Number.-

Line 2:PO Item, Article Number.

What is the most efficient way to do this?

[Former Member]

Hi,

You can use parallel processing techinque..

Ex.

SORT T_EKPO BY EBELN.

DATA: V_TABIX TYPE SYTABIX.

LOOP AT T_EKKO.

WRITE: / 'PO Number - ', T_EKKO-EBELN.

READ TABLE T_EKPO TRANSPORTING NO FIELDS

WITH KEY EBELN = T_EKKO-EBELN

BINARY SEARCH.

IF SY-SUBRC = 0.

• Store the row number.

V_TABIX = SY-TABIX.

LOOP AT T_EKPO FROM V_TABIX.

• Exit condition.

IF T_EKPO-EBELN <> T_EKKO-EBELN.

EXIT.

ENDIF.

WRITE: / 'PO item', T_EKPO-EBELP.

.......

ENDLOOP.

ENDIF.

ENDLOOP.

Thanks,

Naren

[Former Member]

For small tables like this, it doesn't really matter, but for larger ones, check this:

/people/rob.burbank/blog/2006/02/07/performance-of-nested-loops

Rob

[suresh_datti]

try this..

sort t_ekko.

delete adjacent duplicates from t_ekko.

sort t_ekpo.

loop at t_ekko.

write: / t_ekko-ebeln.

read table t_ekpo with key ebeln = t_ekko-ebeln

binary search.

if sy-subrc eq 0.

write: / t_ekpo-EBELP,t_ekpo-EAN11.

endif.

endloop.

~Suresh

[bettinasilveira]

But in t_ekpo table you can have more than 1 record....