Generic extractor FM : taking 5-6 hours time for 3 months to BW urgent:

Question

Dear experts,
I have designed a FM for generic extraction , which is taking 5-6 hoours time for 3 months data i.e 24 lakhs records to BW up to PSA .
i have given the coding below plz provide any modifications to improve the performance.....
FUNCTION zhr_att_analysis.
""Local Interface:
*"  IMPORTING
*"     VALUE(I_REQUNR) TYPE  SRSC_S_IF_SIMPLE-REQUNR
*"     VALUE(I_DSOURCE) TYPE  SRSC_S_IF_SIMPLE-DSOURCE OPTIONAL
*"     VALUE(I_MAXSIZE) TYPE  SRSC_S_IF_SIMPLE-MAXSIZE OPTIONAL
*"     VALUE(I_INITFLAG) TYPE  SRSC_S_IF_SIMPLE-INITFLAG OPTIONAL
*"     VALUE(I_REMOTE_CALL) TYPE  SBIWA_FLAG DEFAULT SBIWA_C_FLAG_OFF
*"  TABLES
*"      I_T_SELECT TYPE  SBIWA_T_SELECT OPTIONAL
*"      I_T_FIELDS TYPE  SBIWA_T_FIELDS OPTIONAL
*"      E_T_DATA STRUCTURE  ZHR_ATT_MAIN OPTIONAL
*"  EXCEPTIONS
*"      NO_MORE_DATA
*"      ERROR_PASSED_TO_MESS_HANDLER
Auxiliary Selection criteria structure
  DATA: l_s_select TYPE sbiwa_s_select.
Maximum number of lines for DB table
  STATICS: l_maxsize TYPE sbiwa_s_interface-maxsize.
Select ranges
  RANGES: l_r_pernr FOR pa9004-pernr,
          l_r_bukrs FOR pa0001-bukrs,
          l_r_persg FOR pa0001-persg,
          l_r_begda FOR pa9004-begda,
          l_r_persk FOR pa0001-persk.
Maximum number of lines for DB table
  STATICS: s_s_if TYPE srsc_s_if_simple,
counter
          s_counter_datapakid LIKE sy-tabix,
cursor
          s_cursor TYPE cursor.
*"Declaration of store data
TYPES : BEGIN OF ty_9004,
         pernr TYPE persno,
         endda TYPE endda,
         begda TYPE begda,
         zrs TYPE zrs,
         zstorecode TYPE zstorecode,
        END OF ty_9004.
*"Declaration of employee data
  TYPES : BEGIN OF ty_0001,
          pernr TYPE pernr_d,
          endda TYPE endda,
          begda TYPE begda,
         AEDTM TYPE AEDAT,
          bukrs TYPE bukrs,
          persg TYPE persg,
          persk TYPE persk,
          END OF ty_0001.
*"Declaration of expected mandays
  TYPES : BEGIN OF ty_0000,
          pernr TYPE persno,
          endda TYPE endda,
          begda TYPE begda,
          aedtm TYPE aedat,
          stat2 TYPE stat2,
          massn TYPE massn,
          END OF ty_0000.
*"Declaration of man days swiped
  TYPES : BEGIN OF ty_teven,
          pernr TYPE pernr_d,
          ldate TYPE ldate,
          satza TYPE retyp,
          aedtm TYPE aedat,
          counter_swiped TYPE i,
          END OF ty_teven.
*"Declaration of Man days regularized
  TYPES : BEGIN OF ty_2002,
          pernr TYPE pernr_d,
          subty TYPE subty,
          endda TYPE endda,
          begda TYPE begda,
          aedtm TYPE aedat,
          END OF ty_2002.
*"Declaration of Man days lostdue to leave
  TYPES : BEGIN OF ty_2001,
          pernr TYPE pernr_d,
          subty TYPE subty,
          endda TYPE endda,
          begda TYPE begda,
          aedtm TYPE aedat,
          END OF ty_2001.
*****Declaration of weekly off
  TYPES : BEGIN OF ty_2003,
          pernr TYPE pernr_d,
          subty TYPE subty,
          endda TYPE endda,
          begda TYPE begda,
          aedtm TYPE aedat,
          tprog TYPE tprog,
          END OF ty_2003.
Auxiliary Selection criteria structure
  DATA :
        it_0001 TYPE TABLE OF ty_0001,
        wa_0001 TYPE ty_0001,
        it_0000 TYPE TABLE OF ty_0000,
        wa_0000 TYPE ty_0000,
        it_teven TYPE TABLE OF ty_teven,
        wa_teven TYPE ty_teven,
        it_2002 TYPE TABLE OF ty_2002 ,
        wa_2002 TYPE ty_2002,
        it_2001 TYPE TABLE OF ty_2001,
        wa_2001 TYPE ty_2001,
        it_2003 TYPE TABLE OF ty_2003,
        wa_2003 TYPE ty_2003,
        wa_target TYPE zhr_att_main.
  DATA : date  TYPE dats,
  doj TYPE dats,
  dol TYPE dats,
  date1 TYPE dats,
  date2 TYPE dats,
         counter(9)  TYPE n.
Initialization mode (first call by SAPI) or data transfer mode
(following calls) ?
  IF i_initflag = sbiwa_c_flag_on.
Initialization: check input parameters
                buffer input parameters
                prepare data selection
Check DataSource validity
    CASE i_dsource.
      WHEN 'ZHR_ATT_ANALYSIS'.
      WHEN OTHERS.
        IF 1 = 2. MESSAGE e009(r3). ENDIF.
        log_write 'E'                  "message type
                  'R3'                 "message class
                  '009'                "message number
                  i_dsource            "message variable 1
                  ' '.                 "message variable 2
        RAISE error_passed_to_mess_handler.
    ENDCASE.
    APPEND LINES OF i_t_select TO s_s_if-t_select.
Fill parameter buffer for data extraction calls
    s_s_if-requnr    = i_requnr.
    s_s_if-dsource   = i_dsource.
    s_s_if-maxsize   = i_maxsize.
Fill field list table for an optimized select statement
(in case that there is no 1:1 relation between InfoSource fields
and database table fields this may be far from beeing trivial)
    APPEND LINES OF i_t_fields TO s_s_if-t_fields.
  ELSE.                 "Initialization mode or data extraction ?
Data transfer: First Call      OPEN CURSOR + FETCH
               Following Calls FETCH only
First data package -> OPEN CURSOR
    IF s_counter_datapakid = 0.
      LOOP AT s_s_if-t_select INTO l_s_select WHERE fieldnm = 'PERNR'.
        MOVE-CORRESPONDING l_s_select TO l_r_pernr.
        APPEND l_r_pernr.
      ENDLOOP.
      LOOP AT s_s_if-t_select INTO l_s_select WHERE fieldnm = 'BUKRS'.
        MOVE-CORRESPONDING l_s_select TO l_r_bukrs.
        APPEND l_r_bukrs.
      ENDLOOP.
      LOOP AT s_s_if-t_select INTO l_s_select WHERE fieldnm = 'PERSG'.
        MOVE-CORRESPONDING l_s_select TO l_r_persg.
        APPEND l_r_persg.
      ENDLOOP.
      LOOP AT s_s_if-t_select INTO l_s_select WHERE fieldnm = 'BEGDA'.
        MOVE-CORRESPONDING l_s_select TO l_r_begda.
        APPEND l_r_begda.
      ENDLOOP.
      LOOP AT s_s_if-t_select INTO l_s_select WHERE fieldnm = 'PERSK'.
        MOVE-CORRESPONDING l_s_select TO l_r_persk.
        APPEND l_r_persk.
      ENDLOOP.
      OPEN CURSOR WITH HOLD s_cursor FOR
populate only store code employess does not have empty store codes
        SELECT apernr bpernr bendda bbegda bbukrs bpersg b~persk FROM pa9004 AS a INNER JOIN pa0001 AS b
                ON  apernr = bpernr
                 WHERE a~pernr IN l_r_pernr AND
                      a~zstorecode <> ''    AND
                      bukrs IN l_r_bukrs  AND
                      persg IN l_r_persg AND
                      persk IN l_r_persk.
    ENDIF.
Fetch records into interface table.
  named E_T_'Name of extract structure'.
    FETCH NEXT CURSOR s_cursor
               APPENDING CORRESPONDING FIELDS
               OF TABLE  it_0001
               PACKAGE SIZE s_s_if-maxsize.
    IF sy-subrc <> 0.
      CLOSE CURSOR s_cursor.
      RAISE no_more_data.
    ELSE.
     break-point.
      IF l_r_begda-high = '00000000' AND l_r_begda-low = '00000000'.
        date1 = sy-datum - 1.
        date2 = sy-datum - 1.
      ELSE.
        date1 = l_r_begda-low .
        date2 = l_r_begda-high.
      ENDIF.
      SORT it_0001 BY pernr persg begda endda bukrs.
      DELETE it_0001 WHERE persg NE 'T' AND
                           persg NE 'K' AND
                           persg NE 'P' AND
                           persg NE 'W'.
      DELETE ADJACENT DUPLICATES FROM it_0001 COMPARING pernr begda endda bukrs.
populate all the employees that are active in pa9004.
      IF NOT it_0001[] IS INITIAL.
        SELECT pernr endda begda aedtm massn FROM pa0000
               INTO CORRESPONDING FIELDS OF TABLE it_0000
               FOR ALL ENTRIES IN it_0001
               WHERE pernr = it_0001-pernr
                 AND ( massn = 'A1' OR massn = '00' OR massn = 'A6' OR massn = 'A3' ).
        SORT it_0000 BY pernr begda DESCENDING.
      ENDIF.
populate SWIPED MAN DAYS data
      IF NOT it_0001[] IS INITIAL.
        SELECT pernr ldate satza aedtm FROM teven
           INTO CORRESPONDING FIELDS OF  TABLE it_teven
           FOR ALL ENTRIES IN it_0001
           WHERE pernr = it_0001-pernr AND
                             satza = 'P01'
                             AND ldate IN l_r_begda.
        SORT it_teven BY pernr ldate.
      ENDIF.
**populate REGULARIZATION DAYS data
      IF NOT it_0001[] IS INITIAL.
        SELECT pernr subty endda begda aedtm FROM pa2002
          INTO CORRESPONDING FIELDS OF  TABLE it_2002
           FOR ALL ENTRIES IN it_0001
           WHERE pernr = it_0001-pernr
            AND  begda >= date1
            AND endda <= date2 .
        SORT it_2002 BY pernr begda endda.
      ENDIF.
**populate LEAVE DAYS data
      IF NOT it_0001[] IS INITIAL.
        SELECT pernr subty endda begda aedtm FROM pa2001
          INTO CORRESPONDING FIELDS OF   TABLE it_2001
           FOR ALL ENTRIES IN it_0001
           WHERE pernr = it_0001-pernr
            AND  begda >= date1
            AND endda <= date2  .
        SORT it_2001 BY pernr begda endda .
      ENDIF.
**populate WEEKLY OFF data
      IF NOT it_0001[] IS INITIAL.
        SELECT pernr subty endda begda aedtm tprog FROM pa2003
          INTO CORRESPONDING FIELDS OF  TABLE it_2003
             FOR ALL ENTRIES IN it_0001
             WHERE pernr = it_0001-pernr AND
                          tprog = 'OFF'
                           AND  begda >= date1
                           AND endda <= date2  .
        SORT it_2003 BY pernr begda endda.
      ENDIF.
      date = sy-datum.
********added changes on 06.04.2008**************action type & date dependent extaction****
loop over it_0001 table
     BREAK-POINT.
      LOOP AT it_0001 INTO wa_0001.
       if sy-tabix = 1.
        counter = 0.
for expected mandays
        LOOP AT it_0000 INTO wa_0000 WHERE pernr = wa_0001-pernr .
          IF wa_0000-massn = 'A1' OR wa_0000-massn = '00' OR wa_0000-massn = 'A3'.
            doj = wa_0000-begda.
           if wa_0000-endda = '99991231'.
          date2  = sy-datum.
           else.
            dol = date2.
           endif.
          ELSEIF wa_0000-massn = 'A6'.
            dol = wa_0000-begda.
          ENDIF.
        ENDLOOP.
        IF  date1 <= wa_0001-begda AND date2 <= wa_0001-endda AND date2 >= wa_0001-begda AND date1 <= wa_0001-endda.
          counter = date2 - wa_0001-begda .
          counter = counter + 1.
          date = wa_0001-begda - 1.
        ELSEIF date1 >= wa_0001-begda  AND date2 >= wa_0001-endda AND date2 >= wa_0001-begda AND date1 <= wa_0001-endda.
          counter =  wa_0001-endda - date1.
          counter = counter + 1.
          date = date1 - 1.
        ELSEIF date1 >= wa_0001-begda AND date2 <= wa_0001-endda AND  date2 >= wa_0001-begda AND date1 <= wa_0001-endda.
          counter = date2  - date1.
          counter = counter + 1.
          date = date1 - 1.
        ELSEIF  date1 <= wa_0001-begda AND  date2 >= wa_0001-endda AND date2 >= wa_0001-begda AND date1 <= wa_0001-endda.
          counter = wa_0001-endda - wa_0001-begda.
          counter = counter + 1.
          date =  wa_0001-begda - 1.
        ELSE.
          CONTINUE.
        ENDIF.
********completed changes on 06.04.2008**************action type & date dependent extaction**
split records from date of joining to till date
        DO counter  TIMES.
          CLEAR : wa_teven , wa_target.
          date = date + 1.
          wa_target-date1 = date.
          wa_target-pernr = wa_0001-pernr.
          wa_target-bukrs = wa_0001-bukrs.
          wa_target-persg = wa_0001-persg.
          wa_target-persk = wa_0001-persk.
for expected mandays count
          IF wa_target-date1 >= doj AND wa_target-date1 <= dol.
            wa_target-expectedmandays = 1.
            wa_target-aedtm = wa_0000-aedtm.
for swiped mandays
            READ TABLE it_teven INTO wa_teven WITH KEY pernr = wa_target-pernr
                                                       ldate = wa_target-date1 BINARY SEARCH.
            IF sy-subrc = 0.
              wa_target-swiped_days = 1.
              wa_target-aedtm = wa_teven-aedtm.
            ENDIF.
for regularized days
            LOOP AT it_2002 INTO wa_2002 WHERE pernr = wa_target-pernr
               AND  ( endda GE wa_target-date1 AND begda LE wa_target-date1 ).
              wa_target-reg_days  = 1.
              wa_target-subty2 = wa_2002-subty.
              wa_target-aedtm = wa_2002-aedtm.
            ENDLOOP.
for leave days
            LOOP AT it_2001 INTO wa_2001 WHERE pernr = wa_target-pernr
               AND  ( endda GE wa_target-date1 AND begda LE wa_target-date1 ).
              wa_target-leave_days  = 1.
              wa_target-subty1 = wa_2001-subty.
              wa_target-aedtm = wa_2001-aedtm.
            ENDLOOP.
for weekly off days
            LOOP AT it_2003 INTO wa_2003 WHERE pernr = wa_target-pernr
               AND  ( endda GE wa_target-date1 AND begda LE wa_target-date1 ).
              wa_target-off_days   = 1.
              wa_target-aedtm = wa_2003-aedtm.
            ENDLOOP.
append work area to e_t_data
            APPEND wa_target TO  e_t_data.
          ENDIF.
        ENDDO.
      ENDLOOP.
clear internal tables
      CLEAR :  it_0000 , it_0001 , it_2001 , it_2002 , it_2003 , it_teven.
    ENDIF.
    s_counter_datapakid = s_counter_datapakid + 1.
  ENDIF.   "Initialization mode or data extraction ?
ENDFUNCTION.

Answer 1

Hi Guys
I can have both your cases looked into for you.
Please send me an email using the contact us form in my profile. The address for this form in the section 'about me'.
Thanks
Stuart
BTCare Community Mod
If we have asked you to email us with your details, please make sure you are logged in to the forum, otherwise you will not be able to see our ‘Contact Us’ link within our profiles.
We are sorry that we are unable to deal with service/account queries via the private message(PM) function so please don't PM your account info, we need to deal with this via our email account :-)