How to write duplicate data into error file?

[Former Member]

Hi Experts,

I have below code to check if any duplicate data:

SORT <ft_csv_data> BY guid.

               
lt_contacts = <ft_csv_data>.

v1 = lines( lt_contacts ).

DELETE ADJACENT DUPLICATES FROM <ft_csv_data> COMPARING guid.

v2 = lines( <ft_csv_data> ).

if v1 NE v2.
     MESSAGE: 'Duplicate data exist!' TYPE 'I'.
ENDIF.

If there is duplicate data in internal table, i would want to write the duplicate records into error text file.

Do anyone has idea how to did that?

Thank you.

Regards,

Honda

[Former Member]

Hi,

In order to populate the duplicate records into error file, please try this one.

loop at it_final. "final table

if ls_final = lt_final. "here give the key fields which needs to be compared.

"populate record into erro table

else.

ls_final = lt_final.

endif.

clear lt_final.

endloop.

[Former Member]

Hi Lakshmi,

I am not so understand on your code.

Can you explain abit more?

Thanks.

[Former Member]

Hey,

okey..what we can do is, just compare the current record with the previous one. if both are same then populate the record to error table. thats it..

Here ls_final is the structure which will hold the previous record. just declare it with the same type of it_final.

loop at it_final. "final table

if ls_final = lt_final. "here give the key fields which needs to be compared.

"populate record into erro table

else.

ls_final = lt_final. "Populating the current record into ls_final for back up.

endif.

clear lt_final.

endloop.

let me know if its not clear..

Thanks,

Lakshmi

[former_member184635]

Hi Kelvin,

The code is simple, no need to use so much variables.

Just use the below code:

SORT <ft_csv_data> BY guid.

               
lt_contacts = <ft_csv_data>.

DELETE ADJACENT DUPLICATES FROM lt_contacts COMPARING guid.

if sy-tfill is not initial.

   Message 'Duplicate data exist' type I.

endif.

regards,

Dadarao.

[Former Member]

Hi Dadarao,

After this how to write the duplicate data into error file?

Thank you.

[Former Member]

Hi,

You can try the follwoing:

data: lv_flag type flag.

lt_contact_temp[] = lt_contacts[].

sort lt_contacts by guid.

sort lt_contacts_temp by guid.

delete adjacent duplicates from lt_contacts_temp comparing guid.

clear: lw_contacts.

loop at lt_contacts into lw_contacts.

clear: lw_contact_temp.

read table lt_contacts_temp into lw_contacts_temp with key guid = lw_contacts-guid binary search.

if sy-subrc  eq 0.

lv_flag = 'X'.

if lv_flag is initial.

move-corresponding lw_contacts to lw_error.

append lw_error to li_error.

clear:lw_error,lv_flag.

endif.

endif.

clear: lw_contacts.

endloop.

Hope this will solve your probelm.

[Former Member]

Hi Prasad,

I have solve the issue with below code:

lt_contacts        = <ft_csv_data>.
   SORT <ft_csv_data> BY guid.
   SORT lt_contacts   BY guid.
   v1 = lines( lt_contacts ).
   DELETE ADJACENT DUPLICATES FROM <ft_csv_data> COMPARING guid.
   v2 = lines( <ft_csv_data> ).

   LOOP AT <ft_csv_data> INTO ls_contacts_temp.
   ENDLOOP.

   IF v1 NE v2.
     MESSAGE: 'Duplicate data exist!' TYPE 'I'.

     LOOP AT lt_contacts INTO ls_contacts WHERE guid NE ls_contacts_temp-guid.

       CONCATENATE 'Record no: ' ls_contacts-client 'contain duplicate data at field <GUID>.'
                   'Duplicate data as:' ls_contacts-guid INTO l_error SEPARATED BY space.

       wa_error-line = l_error.
       APPEND wa_error TO gt_error.

     ENDLOOP.
     PERFORM error_file.
   ELSE.
     MESSAGE: 'No duplicate data!' TYPE 'I'.
   ENDIF.

Its working, but not the best method i think. Any best suggestion?

[Former Member]

Hi,

Try to do the code suggestion of . Then add the following

• Check if duplicates_found is not initial.
• If yes, then display your error message. Get the value of itab2.
• If no, display your success message.

Regards,

Jake.

[RaymondGiuseppi]

You could adapt the following code

SORT itab BY guid.
LOOP AT itab ASSIGNING wa.
  AT NEW guid.
    new_guid_record = abap_true.
  ENDAT.
  IF new_guid_record IS INITIAL.
    APPEND wa TO itab2.
    DELETE itab.
    duplicates_found = abap_true.
  ELSE.
    CLEAR new_guid_record.
  ENDIF.
ENDAT.

Regards,

Raymond

[Former Member]

  Hi Alexender,

following is the entire code you require:

TYPES: BEGIN OF ty_test,

field1(3),

field2(3),

field3(3),

END OF ty_test.

DATA: it_test TYPE STANDARD TABLE OF ty_test WITH HEADER LINE,

old_f1(3),

old_f3(3),

g_error_file TYPE string,

test_string TYPE string,

c_log_path TYPE filepath-pathintern VALUE 'ZOUT',

g_v_uxdir_xml TYPE rlgrap-filename.

it_test-field1 = 'AAA'. it_test-field2 = 'AAA'. it_test-field3 = 'DDD'. APPEND it_test.

it_test-field1 = 'BBB'. it_test-field2 = 'AAA'. it_test-field3 = 'CCC'. APPEND it_test.

it_test-field1 = 'AAA'. it_test-field2 = 'AAA'. it_test-field3 = 'AAA'. APPEND it_test.

it_test-field1 = 'BBB'. it_test-field2 = 'AAA'. it_test-field3 = 'CCC'. APPEND it_test.

SORT: it_test BY field1 field3.

CALL FUNCTION 'ZFILE_GET_PATH_NAME'

EXPORTING

logical_path = c_log_path

IMPORTING

file_name_path = g_v_uxdir_xml

EXCEPTIONS

path_not_found = 1

missing_parameter = 2

operating_system_not_found = 3

file_system_not_found = 4

OTHERS = 5.

CLEAR g_error_file.

CONCATENATE g_v_uxdir_xml 'test_error_file.txt' INTO g_error_file.

OPEN DATASET g_error_file FOR OUTPUT IN TEXT MODE ENCODING UTF-8.

LOOP AT it_test.

IF old_f1 = it_test-field1 AND old_f3 = it_test-field3.

CLEAR test_string.

CONCATENATE test_string it_test-field1 it_test-field2 it_test-field3 INTO test_string SEPARATED BY ' '.

TRANSFER test_string TO g_error_file.

ENDIF.

*endloop.

old_f1 = it_test-field1.

old_f3 = it_test-field3.

ENDLOOP.

* Close the dataset

CLOSE DATASET g_error_file.

IF sy-subrc <> 0.

WRITE : 'Error closing Error file'.

ELSE.

WRITE: 'error file created'.

ENDIF.

Thanks and regards,

Aastha