Sort inernal table using 2 fields and only modify first record from a list

[Former Member]

Hi All,

I have a requirement in BW to transform HR related data for internal use. For this purpose i have created an APD (Analysis Process Designer) which reads data from a database table and writes it into a CSV file by trasnforming only some records using an ABAP routine.

Exmaple. I have ORGUNIT and DATETO as 2 fields in the source table.

ORGUNIT | DATETO |

12345678 | 31.12.2011

12345678 | 31.11.2010

12345678 | 31.10.2009

12345678 | 31.08.2008

87654321 | 31.12.9999

87654321 | 31.12.2012

87654321 | 31.10.2011

Target CSV file should have output like below

12345678 | 31.12.9999

12345678 | 31.11.2010

12345678 | 31.10.2009

12345678 | 31.08.2008

87654321 | 31.12.9999

87654321 | 31.12.2012

87654321 | 31.10.2011

in the above example ORGUIT 87654321 has no change, where as in 12345678 we need to sort the ORGUNIT by DATETO field in DESCENDING ORDER and check the first record's DATETO value. IF first record's DATETO value is not equal to 31.12.9999, we only need to modify the first record with value as 31.12.9999. rest of the records will remain same.

For this requirement i used the below code in my routine and struggling to get the logic working.

DATA: ls_ source type y_source_fields,

           ls_target type y_target_feilds.

SORT itab by ORGUNIT ASCENDING DATETO DESCENDING.

loop at itab into ls_source.

move-corresponsing ls_source to ls_target.

if ls_source-DATETO ne '99991231'

ls_target-DATETO = '99991231'

Append ls_target to et_target.

Endloop.

I am not an ABAPer .but trying to do this by myself. The above code gives me DATETO fields for each record as 31.12.9999.

Can any one help me here please.

Thanks in advance.

[Former Member]

Hi All,

I have a requirement in BW to transform HR related data for internal use. For this purpose i have created an APD (Analysis Process Designer) which reads data from a database table and writes it into a CSV file by trasnforming only some records using an ABAP routine.

Exmaple. I have ORGUNIT and DATETO as 2 fields in the source table.

ORGUNIT | DATETO |

12345678 | 31.12.2011

12345678 | 31.11.2010

12345678 | 31.10.2009

12345678 | 31.08.2008

87654321 | 31.12.9999

87654321 | 31.12.2012

87654321 | 31.10.2011

Target CSV file should have output like below

12345678 | 31.12.9999

12345678 | 31.11.2010

12345678 | 31.10.2009

12345678 | 31.08.2008

87654321 | 31.12.9999

87654321 | 31.12.2012

87654321 | 31.10.2011

in the above example ORGUIT 87654321 has no change, where as in 12345678 we need to sort the ORGUNIT by DATETO field in DESCENDING ORDER and check the first record's DATETO value. IF first record's DATETO value is not equal to 31.12.9999, we only need to modify the first record with value as 31.12.9999. rest of the records will remain same.

For this requirement i used the below code in my routine and struggling to get the logic working.

DATA: ls_ source type y_source_fields,

           ls_target type y_target_feilds.

SORT itab by ORGUNIT ASCENDING DATETO DESCENDING.

loop at itab into ls_source.

move-corresponsing ls_source to ls_target.

if ls_source-DATETO ne '99991231'

ls_target-DATETO = '99991231'

Append ls_target to et_target.

Endloop.

I am not an ABAPer .but trying to do this by myself. The above code gives me DATETO fields for each record as 31.12.9999.

Can any one help me here please.

Thanks in advance.

[Former Member]

Dear Former Member,

Go like this.

DATA: ls_ source type y_source_fields,

           ls_target type y_target_feilds.

SORT itab by ORGUNIT ASCENDING DATETO DESCENDING.

loop at itab into ls_source.

At New  ORGUNIT.

if ls_source-DATETO ne '99991231'

ls_target-DATETO = '99991231'

end at.

Append ls_target to et_target.

Endloop.

With Regards,

Akshay

[Former Member]

Hi thanks for your help.

But that doesn't solve my problem. The result is not what i wanted. if i don't use move corresponding statement at the start of the loop, that will write all other field values to zero except DATETO.

any other thougths please ??

[Former Member]

Hi All,

Can any one please help me with the code here please.

I am using the code as below, but it is not working as i wanted.

DATA: LS_SOURCE TYPE y_source_fields,
          LS_TARGET TYPE y_target_fields.
SORT it_SOURCE BY ORGUNIT ASCENDING DATETO DESCENDING.
LOOP AT IT_SOURCE into ls_source. ..

MOVE-CORRESPONDING

ls_source to ls_target.

at new ORGUNIT.
     if ls_source-dateto ne '99991231'.
     ls_target-dateto = '99991231'.
endif.
endat.

append ls_target to et_target.
SORT ET_TARGET BY NODEID ASCENDING.
ENDLOOP.

[kumud]

Hi,

Could you please show how have you defined the types y_source_fields and y_target_fields?

Regards,

Kumud

[Former Member]

Hi POPS,

What output you are getting..?

Are you getting " ****** " in date field....???

if you are getting " ****** " in date filed than go for the following modified code.

DATA: LS_SOURCE TYPE y_source_fields,
          LS_TARGET TYPE y_target_fields.

DATA: Old_date type  ls_target-dateto,

             New_date type ls_target-dateto.

SORT it_SOURCE BY ORGUNIT ASCENDING DATETO DESCENDING.
LOOP AT IT_SOURCE into ls_source. ..

MOVE-CORRESPONDING

ls_source to ls_target.

old_date = ls_target-dateto.

at new ORGUNIT.
     if old_date ne '99991231'.
       new_date = '99991231'.
    else    

       new_date =  old_date.

    endif.
endat.

ls_target-dateto = new_date.

clear new_date.

clear old_date.

append ls_target to et_target.
SORT ET_TARGET BY NODEID ASCENDING.
ENDLOOP.

With Regards,

Akshay