topic Re: checking duplicate in Application Development and Automation Discussions

checking duplicate

Former Member — Wed, 16 May 2007 04:01:38 GMT

Hi Experts,

I would like to check that, are there some duplicate rows in internal table?

I can chek it by hard code and I think it will not work when working with a lot of row.

Could you please tell me the solutions for this case?

Is there FM for this?

Thanks in advance.

Regards,

Tiwa.

Here is my code

REPORT.

DATA: BEGIN OF t_test OCCURS 0,

f1, f2, f3,

END OF t_test.

DATA: t_tmp1 LIKE t_test OCCURS 0 WITH HEADER LINE,

t_tmp2 LIKE t_test OCCURS 0 WITH HEADER LINE.

DATA found TYPE I.

PERFORM add USING 'a' 'b' 'c'.

PERFORM add USING 'a' 'c' 'c'.

PERFORM add USING 'a' 'b' 'c'.

PERFORM add USING 'a' 'd' 'd'.

PERFORM add USING 'a' 'c' 'c'.

t_tmp1[] = t_test[].

t_tmp2[] = t_test[].

LOOP AT t_tmp1.

CLEAR found.

LOOP AT t_tmp2.

IF ( t_tmp1-f1 EQ t_tmp2-f1 ) AND

( t_tmp1-f2 EQ t_tmp2-f2 ) AND

( t_tmp1-f3 EQ t_tmp2-f3 ).

ADD 1 TO found.

ENDIF.

ENDLOOP.

IF found > 1.

WRITE: / t_tmp1-f1, t_tmp1-f2, t_tmp1-f3, 'Duplicate'.

DELETE t_tmp1 WHERE f1 = t_tmp1-f1

AND f2 = t_tmp1-f2

AND f3 = t_tmp1-f3.

DELETE t_tmp2 WHERE f1 = t_tmp1-f1

AND f2 = t_tmp1-f2

AND f3 = t_tmp1-f3.

ENDIF.

ENDLOOP.

FORM add USING f1 f2 f3.

t_test-f1 = f1.

t_test-f2 = f2.

t_test-f3 = f3.

APPEND t_test.

ENDFORM.

Re: checking duplicate

Former Member — Wed, 16 May 2007 04:05:19 GMT

you can use :

Delete adjacent duplicated from Itab comparing <field>

eg :

Delete adjacent duplicated from Itab comparing BUKRS.

This will give you only one entry in the internal table for a perticular BUKRS

sort the itab first.

please reward if useful.

reagrds,

Message was edited by:

Sachin123

Re: checking duplicate

Sougata — Wed, 16 May 2007 04:06:09 GMT

Use syntax:

DELETE ADJACENT DUPLICATES FROM itab COMPARING field1 field2...

Reward points if useful.

Regards.

Re: checking duplicate

Former Member — Wed, 16 May 2007 04:09:20 GMT

hi,

for checking the duplicates exist..

Loop your internal table..

eg:

loop at itab into wa.

read table itab into wa1 with key f1 = wa-f1.

if sy-subrc = 0.

if ( wa-f2 = wa1-f2 AND wa-f3 = wa1-f3). (LKike this check to no of fields..

write: 'Duplicate exists for row:", wa-f1, wa-f2, wa-f3.

endif.

endloop.

IF you want to delete duplicate entries then see above post..

DELETE ADJACENT DUPLICATES FROM ITAB COMPARING ALL FIELDS.

Sort your table before using this, because once I had a problem in deleteing adjacent duplicates.. when I sorted the table It was working fine..

rewards if useful

regards,

nazeer

Message was edited by:

nazeer shaik

Re: checking duplicate

Former Member — Wed, 16 May 2007 04:12:12 GMT

Hi Sachin123,

Thanks for your reply.

I known that I can use delete adjacent for delete duplicates row by comparing key... but in this case I want to know which rows, data duplicates.

Regards,

Tiwa.

Re: checking duplicate

Former Member — Wed, 16 May 2007 04:15:24 GMT

good

in this case you can use like this

1-first use sort internal table to sort all the fields of internal table

2-second use delete itab to delete the null field of the internal table

thanks

mrutyun^

Re: checking duplicate

Former Member — Wed, 16 May 2007 04:19:39 GMT

HI,

see my previous post..It clearly says that duplicate entry exists for the particular row..

I am posting the code again...

wa and wa are work area of type internal table type

loop at itab into wa.

read table itab into wa1 with key f1 = wa-f1.

if sy-subrc = 0.

if ( wa-f2 = wa1-f2 AND wa-f3 = wa1-f3). (LKike this check to no of fields..

<b>write: 'Duplicate exists for row:", wa-f1, wa-f2, wa-f3.</b> "else you can give index here

endif.

endloop.

rewards if useful

regards,

nazeer

Re: checking duplicate

Former Member — Wed, 16 May 2007 04:20:59 GMT

one more method

loop at itab1.

read table itab2 with key f1 eq itab1-f1 f2 eq itab1-f2 f3 eq itab1-f3.

if sy-subrc eq 0.

write : 'duplicate exists for ', itab1-f1, itab1-f2, itab1-f3.

else.

move-corresponding itab1 to itab2.

append itab12.

endloop.

then use delete adjacent duplicates to remove duplicate entries

if helpful, reward

Sathish. R

Re: checking duplicate

Former Member — Wed, 16 May 2007 04:24:14 GMT

Loop on Itab1.

Read ITAB2 with all keys of ITAB1.

if sy-subrc = 0.

duplicate.else not.

regards,

Re: checking duplicate

Former Member — Wed, 16 May 2007 04:32:05 GMT

Hi All,

If itab2 be assigned data from itab1 then all of them are same,

loop at itab1.

read table itab2 with key f1 eq itab1-f1 f2 eq itab1-f2 f3 eq itab1-f3.

sy-subrc = 0 always

Do you think so?

Regards,

Tiwa.

Re: checking duplicate

Former Member — Wed, 16 May 2007 04:36:05 GMT

Hi,

This code helps you out.

define a variable wrk_tabix like sy-tabix and apply this logic.

loop at t_tmp1.

wrk_tabix = sy-tabix.

loop at t_tmp2 where f1 = t_tmp1-f1 and

f2 = t_tmp1-f2 and

f3 = t_tmp1-f3.

write:/ 'duplicate', t_tmp2-f1,t_tmp2-f2,t_tmp2-f3.

delete t_tmp1 index wrk_tabix.

delete t_tmp2 index wrk_tabix.

endloop.

Regards,

Re: checking duplicate

Former Member — Wed, 16 May 2007 04:40:11 GMT

nope...first time, itab2 will not contain any data of itab1..so subrc is 4, in that loop moves record of itab1...next loop, it reads the itab2 with keys of itab1, if it finds one then, it means a record of same entries in current loop of itab1 exists, which means that the current line of itab1 is a duplicate record...reply back for queries...

execute this code and see the output

DATA: BEGIN OF t_test OCCURS 0,

f1, f2, f3,

END OF t_test.

DATA: itab1 LIKE t_test OCCURS 0 WITH HEADER LINE.

PERFORM add USING 'a' 'b' 'c'.

PERFORM add USING 'a' 'c' 'c'.

PERFORM add USING 'a' 'b' 'c'.

PERFORM add USING 'a' 'd' 'd'.

PERFORM add USING 'a' 'c' 'c'.

LOOP AT t_test.

read table itab1 with key f1 = t_test-f1 f2 = t_test-f2 f3 = t_test-f3.

if sy-subrc eq 0.

write : / 'duplicate exists for ', t_test-f1, t_test-f2, t_test-f3.

else.

move-corresponding t_test to itab1.

append itab1.

endif.

endloop.

FORM add USING f1 f2 f3.

t_test-f1 = f1.

t_test-f2 = f2.

t_test-f3 = f3.

APPEND t_test.

ENDFORM.

if helpful, reward

Sathish. R

Re: checking duplicate

Former Member — Wed, 16 May 2007 04:48:44 GMT

Hi Sathish R,

I use your code to apply and the problem is solved

I have already give you 10 points.

Thanks everyone,

Tiwa.

DATA: w_tmp1 LIKE t_tmp1.

LOOP AT t_tmp1.

IF t_tmp1 EQ w_tmp1.

WRITE: / t_tmp1-f1, t_tmp1-f2, t_tmp1-f3, 'Duplicates'.

DELETE t_tmp1 WHERE f1 = t_tmp1-f1

AND f2 = t_tmp1-f2

AND f3 = t_tmp1-f3.

ENDIF.

w_tmp1 = t_tmp1.

ENDLOOP.

Re: checking duplicate

Sougata — Wed, 16 May 2007 05:22:42 GMT

Its great you have given him 10 points but what about the rest of us who took the time to read your issue and respond trying to help you out?? Do you think we have nothing better to do at work??

All replies should at least be rewarded with 2 points....please read the <b>Rules of Engagement</b> at the top of the forum threads.

Thanks.