topic Re: Replace empty records in a file in Application Development and Automation Discussions

Replace empty records in a file

Former Member — Tue, 23 Apr 2013 09:25:21 GMT

Hi experts,

I am generating a file trough an abap program. I am retrieving my records in a BW infocube (to simplify let's say a SAP table).

The issue is that empty information are generated as ‘#’ in my file. I would like to suppress all ‘#” from my file which contains 5 columns.

The file today is like this:

Field A	Field B	Field C	Field D	Field E
A	AA	AAA	AAAA	20
B	BB	BBB	#	55

What I would lik to do is this:

Field A	Field B	Field C	Field D	Field E
A	AA	AAA	AAAA	20
B	BB	BBB		55

I don’t know if in abap there exit something like this (i am also looking for a performant statement because my file may contain too much records):

Loop at all columns of the file
REPLACE ALL OCCURRENCES OF '#' WITH ' '. .
ENDLOOP.

Thanks for your advices.

Amine

Re: Replace empty records in a file

former_member210266 — Tue, 23 Apr 2013 09:43:35 GMT

Hi Amine

You can use REPLACE ALL OCCURRENCE IN TABLE command

Regards

Swati

Re: Replace empty records in a file

gouravkumar64 — Tue, 23 Apr 2013 09:44:40 GMT

Hi,

As # is some garbage character.

You can try like this with this class .

REPLACE ALL OCCURRENCES OF cl_abap_char_utilities=>cr_lf in text with space.

There are few thread on this,search also.

Thanks

Gourav.

Re: Replace empty records in a file

Former Member — Tue, 23 Apr 2013 09:47:00 GMT

Hi Amine,

The coding works fine in ABAP but here the # is coming while display in Excel.

Even if you write code in Customer Exit will not work.

Re: Replace empty records in a file

RaymondGiuseppi — Tue, 23 Apr 2013 09:52:32 GMT

Use field symbols during LOOP, and as BW set the field to # for initial values, don't use a CPU consuming statement like REPLACE but check for each subfield values.

LOOP AT itab ASSIGNING <record>.
   DO 5 TIMES.
     ASSIGN COMPONENT sy-index OF STRUCTURE <record> TO <fs>.
     IF <fs> EQ '#'.
       CLEAR <fs>.
     ENDIF.
   ENDDO.
 ENDLOOP.

Regards,

Raymond

Re: Replace empty records in a file

Former Member — Tue, 23 Apr 2013 12:55:51 GMT

Hi Sai,

Actually, when i consult my file in AL11, the '#" occurences are already present...

Amine

Re: Replace empty records in a file

Former Member — Tue, 23 Apr 2013 12:57:50 GMT

Hi Raymond,

Thank you for your answer.

What would be <record>? Table/Column..... How should i declare it?

And <fs> i guess it's field symbol.How do you advice me to declare it? (type all?)

Thanks.

Amine

Re: Replace empty records in a file

RaymondGiuseppi — Tue, 23 Apr 2013 13:37:06 GMT

Both Field symbols,for <FS> TYPE ANY, for <RECORD>, eithe ANY or actual type.

Regards,

Raymond

Re: Replace empty records in a file

Former Member — Tue, 23 Apr 2013 13:41:24 GMT

Ok thanks. I will let you know after my tests on the system.

Amine

Re: Replace empty records in a file

Clemenss — Tue, 23 Apr 2013 13:51:47 GMT

This really scares me: I don't know where is documentation for this: In SAP, # usually means the character can not be displayed because it is a control character like tab, newline etc.

Please use debugger, check hexadecimal value of what looks like # and replace using CL_ABAP_CHAR_UTILITIES attributes, i.e. HORIZONTAL_TAB, VERTICAL_TAB, NEWLINE, CR_LF, FORM_FEED or BACKSPACE.

Regards

Clemens

Re: Replace empty records in a file

Former Member — Tue, 23 Apr 2013 15:44:11 GMT

Thx Clemens for your advice.

Amine

Re: Replace empty records in a file

Former Member — Wed, 24 Apr 2013 12:28:21 GMT

Hi Raymond,

I have a question regarding the statment : DO 5 times

It's because i have 5 columns, right?

Amine

Re: Replace empty records in a file

RaymondGiuseppi — Wed, 24 Apr 2013 12:47:55 GMT

Yes, you could also use RTTI tools (like class CL_ABAP_TYPEDESCR) to calculate number of sub-fields for a more generic tool.

Regards,

Raymond

Re: Replace empty records in a file

Former Member — Wed, 24 Apr 2013 12:51:10 GMT

Ok thanks for your explanations.

Amine