Problem processing results from SO_OBJECT_READ

[Former Member]

I am calling function SO_OBJECT_READ and it is returning to me a table called OBJCONT which has one field which is a 255 char field. This is the message content of the SBWP email. It seems to split each line using two # like this:

This is my first line of text.##This is my second line of text.##This is my third line of text.

I am trying to process this string by finding the # symbol in the field. I have tried the following but nothing works:

FIND '#' IN OBJCONT-LINE.

IF OBJCONT-LINE CS '#'.

VAR-POS = SY-FDPOS.

ENDIF.

SPLIT OBJCONT-LINE AT '#' INTO TABLE OBJECT_CONTENTS.

None of the above examples seem to process the '#' characters. It is almost like it is a placeholder for a Carriage return or Line Feed? Has anyone come across this problem before? If I declare another field within my program I can process '#' fine using exact same examples described above.

EG.

DATA: VAR-TEST(10) TYPE C.

VAR-TEST = '123###456'.

FIND '#' IN VAR-TEST MATCH OFFSET VAR-POS.

So why does it not work for the table that I have returned from SO_OBJECT_READ?

[former_member156446]

when ever the person hits an enter in the text field there the # is triggered in the field.. which is called a line field as u suspected.

I had a scenario where I had a # at the end so I found the string length and removed the last character..

If I remember there is some alternative like decelaring the # like

data: c_feed type X value '/OD'.

this doesn't work for a unicode system though.

[Former Member]

The problem is that the ## is repeated throughout the string just not at the end. I can not change the results that the standard SAP function gives me so I need to somehow process the two ##'s?

[venkat_o]

Hi Brad, Just remove ## using the following code. It just replaces with space. Try it . It works i believe.

DATA objcont LIKE soli  OCCURS 0 WITH HEADER LINE.

objcont-line = 'This ##is my 1 line##'.
APPEND objcont.
CLEAR  objcont.
objcont-line = 'This ##is my 2 line##'.
APPEND objcont.
CLEAR  objcont.
objcont-line = 'This ##is my 3 line##'.
APPEND objcont.
CLEAR  objcont.
objcont-line = 'This ##is my 4 line##'.
APPEND objcont.
CLEAR  objcont.
objcont-line = 'This ##is my 5 line##'.
APPEND objcont.
CLEAR  objcont.

REPLACE ALL OCCURRENCES OF REGEX '##'
  IN TABLE objcont WITH space.

LOOP AT objcont .
  WRITE / objcont.
ENDLOOP.

Regards, Venkat.O

[Former Member]

Hi Venkat

That was a good suggestion, I tried your REGEX command but it said "the word regex is reserved"..?

I then tried:

 REPLACE ALL OCCURENCES OF '##' IN OBJCONT-LINE WITH SPACE. 

This gave me SY-SUBRC of 4.

I then tried:

 TRANSLATE OBJCONT-LINE USING '# '. 

This did not return errors but it did not change the contents of the line at all. This is definitely not a normal # symbol we are dealing with...

[venkat_o]

Hi Brad, "the word regex is reserved" ..Is this syntax error or what ? I am able to do in ECC 6.0 . Regards, Venkat.O

[Former Member]

Yes a syntax error, I am on ECC 5

[JayChan]

hi,

I also ran into the same problem about yours.

My system runs on ECC 5.0

please refer my post:

[|]

any idea would be appreciated.

Chan Chih-Chieh

[Former Member]

This will work.

1. is Line Feed Character

REPLACE ALL OCCURRENCES OF CL_ABAP_CHAR_UTILITIES=>CR_LF IN wa_objcont-line WITH ' '.

[Former Member]

Also just to clarify other things I have tried...

I thought it may have something to do with the character field being 255 characters. So I took the first 72 characters and placed in another temporary field and tried to remove the '#' symbols from the new 72 character field. Still no success...

Declaring new fields with # symbol works fine though and allows removal.

[Former Member]

Anyone else have any ideas?

Would be greatly appreciated.

Thanks...

[venkat_o]

Hi Brad, Try this. It is working.

REPORT  zvenkat_notepad2.

DATA objcont LIKE soli  OCCURS 0 WITH HEADER LINE.

objcont-line = 'This ##is my 1 line##'.
APPEND objcont.
CLEAR  objcont.
objcont-line = 'This ##is my 2 line##'.
APPEND objcont.
CLEAR  objcont.
objcont-line = 'This ##is my 3 line##'.
APPEND objcont.
CLEAR  objcont.
objcont-line = 'This ##is my 4 line##'.
APPEND objcont.
CLEAR  objcont.
objcont-line = 'This ##is my 5 line##'.
APPEND objcont.
CLEAR  objcont.

DATA: offset TYPE i.

LOOP AT objcont .
  REPLACE ALL OCCURRENCES OF '##' IN SECTION LENGTH 255 OF  objcont WITH ''.
  MODIFY objcont INDEX sy-tabix.
  CLEAR   objcont.
ENDLOOP.

LOOP AT objcont .
WRITE / objcont.
ENDLOOP.

regards, Venkat..O

[Former Member]

Hi Venkat

This didn't work for me either, I am thinking this is a special character passed from the SO_OBJECT_READ function. However I have managed to get a temporary solution working, the problem is I can not determine where the new line begins as no commands are processing the ## symbols.

[Former Member]

I have also found that if I write to my ZTABLE and try to select from the ZTABLE using ## as my selection criteria it does not find the records.

This raises a new question.. What does ## mean when coming from the Database? Is this some sort of Unicode conversion? I think it is just a placeholder for a CRLF.

[Former Member]

I ended up processing it by writing the character to db table then referencing it. I think that it was a hex character and their is a class which can process this character but I didn't research any further.