changing some characters in string

[marcin_cholewczuk]

Hi all,

I want to change in string occurance of some unwanted characters with '_', but I know only good characters. Any idea how it can be done simply ?

[anversha_s]

hi,

DATA: int_char(10) type c value '0123456789'.

if var CO sy-abcde.
* only alphabets
elseif var CO int_char.
* only numberics
else.
Special cases.

you have to chk that <b>special cases</b>

Regards

Anver

[marcin_cholewczuk]

uhhhhhh, but that was not my question. I want to <b>change</b> all occurance of bad characters with '_'.

[Former Member]

Hi,

DATA: String(20) type c value 'sautyn_nbj_jjk',
          string1(20) type c,
          str(1) type c value '_'.

REPLACE str with space INTO string1.
CONDENSE string1 NO-GAP.

Ok so u can

DATA: String(20) type c value 'sautyn@nbj,jjk'.

<b>SPLIT STRING AT DELIMITERS INTO str1 str2 str3...</b>

depending on number of delimiters.

Then

<b>CONCATENATE str1 str2 str3 INTO string SEPERATED BY '_'.</b>

Hope this helps.

[Former Member]

Marcin,

To replace a string in a field with a different string, use the REPLACE statement.

REPLACE <str1> WITH <str2> INTO <c> [LENGTH <l>].

The statement searches the field <c> for the first occurrence of the first <l> positions of the pattern <str1>. If no length is specified, it searches for the pattern <str1> in its full length.

Then, the statement replaces the first occurrence of the pattern <str1> in field <c> with the string <str2>. If a length <l> was specified, only the relevant part of the pattern is replaced.

If the return code value of the system field SY-SUBRC is set to 0, this indicates that <str1> was found in <c> and replaced by <str2>. A return code value other than 0 means that nothing was replaced. <str1>, <str2>, and <len> can be variables.

DATA: T(10) VALUE 'abcdefghij',

STRING LIKE T,

STR1(4) VALUE 'cdef',

STR2(4) VALUE 'klmn'.

REPLACE STR1 WITH STR2 INTO STRING.

WRITE / STRING.

Pls. mark if useful

[anversha_s]

hi,

* Replacing and translating characters in strings
DATA: STRING(80).

STRING = 'Variable_The variable _ is substituted later.'.

REPLACE '_' WITH ' ' INTO STRING.
condense string no-gaps.

WRITE / STRING.

Regards

Anver

[Former Member]

hi,

*Code to Replace one char with another using TRANSLATE command

*----

---

DATA: ld_date(10) type c.

ld_date = '28.10.1979'.

*Finds all occurences of first char(.) & replaces them with second(-)

TRANSLATE ld_date using '.-'.

*ld_date will now contain '28-10-1979'

Regards,

Bijal

[Former Member]

Hi,

lets say yout string name is STRING1. and i am giving the code for Numbers, if you want alphabets then chage the code according, only change the INT_CHAR to length 26 and values should be a to z.

DATA: int_char(10) type c value '0123456789'.
Data: len type i.
data: first type i, 
        last type i.
first = 0.
last = 1.
describe field STRING1 length LEN.
do LEN times.
 STRING+first(last) CO INT_CHAR.
 if sy-subrc <> 0.
 replace STRING1+first(last)  with '_' into STRING1.
 endif.
first = first + 1.
last = last + 1.
Enddo.

Regards

Sudheer

Message was edited by:

Sudheer Junnuthula

[marcin_cholewczuk]

Hi

Sudheer your solution is what I thought of, but I was hoping that there is possibility to do this in simplier way. If I knew bad characters I would've done it like bijal proposed or by using OVERLAY instruction - that was kind of solution I was thinking about.

[Former Member]

Hi,

In Sudheer's reply u can do like this

<b>DATA: int_char(30) type c value '`~!@#$%^&*()_-+=|\"':;?/>.<,'.</b>

Data: len type i.
data: first type i, 
        last type i.
first = 0.
last = 1.
describe field STRING1 length LEN.
do LEN times.
 STRING+first(last)

<b>CA</b>

INT_CHAR.
 if sy-subrc <> 0.
 replace STRING1+first(last)  with '_' into STRING1.
 endif.
first = first + 1.
last = last + 1.
Enddo.

Hope this helps. Include all unwanted characters in the int_char.

[Former Member]

Hi marcin,

if u have good characters in variables say v1 v2 v3.

concatenate v1 '_' v2 '_' v3 into str.

regards,

keerthi