Using negation in Regular expression

[Former Member]

Hi All,

I need to write a pattern using regular expression where the word does not end with some characters.

Suppose I have a word ZBH_MM_CHECK, the word should start with ZBH I can write the pattern as ^ZBH[_][A-Z]{2}[_]\w  . If the end of the word should not contain _X how should we write such pattern. Here I should check that the word should not end with _X using regular expression. ZBH_MM_CHECK_X is invalid but ZBH_MM_CHECK is valid. I tried using (?!_X)$ at the end but failed. Please suggest.

Thanks in Advance.

[former_member201285]

This sample code might help:

    DATA: suffix_to_check(2) TYPE c VALUE '_X',
        pattern(50) TYPE c VALUE 'ZBH_MM_CHECK_X',
        offset TYPE i.

  offset = strlen( pattern ) - strlen( suffix_to_check ).

  IF pattern+offset = suffix_to_check.

    WRITE: 'Not valid!'.

  ENDIF.

[hendrik_brandes]

Hello,

I would suggest, you make an expression like this:

(^ZBH[_][A-Z]{2}[_]\w{5})(_X)?

Then you will have the _X in the second subgroup, if the expression match otherwiuse you have just one subgroup.

Try your expression within the report DEMO_REGEX_TOY.

Kind regards,

Hendrik

[Former Member]

While using regex, character negation can be done but not word negation.

Alternative would be to check value stored in substring like this.

DATA lv_substr TYPE string.
FIND REGEX '^ZBH_[A-Z]{2}_\w+_(.*)'
  IN 'ZBH_MM_CHECK_X'
  SUBMATCHES lv_substr .
IF sy-subrc EQ 0 AND lv_substr NE 'X'.
  "match
ENDIF.