获取 FIND OF REGEX 语句的子匹配项

Question

I would not only check a string for any occurrence of a substring using regex but also get the result ("45" in my example) of the found regex.

I tried:

 DO.
    FIND FIRST OCCURRENCE OF REGEX '[$][0-9]+[$]' IN 'safmaoigrevnwnfuwifhd$45$sdffge' 
        MATCH OFFSET lv_match_offset MATCH LENGTH lv_match_len 
        SUBMATCHES DATA(s1) DATA(s2) DATA(s3).
    IF sy-subrc = 0.
      " I expected s2 to contain the value between the "$"
      lv_match-value = s2.
      " remove found substring from annotation name for next loop
      lv_annotation_name = lv_annotation_name+lv_match_offset(lv_match_len).
    ELSE.
      EXIT.
    ENDIF.
  ENDDO.

s1, s2 and s3 are always initial but why? I thought that they would contain the value of the found regex, eg "$", "45", "$". What am I doing wrong? Is there a more elegant way than my approach?

Following Sandra's suggestion I changed the code to

FIND ALL OCCURRENCES OF REGEX lv_index_regex IN lv_annotation_name RESULTS lt_matches. 

Note that FIND ALL ... MATCHES does not exist, I assumed Sandra meant "RESULTS". lt_matches then contains the itab submatches but its always empty....

Answer 1

The word SUBMATCHES in the statement FIND [FIRST OCCURRENCE OF] REGEX is used to extract directly the registered subgroups (...) (AKA submatches) of the match into variables.

So, just put some parentheses around each of them:

FIND FIRST OCCURRENCE OF REGEX '([$])([0-9]+)([$])' 
    IN 'safmaoigrevnwnfuwifhd$45$sdffge' 
    MATCH OFFSET lv_match_offset 
    MATCH LENGTH lv_match_len 
    SUBMATCHES DATA(s1) DATA(s2) DATA(s3).

Note that your code will do an endless loop, but that's another question (it will be easier to find all matches at a time with FIND ALL OCCURRENCES OF REGEX ... RESULTS itab and loop at the internal table, each line being one match, and each line also contains the list of the submatches).