Need to compare two fields and populate the other fields.

Question

Hi All,
I have scenario like this.
DATA: text1(150) type C
VALUE 'Step: SHORT DUMP; Scenario: PRE; Date :09.07.2007. Time :13:08:33.'.
DATA: text2(150) TYPE C
      VALUE 'Step: &; Scenario: &; &',
DATA: text_A(150) TYPE C,
          text_B(150) TYPE C,
           text_C(150) TYPE C.
Now I need to compare text1 and text2 and populate text_A, text_B and text_C as follows.
text_A = SHORT DUMP
text_B = PRE
text_C = Date :09.07.2007. Time :13:08:33.
Basically I need to fill text_A, text_A and text_A with the values in text1 which are replaced by '&'.
Can anyone suggest me a code snippet to do it?
Thanks in advance.
Regards
Ankit

Answer 1

Sorry for slow response - I'm in a different timezone I think... Well it's not a short algorithm, but you could try something like the following (there's probably a sweeter algorithm, but this is what occured to me overnight)... I realise that it's not foolproof as you could have words substituted in for the "&" that also appear in the original message text and that would skew the results...
Jonathan
report zlocal_jc_get_msgvars.
types:
  gty_char150(150)      type c.
parameters:
  p_text_1              type gty_char150 lower case,
  p_text_2              type gty_char150 lower case.
initialization.
  perform initialization.
start-of-selection.
  perform extract_msgv.
*&      Form  initialization
form initialization.
*" Default screen variables
  concatenate
    'Step: SHORT DUMP; Scenario: PRE;'
    'Date :09.07.2007. Time :13:08:33.'
    into p_text_1 separated by space.
  p_text_2 = 'Step: &; Scenario: &; &'.
endform.                    "initialization
*&      Form  extract_msgv
form extract_msgv.
*" Work out the values that have been substituted
*" into p_text2 to make p_text1
  data:
    l_text_a            type gty_char150,
    l_text_b            type gty_char150,
    l_text_c            type gty_char150,
    l_text_d            type gty_char150,
    l_tabix_1           type i,
    l_tabix_2           type i,
    l_word_1            type gty_char150,
    lt_word_1           type gty_char150 occurs 10,
    l_word_2            type gty_char150,
    lt_word_2           type gty_char150 occurs 10,
    l_result            type gty_char150,
    l_tabix             type sy-tabix,
    l_tabix_next        type sy-tabix,
    l_tabix_from        type sy-tabix,
    l_tabix_to          type sy-tabix,
    lt_tabix            type sy-tabix occurs 4,
    lt_result           type gty_char150 occurs 10.
*" Break the strings into words
  split p_text_1 at space into table lt_word_1.
  split p_text_2 at space into table lt_word_2.
  format reset.
  format color col_total.
  write: / p_text_1(80).
  write: / p_text_2(80).
  format reset.
*" Look at where the words in the shorter string appear in the longer
  loop at lt_word_2 into l_word_2.
    add 1 to l_tabix_2.
    write: / 'P_TEXT_2, Word', l_tabix_2, l_word_2.
    if l_word_2+0(1) = '&'.  "placemarker
      continue.
    endif.
*" examine the expanded version
    read table lt_word_1 into l_word_1
      with key = l_word_2.
    if not sy-subrc is initial.
      write: /
        'ERROR: Word not found in expanded version' color col_negative.
      continue.
    endif.
    l_tabix_1 = sy-tabix. "Where we found word from short msg
    append l_tabix_1 to lt_tabix.
  endloop.
*" add pointer to end of list of words too
  describe table lt_word_1 lines l_tabix_1.
  add 1 to l_tabix_1. "because we subtract 1 below...
  append l_tabix_1 to lt_tabix.
  loop at lt_tabix into l_tabix_from.
    l_tabix = sy-tabix.
    write: / l_tabix_from.
*" get the range of words we want by getting next tabix row
*" and subtracting 1
    l_tabix_next = l_tabix.
    add 1 to l_tabix_next.
    read table lt_tabix into l_tabix_to index l_tabix_next.
    if not sy-subrc is initial. "no more words...
      exit.
    endif.
    add      1 to   l_tabix_from.
    subtract 1 from l_tabix_to.
    loop at lt_word_1 into l_word_1
      from l_tabix_from to l_tabix_to.
      write: / l_word_1 color col_positive.
* Push into text_A, text_B, text_C, text_D
      case l_tabix.
        when 1.
          concatenate l_text_A l_word_1 into l_text_a
            separated by space.
        when 2.
          concatenate l_text_b l_word_1 into l_text_b
            separated by space.
        when 3.
          concatenate l_text_c l_word_1 into l_text_c
            separated by space.
        when 4.
          concatenate l_text_d l_word_1 into l_text_d
            separated by space.
      endcase.
    endloop.
  endloop.
  uline.
  format reset.
  format color col_group.
  write: / 'TEXT_A', l_text_a(80).
  write: / 'TEXT_B', l_text_b(80).
  write: / 'TEXT_C', l_text_c(80).
  write: / 'TEXT_D', l_text_d(80).
  format reset.
endform.                    "extract_msgv