Last week I ended up with a task to convert special characters before writing them into a text file for a legacy system ( no #utf-8 support ) using :
OPEN DATASET lv_filename FOR APPENDING IN TEXT MODE ENCODING NON-UNICODE.
(Don't forget to check the authority upfront
link )
Some special letters as German "Ä ü ..." are accepted but others like "šũvá" are to be replaced. Replaced at a best by a close "friend": ũ -> u; š -> s to avoid names Like "#av#"
CALL FUNCTION 'SCP_REPLACE_STRANGE_CHARS'
helps a lot, but I hazzled so much to find a code page that fitted to my needs.
Here is the code of a small report that converts a "dirty" string and test if it fits to my requirements.
START-OF-SELECTION.
DATA lv_clean TYPE string.
DATA(lv_dirty) = 'Jáošũvá Алксй Äü'.
WRITE lv_dirty.
SELECT cpcodepage FROM tcp00a INTO TABLE @DATA(lt_codepages)
GROUP BY cpcodepage.
LOOP AT lt_codepages ASSIGNING FIELD-SYMBOL(<s_codepage>).
CALL FUNCTION 'SCP_REPLACE_STRANGE_CHARS'
EXPORTING
intext = lv_dirty
inter_cp = <s_codepage>-cpcodepage
replacement = '35' " Error Chracters are replaced as #
IMPORTING
outtext = lv_clean
EXCEPTIONS
invalid_codepage = 1
codepage_mismatch = 2
internal_error = 3
cannot_convert = 4
fields_not_type_c = 5
OTHERS = 6.
IF sy-subrc <> 0.
WRITE <s_codepage>-cpcodepage && ' error'.
CONTINUE.
ENDIF.
" Fo me only codepages that output german special characters are relevant.
IF contains( val = lv_clean sub = 'Äü' ).
WRITE / |{ <s_codepage>-cpcodepage } && ` : ` && { lv_clean } | .
CLEAR lv_clean.
ENDIF.
ENDLOOP.
EXIT.
With the report I found the SAP Codepage 1157 for my requirments fast.
Be aware, that the conversion takes a while. #performance
There is a
helpful wikipage on this topic as well by
sandrarossi Showing other classes that help on this topic as well. Thank for posting!
