How to Use AMDP Code to Handle Special Characters in SAP BW/4HANA Data Loads

If you are working with SAP BW/4HANA, you may encounter some issues with special characters in your flat files. Special characters are symbols or characters that are not part of the standard alphabet or numeric system, such as !, #, &, etc. These characters can cause errors or inconsistencies when loading data into SAP BW/4HANA, especially in the description fields.

In this blog post, we will show you how to use AMDP code to handle special characters in SAP BW/4HANA data loads. AMDP stands for ABAP Managed Database Procedures, which are database procedures written in ABAP and executed on the SAP HANA database. AMDP code can help you process and transform data faster and more efficiently than ABAP code.

Why Use AMDP Code for Handling Special Characters?

There are several reasons why you may want to use AMDP code for handling special characters in SAP BW/4HANA data loads:

AMDP code can run directly on the SAP HANA database, which means it can leverage the parallel processing and in-memory capabilities of SAP HANA. This can improve the performance and scalability of your data loads.
AMDP code can access and manipulate data from different sources and targets, such as tables, views, InfoObjects, etc. This can give you more flexibility and control over your data transformations.
AMDP code can use SQLScript, which is a scripting language for SAP HANA that supports advanced features such as loops, variables, expressions, etc. This can make your code more readable and maintainable.

How to Use AMDP Code for Handling Special Characters?

To use AMDP code for handling special characters in SAP BW/4HANA data loads, you need to follow these steps:

Create an AMDP class and method that contains the logic for handling special characters. You can use the example code below as a reference. The code uses regular expressions to replace any invalid or unwanted characters with a blank space or a configurable replacement. It also checks if the first character of the string is !, which indicates a comment or a header line, and removes it accordingly. The code also converts the string to uppercase or lowercase depending on the InfoObject settings.

class zcl_invalid_cha_rem_01 definition   public   final   create public .
  public section.
  interfaces if_amdp_marker_hdb.
  CLASS-methods INVALID_CHAR_REMOVAL importing VALUE (IV_IOBJ_NAME) TYPE RSCHABASNM   VALUE (IV_text) TYPE string   value (IV_lowr_case) type CHAR1   exporting VALUE (RESULT) TYPE string.
  protected section.
  private section.
endclass.

class zcl_invalid_cha_rem_01 implementation.
  method INVALID_CHAR_REMOVAL by database procedure   for hdb   language sqlscript.
    declare lv_flag varchar ( 1 );
    declare lv_length integer := 0;
    declare lv_counter integer := 1;
    RESULT = :iv_text;
    -- If the incoming string is not initial
    IF :RESULT != '' then
      -- To get the length of the string
      SELECT LENGTH ( :result ) into lv_length FROM "SYS"."DUMMY";
      -- Remove leading and trailing blank spaces
      select trim ( both '' from :result ) into result from "SYS"."DUMMY";
      -- if incoming string is greater than 1
      IF ( SELECT LENGTH ( :result ) FROM "SYS"."DUMMY" ) > 1 THEN
        -- check if the first char of the string is !, then take the string
        -- from the first char != ! to end of the string
        if SUBSTRING (:result, 1 , 1) = '!' then
          while :lv_counter <= :lv_length do
            if SUBSTRING (:result,:lv_counter,1) != '!' then
              break;
            end if ;
            lv_counter = :lv_counter + 1;
          end while ;
          if :lv_counter = :lv_length then
            result = '';
          else
            select SUBSTRING (:result,:lv_counter,:lv_length) into result from "SYS"."DUMMY";
          end if;
        end if ;
        -- Eliminate the # sign completely
        select REPLACE_REGEXPR (' ( [^ [:print:]|^ [\x {00C0}-\x {017F}]| [#])' IN :result with ' ' occurrence all) into result from "SYS"."DUMMY";
        if :iv_iobj_name != '' then
          -- Remove the select statement from AMDP
          -- instead pass it when calling the AMDP Procedure
          -- SELECT b.lowercase INTO lv_flag FROM rsdcha AS a INNER JOIN rsdchabas AS b
          -- ON a.chabasnm = b.chabasnm
          -- WHERE a.chanm = :IV_IOBJ AND
          -- a.objvers = 'A' AND
          -- b.objvers = 'A';
          -- if lowercase flag is not set then convert to upper case
          if :iv_lowr_case = '' then
            select upper ( :RESULT ) into RESULT from "SYS"."DUMMY" ;
          end if;
        end if;
      -- if incoming string is equal to 1
      else
        select replace ( :RESULT, '#', '') into result from "SYS"."DUMMY";
        if SUBSTRING (:result, 1 , 1) = '!' then
          RESULT = '';
        end if ;
      END IF;
    END IF;
    -- Trim to remove leading and trailing blank spaces before returning the result
    select trim ( both '' from :result ) into result from "SYS"."DUMMY";
  endmethod.
endclass.

class zcl_invalid_cha_rem_01 definition public final create public .

public section.

interfaces if_amdp_marker_hdb.

CLASS-methods INVALID_CHAR_REMOVAL importing VALUE (IV_IOBJ_NAME) TYPE RSCHABASNM VALUE (IV_text) TYPE string value (IV_lowr_case) type CHAR1 exporting VALUE (RESULT) TYPE string.

protected section.

private section.

endclass.

class zcl_invalid_cha_rem_01 implementation.

method INVALID_CHAR_REMOVAL by database procedure for hdb language sqlscript.

declare lv_flag varchar ( 1 );

declare lv_length integer := 0;

declare lv_counter integer := 1;

RESULT = :iv_text;

-- If the incoming string is not initial

IF :RESULT != '' then

-- To get the length of the string

SELECT LENGTH ( :result ) into lv_length FROM "SYS"."DUMMY";

-- Remove leading and trailing blank spaces

select trim ( both '' from :result ) into result from "SYS"."DUMMY";

-- if incoming string is greater than 1

IF ( SELECT LENGTH ( :result ) FROM "SYS"."DUMMY" ) > 1 THEN

-- check if the first char of the string is !, then take the string

-- from the first char != ! to end of the string

if SUBSTRING (:result, 1 , 1) = '!' then

while :lv_counter <= :lv_length do

if SUBSTRING (:result,:lv_counter,1) != '!' then

break;

end if ;

lv_counter = :lv_counter + 1;

end while ;

if :lv_counter = :lv_length then

result = '';

else

select SUBSTRING (:result,:lv_counter,:lv_length) into result from "SYS"."DUMMY";

end if;

end if ;

-- Eliminate the # sign completely

select REPLACE_REGEXPR (' ( [^ [:print:]|^ [\x {00C0}-\x {017F}]| [#])' IN :result with ' ' occurrence all) into result from "SYS"."DUMMY";

if :iv_iobj_name != '' then

-- Remove the select statement from AMDP

-- instead pass it when calling the AMDP Procedure

-- SELECT b.lowercase INTO lv_flag FROM rsdcha AS a INNER JOIN rsdchabas AS b

-- ON a.chabasnm = b.chabasnm

-- WHERE a.chanm = :IV_IOBJ AND

-- a.objvers = 'A' AND

-- b.objvers = 'A';

-- if lowercase flag is not set then convert to upper case

if :iv_lowr_case = '' then

select upper ( :RESULT ) into RESULT from "SYS"."DUMMY" ;

end if;

-- if incoming string is equal to 1

else

select replace ( :RESULT, '#', '') into result from "SYS"."DUMMY";

if SUBSTRING (:result, 1 , 1) = '!' then

RESULT = '';

end if ;

END IF;

-- Trim to remove leading and trailing blank spaces before returning the result

select trim ( both '' from :result ) into result from "SYS"."DUMMY";

endmethod.

endclass.

Create a transformation that uses the AMDP class and method for handling special characters. You can use the transformation editor in SAP BW/4HANA to create a transformation between your source and target objects. In the transformation, you need to call the AMDP method for each field that may contain special characters. You can use the example code below as a reference. The code passes the InfoObject name, the field value, and the lowercase flag as parameters to the AMDP method.

METHOD START_ROUTINE.
  DATA: lv_lowr_case TYPE char1.
  SELECT SINGLE lowercase INTO lv_lowr_case FROM rsdcha AS a INNER JOIN rsdchabas AS b
    ON a.chabasnm = b.chabasnm
    WHERE a.chanm = 'ZIOBJ' AND
    a.objvers = 'A' AND
    b.objvers = 'A'.
ENDMETHOD.

METHOD END_ROUTINE.
  DATA: lt_result TYPE STANDARD TABLE OF _ty_s_TG_1,
        ls_result LIKE LINE OF lt_result.
  LOOP AT RESULT_PACKAGE ASSIGNING FIELD-SYMBOL(<fs_result>).
    ls_result = <fs_result>.
    CALL METHOD zcl_invalid_cha_rem_01=>INVALID_CHAR_REMOVAL
      EXPORTING
        IV_IOBJ_NAME = 'ZIOBJ'
        IV_text      = <fs_result>-ZFIELD1
        IV_lowr_case = lv_lowr_case
      IMPORTING
        RESULT       = ls_result-ZFIELD1.
    CALL METHOD zcl_invalid_cha_rem_01=>INVALID_CHAR_REMOVAL
      EXPORTING
        IV_IOBJ_NAME = 'ZIOBJ'
        IV_text      = <fs_result>-ZFIELD2
        IV_lowr_case = lv_lowr_case
      IMPORTING
        RESULT       = ls_result-ZFIELD2.
    APPEND ls_result TO lt_result.
  ENDLOOP.
  CLEAR RESULT_PACKAGE.
  RESULT_PACKAGE[] = lt_result[].
ENDMETHOD.

METHOD START_ROUTINE.

DATA: lv_lowr_case TYPE char1.

SELECT SINGLE lowercase INTO lv_lowr_case FROM rsdcha AS a INNER JOIN rsdchabas AS b

ON a.chabasnm = b.chabasnm

WHERE a.chanm = 'ZIOBJ' AND

a.objvers = 'A' AND

b.objvers = 'A'.

ENDMETHOD.

METHOD END_ROUTINE.

DATA: lt_result TYPE STANDARD TABLE OF _ty_s_TG_1,

ls_result LIKE LINE OF lt_result.

LOOP AT RESULT_PACKAGE ASSIGNING FIELD-SYMBOL().

ls_result = .

CALL METHOD zcl_invalid_cha_rem_01=>INVALID_CHAR_REMOVAL

EXPORTING

IV_IOBJ_NAME = 'ZIOBJ'

IV_text = -ZFIELD1

IV_lowr_case = lv_lowr_case

IMPORTING

RESULT = ls_result-ZFIELD1.

CALL METHOD zcl_invalid_cha_rem_01=>INVALID_CHAR_REMOVAL

EXPORTING

IV_IOBJ_NAME = 'ZIOBJ'

IV_text = -ZFIELD2

IV_lowr_case = lv_lowr_case

IMPORTING

RESULT = ls_result-ZFIELD2.

APPEND ls_result TO lt_result.

ENDLOOP.

CLEAR RESULT_PACKAGE.

RESULT_PACKAGE[] = lt_result[].

ENDMETHOD.

Execute the data load and check the results. You can use the InfoPackage or the Data Transfer Process (DTP) to execute the data load from your flat file to your target object. You can also monitor the data load status and logs in SAP BW/4HANA. After the data load is completed, you can check the results in your target object and verify that the special characters are handled correctly.

Conclusion

In this blog post, we have shown you how to use AMDP code to handle special characters in SAP BW/4HANA data loads. AMDP code can help you improve the performance, flexibility, and readability of your data transformations. We hope you find this blog post useful and informative. If you have any questions or feedback, please feel free to leave a comment below.

Disclaimer: This content is generated by AI.