Wrong Characters returned by the WebService

[Former Member]

Hi!

I have an scenario where i have one RFC Function module that call a WebService (Java/EJB) that brings the information from a Oracle DB. The RFC parse the results and populate my return tables.

When i call the WebService from wsnavigator in portal(http://url:port/wsnavigator), it returns the words with the special characters OK, eg: "Pend

ê

ncia".
But, when i call the WebService from the RFC Module in ABAP, using the interface if_http_client to make the WebService Call, it returns bogus characters, eg: "Pendência".
So, if my table in Oracle DB have the word "Pendência", the RFC call result will be "Pend

ê

ncia". The same happens with the following characters: áéíóúãàòìùäëïöüãõâêîôûçñ...

This situation only happens when i call the WebService from ABAP using the interface if_http_client, and it looks like the charset is wrong, but when i set the header of the HTTP REQUEST, i set the charset to iso-8859-1, as follows:

-

-

---

...

CALL METHOD http_client->request->set_header_field

EXPORTING

name = '~request_method'

value = 'POST'.

CALL METHOD http_client->request->set_header_field

EXPORTING

name = '~server_protocol'

value = 'HTTP/1.1'.

CALL METHOD http_client->request->set_header_field

EXPORTING

name = 'Content-Type'

value = 'text/xml; charset=iso-8859-1;'.

CALL METHOD http_client->request->set_header_field

EXPORTING

name = 'Content-Length'

value = txlen.

CALL METHOD http_client->request->set_header_field

EXPORTING

name = 'SOAPAction'

value = ''.

...

-

-

---

Any idea?

Thanks in advance.

[Former Member]

May be you don't have the right code page. I am not an expert, but check on that aspect with your basis administrators.

Here is the documentation related to code pages.

SAP character set identification

Definition

The 4-character name of an SAP character set as defined

in SAP character set maintenance.

The following explains the naming convention in more detail:

First digit: Code

0 EBCDIC character sets

1 ASCII character sets

2 mixed single byte / double byte character sets

4 double-byte character sets

6 mixed character sets

8 double byte and multibyte character sets

9 reserved for code pages you define

Second digit: Country

1-3 countries that use the Latin alphabet (Western Europe, North and South America, Australia, Africa)

4-6 countries that use non-Latin alphabets and writing

systems (Eastern Europe, Asia, Arabic countries in Africa)

7-9 reserved for special languages

Third and fourth digits: Sequential number

Example

0100 IBM 00697/00273 (Latin 1 - Germany/Austria)

0401 SNI BS2000 8859-5 EHCLC (cyrillic - multiple languages)

The table is TCP00.

Message was edited by: Srinivas Adavi

[Former Member]

I solved it by converting the code-page in the return of the webservice call...

* After receive the return from the WebService,
* move the result to the string wf_string1.
  clear wf_string1 .
  wf_string1 = http_client->response->get_cdata( ).
* Some variables
data: conv type ref to cl_abap_conv_in_ce,
        wf_stringx type xstring,
        viewoff type ref to cl_abap_view_offlen,
        len type i,
        content type string.

* Convert the return string of the webservice to Hexadecimal
  call function 'SCMS_STRING_TO_XSTRING'
    exporting
      text     = wf_string1
      mimetype = ''
    importing
      buffer   = wf_stringx.
* call the method to convert UTF-8(Code-page 4110) to ISO-8859-1(1100)
  conv = cl_abap_conv_in_ce=>create(
                        encoding = '4110'
                          endian = 'L'
                           input = wf_stringx ).

* Read the conversion result
  conv->read(
   exporting
     view = viewoff
   importing
     data = content
     len  = len ).
* Result string...
  wf_string1 = content.