axis-c-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Stadelmann Josef" <>
Subject Axis2/C ISO-8859-1 to UTF-8 conversion during (de)serialization
Date Thu, 13 Nov 2008 07:40:36 GMT
Hi all,

maybe someone more known to the subject can tell me that Axis2/C release 1.5.0 and base libraries
support the following two conversions

a) dumping/deserializing an Axiom OM element consisting of characters belonging to the ISO-8859-1
chara set into an UTF-8 / UTF-16 XML string/stream

b) parsing/deserializing an UTF-8 / UTF-16 XML string/stream into an Axiom OM Element storing
spezial characters as members of the ISO-8859-1 char set

c) deliver a short example for deserializing and serializing this in Axis2/C

I have seen that Axis2/C 1.0 has an issue logged in that it does not support UTF-8 and other
char sets. 
And in fact this is the case with Axis2/C Version 1.0, the version I currently have in use.
I ported all of it to OpenVMS. 
I like to know that Axis2/C 1.5 and related base libs support UTF-8 before I go one with an
upgrade and PORT of 
Axis2/C Version Release 1.5 and relatde base libs from Windows to OpenVMS again. 

My Axis2/Java web service serializes an Axiom OM Element into a SOPA XMS string and passes
this via JVM and WSIT lib from HP to a legacy wrapper written in Axis2/C on OopenVMS. Then
when all legacy work is done, the Axis2/C Axiom element is serialized into a XML stream which
is returned to the calling Axis2/Java web service routine. Now when it arrives at this routine,
the XML string has to be deserialized into an Axiom OM element and it fails because Axis2/Java
supports UTF-8 while Axis2/C supports ISO-8859-1 only.

Now; because all remaining code down to the Windows Vista Webservice WCF Client support UTF-8,
I would like to use UTF-8 and make it the ultimative future character set and have one conversion
only in the wrapper which wrappes our legacy server in OpenVMS Pascal, which supports only

Today: I serialize an Axis2/C Axiom OM Element consiting of spezial characters such as the
German Umlauts "ä ö ü" to an XML string which then arrives at an Axis2/J web service, where
it has to be deserialized again into an Axiom OM element, and hence fails because the Axis2/J
part supports UTF-8.

So I like to know if Axis2/C and base libs support UTF-8 with Release 1.5 and how this conversions
have to be invoked.

many thanks


View raw message