Corrupted chars/wrong charset?

Hello!

We send a TextMessage from Weblogic 8.1 (over a Messaging bridge in WL) to a Queue at MQ. The messages at the queue is later consumed from a app at WebSphere.

The text is in fact a XML-structure. Unfortunately it may contain swidish signs like "åäö". These chars is corrupted when the message is processed at MQ.

Some where along the way the chars get corrupted. Unfortunately I can not say exactly where the corruption takes place, but I guess it is at MQ.

So now is my question if any one got some hints how to solve this.

Hopefully this is some clues.
The text inside the message is a xml-string. The xml got this encoding:
<?xml version="1.0" encoding="UTF-8"?>
or we have also tried with
<?xml version="1.0" encoding="ISO-8859-1"?>

Acctually we do not think this matters since this is "folded" inside a JMS TextMessage.

We have there for also set this (both utf-8 and iso-8859-1) on the protocol iiop at the server inside WebLogic.

When the message is send we write the xml to file and everything looks fine:

SalesItemLegacyId="ÅÄÖåäöÅÄÖ"

Hopefully you are able to se the chars, in case I have html-encode them like:
&Aring;&Auml;&Ouml;


How ever when the message reach MQ it looks like and we browse the queue the messsage looks like:

001D0  42 75 6E 64 6C 69 6E 67--3D 22 22 20 53 61 6C 65  |Bundling....Sale|
001E0  73 49 74 65 6D 4C 65 67--61 63 79 49 64 3D 22 C3  |sItemLegacyId..+|
001F0  85 C3 84 C3 96 C3 A5 C3--A4 C3 B6 C3 85 C3 84 C3  |.+.+.+.+.+.+.+.+|
00200  96 22 20 53 61 6C 65 73--49 74 65 6D 4E 61 6D 65  |...

And if we browse it with a client tool it looks like:

SalesItemLegacyId="ÅÄÖåäö"

Inside MQ it looks like the "code set"/"page set" is set to 437 or IBM-437 (I hope I put that right since I am at home now).

Our gues is that IBM-437 and UTF-8/ISO-8859-1 give us a missmatch.

So if any one know how to solve this please let us know!

Best regards
Fredrik

Best regards
Fredrik