| Author |
Message
|
| CHERIANV |
 Posted: Mon Feb 13, 2006 9:41 am Post subject: Japanese XML chars. via MQ |
 |
|
Apprentice
Joined: 15 Feb 2002
Posts: 43
Location: New York
|
Hello,
We're receiving a UTF-8 XML message with Japanese chars. via MQ but the MQMD codepage is 819 within the message . We parse this message through a Message Flow hosted by a Z/OS WMQI 2.1 Broker (QM codepage 500). The message flow sets to output message MQMD code page = 930 to get the Japanese field in the output message . When we test this, we're getting some "?? " invalid chars. in associated fields that should have valid Japanese chars. in the output message.
Is there a way for the receiving application (ie. the message flow) to manipulate this message (MQMD = 819) it so we get the right Japanese chars. in the output message.
Should we ask the source application to explicitly set the MQMD codepage to 1208 ? If so, will that resolve the problem ?
Any input to resolve the problem is appreciated. |
|
| Back to top |
|
 |
| jefflowrey |
| Posted: Mon Feb 13, 2006 10:03 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
At all times, the CCSID of the message should accurately represent the contents of the message. So if the message contains data that is represented in 1208, then the CCSID should be 1208.
If you are getting bad characters when you output to a codepage of 930, that does not in any way imply that you have received bad characters on the input message.
It is entirely possible that converting the message TO codepage 930 has caused the problem.
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| CHERIANV |
| Posted: Mon Feb 13, 2006 10:57 am Post subject: |
|
|
Apprentice
Joined: 15 Feb 2002
Posts: 43
Location: New York
|
Thanks Jeff. It appears that the message we are receiving is defaulted to the originating Queue Manager's copepage ie. 819. Is it possible to CAST the incoming message that we currently receive with CCSID 1208 and set the MQMD CCSID to 1208 as well within the message flow and reparse it as an XML before output to a codpage of 930.
Was wondering if there were any alternatives at all for handling the conversion at our end or is it not possible ? |
|
| Back to top |
|
|
| jefflowrey |
| Posted: Mon Feb 13, 2006 11:02 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
Broker automatically converts data into Unicode (and now someone from Hursley will likely correct this generalization) when it parses it into a logical message tree.
You need to determine if this is the cause of the problem, or if the cause of th e problem is casting back to codepage 930 on the output side.
Until you do that, you don't know what the problem is and you can't solve it.
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| jefflowrey |
| Posted: Tue Feb 14, 2006 5:45 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
| mquseless wrote: |
| The solution is to define a custom msg format, and a custom msg converter. It is not too difficult; WMQ supplies an app (crtmqcvx on UNIX) which generates most of the code required. See the APG. |
That's certainly *a* solution.
Ensuring that the CCSID properly identifies the message is another.
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| jefflowrey |
| Posted: Tue Feb 14, 2006 6:51 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
I was assuming the "Japanese" characters were in UTF-8.
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
|
|
