MQSeries.net :: View topic

atheek · Posted: Mon Oct 20, 2008 11:12 pm Post subject:

Hi All,

We are getting a message from a TPF system to a unix queue manager. We are facing issues in conversion of certain characters:

The byte sequence in unconverted format is:
156C D8D2 40E2 ( Message had a CCSID 500)

which when converted gets

0A25 514B 2053 ( Message has a CCSID 819) and in character form this is as: .%QK SY

15 is the ebcdic new line character which is getting to ascii line feed charcater 0A. 6C is ebcidic % charcter which is getting correctly transformed to % symbol. So the receiving application is interpreting that there is a % symbol in the message.

Now the TPF guys tell that 156C is a special combination and has its own meaning in the TPF world. 156C converts to a new line for them.
So I suspect the message is a mix of double byte and single byte characters which I think MQ is unable to convert correctly in its original form. The charcter conversion needs to be handled either at the receiver side or at sender side. Preference should be to use sender since the mainframe may be able to understand ebcdic better

I would like to know if you guys have faced any similar issue.

A few basic questions:
1. Does CCSID500 support both single byte and double byte characters?
2. What will be the best way to handle such a situation?

There are quite a few other combinations like this eg 5E5E, 005E etc

Thanks,
Atheek