ASG
IBM
Zystems
Cressida
Icon
Netflexity
 
  MQSeries.net
Search  Search       Tech Exchange      Education      Certifications      Library      Info Center      SupportPacs      LinkedIn  Search  Search                                                                   FAQ  FAQ   Usergroups  Usergroups
 
Register  ::  Log in Log in to check your private messages
 
RSS Feed - WebSphere MQ Support RSS Feed - Message Broker Support

MQSeries.net Forum Index » WebSphere Message Broker (ACE) Support » Problems with extended ascii characters in unicode

Post new topic  Reply to topic
 Problems with extended ascii characters in unicode « View previous topic :: View next topic » 
Author Message
m00300
PostPosted: Fri Jan 31, 2003 8:10 am    Post subject: Problems with extended ascii characters in unicode Reply with quote

Apprentice

Joined: 01 May 2002
Posts: 31
Can anyone help with the following problem :

We have a Java GUI front end sending CWF messages to our MQSI (v2.0.2) processing hub. These messages are being sent in CCSID 1208.

All was going well until they started testing unusual characters including :
£éíóáú

This caused the following error :
Unconvertable character BIPMsg[1] - c3 BIPMsg[2] - 2122c2a324255e26282a29407e233f2f7b7d5b5d2d5f3d2b5c7c3ce282acc3a9c3adc3 BIPMsg[3] - 1208

After some research (mainly on this site) we tried CCSID 1200 instead which changed the error message to :

CWF Invalid Extended Decimal BIPMsg[1] - -30 BIPMsg[2] - GetMessageFromLegacyBitstream BIPMsg[3] - DN0UURS0A0001 BIPMsg[4] - CLAIMS_NOTN_INPUT_MSG BIPMsg[5] - CR1_INCIDENT_TABLE BIPMsg[6] - CR1_I_TP_VEHICLE_CNT

This suggested to us that some character parts of the message where longer then they should have been (due to incorrectly converted mult-byte characters?) and were over flowing into a numeric field giving the error above when parsed through the MRM.

Yet more research here suggested that we should be using CCSID 1252.
Got the Java front end to enter this as the CCSID in the MQMD when writing to MQ and hey presto everthing works fine - character converted perfectly and message goes onto Unix and mainframe platforms post MQSI fine.
BUT.....our Java team are now saying they dont want to implement this in
production as its a bodge becuase JAVA is really working in CCSID 1200 and we are lying to MQ by telling it we are using 1252. The front end is currently not doing a conversion between 1200 and 1252 and I think i'm right in saying that by default MQ doesnt do a convert on an MQPUT.

So my question is - are they right ?
Is it a bodge ?
Is there a better way to do it and should MQSI be able to handle these characters in CCSID 1200.

(we have no conversion on channels switched on in MQ and are messages do not have RFH2 headers on. We tried a couple of experiments using RFH2 headers with confusing results : adding an RFH2 header to the CWF message made no difference, but if we send XML with an RFH2 header the characters seem to be converted correctly!?)

I'd be grateful for any advice/explanation of why these problems are occurring.

Thanks.
Kevin.
Back to top
View user's profile Send private message Send e-mail
m00300
PostPosted: Tue Feb 04, 2003 3:00 am    Post subject: ...after more testing (unicode problems) Reply with quote

Apprentice

Joined: 01 May 2002
Posts: 31
I've been experimenting with the convert option on the Input queue node and i've think i've solved my problem :

- sending a test message in with CCSID 1208 and Encoding 273 with convert option unchecked fails (due to errors mentioned in previous post)

- sending a test message in with CCSID 1208 and Encoding 273 with convert option checked and Encoding set to 273 works and gives an
outgoing message of CCSID 850 and Encoding 273

- sending a test message in with CCSID 1208 and Encoding 273 with convert option checked but no CCSID or Encoding specified also works and gives an outgoing message of CCSID 850 and Encoding 0

Not to sure about an Encoding of 0 - I havent checked yet that these messages can be processed by the receiving platforms correctly but the messages look well formed.

So I think this come down to the fact that none of the other dozen or so systems/platforms we receive messages from has ever sent us unicode and therefore we've never had to worry about using the convert option on the input node before.

Looks like defaulting to checking convert on the input node may be the way to go.
Back to top
View user's profile Send private message Send e-mail
Display posts from previous:   
Post new topic  Reply to topic Page 1 of 1

MQSeries.net Forum Index » WebSphere Message Broker (ACE) Support » Problems with extended ascii characters in unicode
Jump to:  



You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
Protected by Anti-Spam ACP
  
 


Theme by Dustin Baccetti
Powered by phpBB © 2001, 2002 phpBB Group

Copyright © MQSeries.net. All rights reserved.