Doug -
We had a similar issue (exact same OS's on both sides). I ended up in a
crash course on character sets. Best place for an overall view of your
issue, and what IBM wants you to do about it is at :
http://www-3.ibm.com/software/ts/mqseries/support/faqs/convert.html
Depending on which direction you're going you may find problems with
open/closed parens, ampersand - &, and open/closed square brackets as well.
For several reasons (that I don't really want to get into here) we are
doing transaction on the sender channels. As a result, IBM chastised us
for not using translation on the GET, conceded the issue was still
legitimate, then suggested we change our OS/390 CCSID to 37.
That change fixed our issues so far.
Good Luck
Kelly Urban
First Data Corp
Douglas Pierson
<[EMAIL PROTECTED] To: [EMAIL PROTECTED]
OM> cc:
Sent by: MQSeries bcc:
List Subject: One bad character during
ASCII to EBCDIC conversion
<MQSERIES@AKH-Wie
n.AC.AT>
01/15/2003 11:49
AM
Please respond to
MQSeries List
Hi Everyone,
I have an MQI application that sends character data from Win2K (MQ 5.2,
CCSID 437) to OS/390 (MQ 2.1, CCSID 500). Conversion is requested of the
qmgr by the application on the MQGET.
The data is converted, except for one problem. An exclamation point, ASCII
x'21', is converted to a vertical bar, EBCDIC x'4F'. Shouldn't it instead
convert to EBCDIC x'5A'?
While Appendix H of MQ's Application Programmer's Reference identifies the
supported CCSID's, it doesn't show a character-by-character mapping. Is
such a mapping available somewhere? If the ASCI x'21' to EBCDIC x'4F' is
"correct", must I resort to a data conversion exit to solve my problem?
Thanks,
Doug Pierson
"{-??-???~??jv??x2??æj)b? b??.n?+??b?v?z??'^v)?â?Û ??êڕK
?n?ך???i?^j?m
'rبfj*.f)+-r*'jub^.+-ޭTyڊV歆i0[(~(
܆+ކijHpq1'j!
