Something is amiss, I think.
The usual substitution character for any character set is something resembling "?". "?" in CCSID 870 is X'6F'. X'3F' is "?" in ASCII. Are you sure you're not double translating into ASCII? I found something that says the substitution character is determined from the character set. Character set for CCSID 870 is 959. I don't see anything that shows how to override the substitution character for the character set.
ftp://ftp.software.ibm.com/software/globalization/gcoc/attachments/CS00959.pdf
ftp://ftp.software.ibm.com/software/globalization/gcoc/attachments/CP00870.pdf
I haven't found a page for DB2 IBMi but this might be helpful: "Rules for String Conversions":
http://publib.boulder.ibm.com/infocenter/db2luw/v9r5/index.jsp?topic=%2Fcom.ibm.db2.luw.sql.ref.doc%2Fdoc%2Fr0008481.html
-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Åke Olsson
Sent: Thursday, April 18, 2013 9:12 AM
To: (midrange-l@xxxxxxxxxxxx)
Subject: Can you select replacement character in translation from unicode (UTF-16)
Problem is as follows:
A file with text field in Unicode (UTF-16) is read by a job running under CCSID 870 (Czech).
The text field has a character value that is valid in many language locales but not in Czech.
As a result these characters are replaced with X'3F' in the conversion process between Unicode and Ebcdic. 3F causes lots of problem when it is passed on later in the chain.
Question: Is there anywhere where you can configure the system to replace non-existing characters with some other hex code (preferably X'40' since that will not cause problems later in the chain)?
Med vänlig hälsning / Best regards
Åke H Olsson
[cid:image001.png@01CA1FE6.387A03A0]
Box 433 SE 551 16 Jönköping Sweden visit: Brunnsgatan 11
phone: +46 (0)36 342976 mobile: +46 (0)705 482976 fax: +46 (0)36 34 29 29
ake.olsson@xxxxxx<mailto:ake.olsson@xxxxxx> www.pdb.se
midrange.com
