the issue is not with japanese characters
it is that You are trying to process data outside the current record length
somewhere in Your logic ( application design ) there is a flaw,
You are expecting records with a 1xxx something length,
but You get also some record which are shorter, but most important they do not seem to contain the info You expect
Joined: 15 Feb 2005 Posts: 7129 Location: San Jose, CA
kaj_mei,
If you use DFSORT's VLSHRT option, the "short" records will be sorted as if the missing characters are binary zeros. If that's what you want to do, then you can add the following to your job:
Code:
//DFSPARM DD *
OPTION VLSHRT
/*
If you want to handle the "short" records in some other way, you need to explain how you want to handle them.
expat,
VLSHRT applies to SORT/MERGE, SUM, and INCLUDE/OMIT.
VLSCMP only applies to INCLUDE/OMIT, not SORT/MERGE, so it's not relevant in the situation described here.
