Archives of the TeradataForum
Message Posted: Mon, 14 Dec 2009 @ 12:39:50 GMT
Subj: | | Re: UDFs and unicode symbols |
|
From: | | Frydryszak, Marek |
You ought to use character set unicode for [var]char variable(s) in create/replace function SQL statement and CHARACTER_UNICODE ( or
VARCHAR_UNICODE ) type in C code.
In other case 'abc' - even if default type for literals is unicode - will be converted to latin before passing to UDF and during the conversion
from Cyrillic to one-byte character you may get such error.
Certainly in C code you have to take into account you operate on 2-byte UTF16 characters ( size of buffers, don't use strcpy,strcmp etc. )
Regards
Marek
|