Archives of the TeradataForum

Message Posted: Thu, 14 Apr 2016 @ 16:53:33 GMT

	<Prev	Next>		<<First	<Prev	Next>	Last>>

Dave, I only get the digest, so forgive me if I repeat something said by someone else. The correct way to solve this problem, it seems to me, is to map the Unicode characters to Latin or some other canonical-ish form (perhaps with a UDF), then compare. For instance, you would have a translation table that mapped ? ? ? ? to e, ? to n, and so on.

Of course, writing such a thing would be onerous. You could probably find open-source code that does it and could be modified for use within Teradata.

I hope this gives you some ideas.

	<Prev	Next>		<<First	<Prev	Next>	Last>>