> definitions. There is already a built-in NOCASE collation that works for > US-Latin characters. You'll need to do something more advanced for
I am using US-Latin characters. > international characters sets, however. If you need i18n support, > you can still use the built-in NOCASE collation as an implementation > guide (it isn't hard) or to experiment with the COLLATE clause on table > definitions. I thought I saw mention of that somewhere, but I couldn't find any occurances of "NOCASE" in the syntax document. I was hoping for some database wide setting (using 3.x). Is there not one? I presume using the collate clause is something on the order of: create table participant ( partkey integer primary key not null, lastname varchar(40) not null collate nocase, firstname varchar(40) null collate nocase );