On Jun 21, 2006, at 18:19, Josh Berkus wrote:
> Well, current case-insensitivity hacks definitely aren't compatible
> with
> LIKE as far as "begins with" indexes are concerned.
Yes, currently I use LOWER() for my indexes and for all LIKE, =, etc.
queries. This works well, but ORDER by of course isn't what I'd like.
That's one of the things that Elein's email domain addresses, albeit
with a USING keyword, which is unfortunate.
> Of course, floating
> LIKEs (%value%) are going to suck no matter what data type you're
> using.
Yes, I know that. :-) I avoid that.
> I created an operator for CI equality ... =~ ... which performs
> well on
> indexed columns. But it doesn't do "begins with".
Oops. So how could it perform well on indexed columns?
> ITEXT is a TODO, but there are reasons why it's harder than it looks.
I'm sure. I should bug potential future SoC students about it. ;-)
Best,
David
