On Sat, Dec 17, 2005 at 05:01:15PM -0500, Tom Lane wrote:
> Martijn van Oosterhout <kleptog@svana.org> writes:
> > I think the real solution is to implement COLLATE support.
>
> Maybe so, but we still need to figure out what we're doing for the back
> branches, and that won't be it ...
To be honest, there are really only a handful of locales that suffer
from this issue, so perhaps we should document it and move on. Perhaps
we should add some sanity checks to initdb to ensure that the locale is
sane. Basically things like:
FOO != foo (case-sensetivity)
tty != tyty (wierd hungarian case)
aei != äëï (accent-sensetivity)
Until COLLATE support is working, I'm not sure if we should put that
much effort into dealing with these cases. Even after we have COLLATE
support we'll probably have to start using strxfrm to make these
locales usable for indexes.
Using a binary string compares for tie-breaks seems a waste of cycles
for the 99% of locales that don't need it.
Have a nice day,
--
Martijn van Oosterhout <kleptog@svana.org> http://svana.org/kleptog/
> Patent. n. Genius is 5% inspiration and 95% perspiration. A patent is a
> tool for doing 5% of the work and then sitting around waiting for someone
> else to do the other 95% so you can sue them.