Re: Hostnames, IDNs, Punycode and Unicode Case Folding

On Tue, Dec 30, 2014 at 12:18:58AM +0000, Mike Cardwell wrote:
>
> This is exactly the same method that we commonly use for performing case
> insensitive text searches using lower() indexes.

Hmm.  How did you get the original, then?  If you have the original
Unicode version, why don't you switch to IDNA2008 publication rules,
which are way more reliable?  In that case, you do have a 1:1 lookup
and you shouldn't have a problem.

If you need variants, then you have a different problem, but that
actually can be specified for the much narrower range of UTF-8
permissible under IDNA2008.

A

--
Andrew Sullivan
ajs@crankycanuck.ca