Re: Case mapping of sharp s

[Eli Zaretskii]

> From: Kenichi Handa <address@hidden>
> Cc: address@hidden, address@hidden
> Date: Wed, 18 Nov 2009 10:00:57 +0900
> 
> In article <address@hidden>, Eli Zaretskii <address@hidden> writes:
> 
> > > From: Kenichi Handa <address@hidden>
> > > Cc: address@hidden, address@hidden
> > > Date: Tue, 17 Nov 2009 16:49:53 +0900
> > > 
> > > > Wouldn't the technique described in UTS 18
> > > > (http://www.unicode.org/reports/tr18/) help here?
> > > 
> > > Which technique do you mean?
> 
> > The one described under "3.10 Folded Matching".
> 
> But, that section just tells the way of providing pre-folded
> text in matching, which, I think, is not an acceptable way
> for searching a big buffer.

My idea was to use the same technique to fold the text as you search
through it.  You could, for example, fold one character at a time, so
that instead of comparing against X you compare against fold(X).

I admit that I didn't think about this more than a few seconds,
though, so perhaps this idea doesn't "hold water".