On 4 Dec 2015 4:27 pm, "Eli Zaretskii" <eliz@gnu.org> wrote:
>
> > Date: Fri, 4 Dec 2015 16:06:40 +0000
> > From: Artur Malabarba <bruce.connor.am@gmail.com>
> > Cc: Alan Mackenzie <acm@muc.de>, 22090@debbugs.gnu.org
> >
> > > Thanks.  Is there any reasonably simple way of describing the
> > > resulting limitations (i.e. what will NOT match) on the user manual
> > > level?
> >
> > Basically, 'a' will match similar characters (like '𝑎' and 'á') and
> > their upper-case equivalents (like 'Á'). 'a' will NOT match characters
> > similar to 'A' that don't have a lower-case equivalent (like '𝔸') in
> > the unicode standard.
>
> What about ligatures, or symbols like ℻?

Won't match cross-case.

> Also, by "lower-case equivalent" do you mean a case mapping defined by
> the UCD

Yes. Visual appearance is irrelevant.

Strictly speaking, to match a general character, you need to search for its
decomposition. If this character also has a case "equivalent" (as per
current-case-table), you can also search for the decomposition of this
equivalent.