Re: [bug #36567] grep -i (case-insensitive) is broken with UTF8

[Paul Eggert]

Ah, OK, thanks, I see what you're saying now.  Clearly grep does not
behave the way you're asking for.  I don't think POSIX requires that
behavior either.  The relevant part of POSIX says that for -i

  not only the character, but also its case counterpart (if any),
  shall be matched

http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap09.html#tag_09_02

which indicates that the POSIX folks did not consider examples such as
Greek sigma, where a single letter can have multiple case
counterparts.

Even if POSIX doesn't specify the behavior, it might be nice for
GNU grep to do it anyway, if someone could take the time to
implement it without undue performance loss.  You're welcome to
file a bug report on the Savannah bug tracker about this.
<http://savannah.gnu.org/bugs/?group=grep>