Re: [bug #13859] matching '([A]|[B]){2}' in different locales

[Charles Levert]

* On Wednesday 2005-07-20 at 17:28:39 +0000, anonymous wrote:
> 
> Follow-up Comment #2, bug #13859 (project grep):
> 
> Thank you, that version works correctly regardless of locale for me too. The
> problem seems to be worked around by your dfa_optional patch. If I keep all
> other patches, but unapply that, it starts failing for me again. After reading
> http://www.mail-archive.com/address@hidden/msg00068.html

This referenced messages states:

> The full story with this patch is that the grep built-in DFA actually
> pessimises rather than optimises when processing UTF-8.  The idea of this
> patch is: if we are using UTF-8, disable grep's built-in DFA altogether.
> 
> (The DFA in glibc handles UTF-8 correctly, and fast.)

This begs the question:  what does grep's
built-in DFA actually provides (in terms of
performance, chiefly) once the regexp code has
been updated to glibc/gnulib's current version?
Does the answer depend on the chosen locale?
Is there still a justification for grep's DFA?