Re: [PATCH 3/3] dfa: optimize UTF-8 period

[Paolo Bonzini]

On 04/19/2010 10:07 AM, Jim Meyering wrote:
> Paolo Bonzini wrote:
> > On 04/17/2010 09:27 AM, Jim Meyering wrote:
> > > Paolo Bonzini wrote:
> > > > * NEWS: Document improvement.
> > > > * src/dfa.c (struct dfa): Add utf8_anychar_classes.
> > > > (add_utf8_anychar): New.
> > > > (atom): Simplify if/else nesting.  Call add_utf8_anychar for ANYCHAR
> > > > in UTF-8 locales.
> > > > (dfaoptimize): Abort on ANYCHAR.
> > > > ---
> > > >    NEWS      |    6 ++++++
> > > >    src/dfa.c |   46 +++++++++++++++++++++++++++++++++++++++++++---
> > > >    2 files changed, 49 insertions(+), 3 deletions(-)
> > >
> > > Only quick superficial feedback for now:
> >
> > I pushed all patches but this.
>
> Thanks!
>
> Would you please add comments describing this one in more detail?
> I ran out of time trying to understand how it works.

Something like this?

 /* For UTF-8 expand the period to a series of CSETs that define a valid
    UTF-8 character.  This avoids using the slow multibyte path.  I'm
    pretty sure it would be both profitable and correct to do it for
    any encoding; however, the optimization must be done manually as
    it is done above in add_utf8_anychar.  So, let's start with
    UTF-8: it is the most used, and the structure of the encoding
    makes the correctness more obvious.  */