On Wed, Nov 07, 2007 at 11:21:30PM +0000, Ren=C3=A9 Scharfe wrote:
quoted text
> Pierre Habouzit schrieb:
> > {
> >     const char *percent =3D strchrnul(fmt, '%');
> >     while (*percent) {
> >         strbuf_add(sb, fmt, percent - fmt);
> >         fmt =3D percent + 1;
> >=20
> >         /* do your stuff */
> >=20
> >         percent =3D strchrnul(fmt, '%');
> >     }
> >     strbuf_add(sb, fmt, percent - fmt);
> > }
> >=20
> >=20
> > Which would require strchrnul, but it's trivial compat/ material for su=
re.
quoted text
>=20
> Grepping through the source I see several places that can be simplified
> by converting them to strchrnul(), so I think introducing this GNU
> extension is a good idea in any case.

  Yes, strchrnul is _very_ useful for parsing.

quoted text
> Using strchr()/strchrnul() instead of strbuf_addch()'ing is sensible, of
> course.  I don't like the duplicate code in your sketch above, though.
> I'll try to look into it later today.

Well you can alternatively write it this way of course:

for (;;) {
    const char *percent =3D strchrnul(fmt, '%');
    strbuf_add(sb, fmt, percent - fmt);
    if (!*percent)
        break /* or return probably */;
    fmt =3D percent + 1;

    /* do your stuff */
}

--=20
=C2=B7O=C2=B7  Pierre Habouzit
=C2=B7=C2=B7O                                                madcoder@debia=
n.org
OOO                                                http://www.madism.org