This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: C++: Tag transparent binding contour
- From: Mark Mitchell <mark at codesourcery dot com>
- To: Gabriel Dos Reis <gdr at integrable-solutions dot net>
- Cc: Jason Merrill <jason at redhat dot com>, gcc at gcc dot gnu dot org
- Date: 19 Sep 2003 00:25:52 -0700
- Subject: Re: C++: Tag transparent binding contour
- References: <m3brxfcftc.fsf@uniton.integrable-solutions.net><wvlr86brtcz.fsf@prospero.boston.redhat.com><m3llwj9jlk.fsf@uniton.integrable-solutions.net><wvladczrruo.fsf@prospero.boston.redhat.com> <m3pthxjzsz.fsf@uniton.integrable-solutions.net>
On Thu, 2003-09-18 at 19:07, Gabriel Dos Reis wrote:
> Jason Merrill <jason@redhat.com> writes:
>
> | On 03 Jun 2003 19:06:15 +0200, Gabriel Dos Reis <gdr@integrable-solutions.net> wrote:
> |
> | > Thank you. Is it OK to describe the binding contour created by
> | > maybe_push_cleanup_level as just a bare sk_block? Or do you think it
> | > is necessary to introduce a separate scope_kind enumerator? (I would
> | > say sk_block is sufficient),
> |
> | Well, it isn't a block. It's transparent to all names. If you're working
> | to distinguish different types of binding levels, this seems worth
> | distinguishing.
The right thing to do would be to get rid of these fake post-label
binding levels. The binding level structures should mirror the scopes
required by the language, and there is no new scope required after a
label.
--
Mark Mitchell
CodeSourcery, LLC
mark@codesourcery.com