This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: PATCH (libstdc++-v3): Fix libstdc++/7680 (and a reference fix for the dynamic visibility of C99 features in system headers)
- From: Gabriel Dos Reis <gdr at integrable-solutions dot net>
- To: Zack Weinberg <zack at codesourcery dot com>
- Cc: Loren James Rittle <rittle at latour dot rsch dot comm dot mot dot com>, phil at jaj dot com, CLR004 at motorola dot com, gcc-patches at gcc dot gnu dot org, libstdc++ at gcc dot gnu dot org
- Date: 16 Apr 2003 03:24:28 +0200
- Subject: Re: PATCH (libstdc++-v3): Fix libstdc++/7680 (and a reference fix for the dynamic visibility of C99 features in system headers)
- Organization: Integrable Solutions
- References: <200304160031.h3G0VpBt043952@latour.rsch.comm.mot.com><87of379qzj.fsf@egil.codesourcery.com>
Zack Weinberg <zack at codesourcery dot com> writes:
| Loren James Rittle <rittle at latour dot rsch dot comm dot mot dot com> writes:
|
| >>I'm wondering how much typing you would save with CPP...
| >
| > I wondered the same thing. How do I put a #if region inside a macro
| > defined with #define?
| >
| > I need to define something like this to tighten it:
| >
| > #define _GLIBCXX_C99_BIND(a,b) \
| > #if (a) \
| > using ::b; \
| > #endif \
|
| This, I'm afraid, is completely impossible.
|
| If you want to talk about extensions to allow it, I would suggest that
| you instead allow
|
| if (a) using ::b;
This, unfortunately won't solve the original concern since in that
specific case the using-declaration ought to appear at a
namespace-scope -- where one can't use if().
| where (a) is an integer-constant-expression. This is a limited form
| of the 'if-goes-everywhere' functionality I've been thinking about off
| and on - the idea is to supersede #if completely (except for excluding
| other preprocessing directives), by allowing if (x) { ... } anywhere,
I believe you're going far beyond where most of us, zealous C++
programmers wanting to get rid of CPP, didn't even go -- we just
dreamt it ;-)
At any rate, it is interesting to note that it is somehow part of
Daveed Vandevoorde meta-programming extension to C++ (except that you
need to prefix it with "metacode") he talked about both at the ACCU
conference and the ISO C++ committee meeting.
-- Gaby