c++/359

[Jean-Francois Panisset]

The following reply was made to PR c++/359; it has been noted by GNATS.

From: Jean-Francois Panisset <panisset@discreet.com>
To: loewis@gcc.gnu.org
Cc: gcc-gnats@gcc.gnu.org, nobody@gcc.gnu.org, panisset@discreet.com
Subject: Re: c++/359 
Date: Thu, 29 Jun 2000 04:16:12 -0400

  <20000629075030.21549.qmail@sourceware.cygnus.com>loewis@gcc.gnu.org writes
 >Synopsis: g++ incorrectly grants access to private members to nested classes
 >
 >State-Changed-From-To: open->suspended
 >State-Changed-By: loewis
 >State-Changed-When: Thu Jun 29 00:50:29 2000
 >State-Changed-Why:
 >    According to the proposed resolution of Core Issue 45,
 >    http://www.informatik.hu-berlin.de/~loewis/cwg_active.html#45
 >    your example is well-formed, so it is not a bug in the compiler.
 >    If this resolution won't be listed in the next technical corrigendum,
 >    this report would need to be reviewed.
 >
 > http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view&pr=359&database=gcc
 
 Thank you very much for the answer, sorry to have submitted a non-bug.
 Talk about a moving target though...
 
 I initially got confused since gcc 2.95 refused to compile the following code:
 
 class foo {
 private:
   void foo_method();
 
   class foo_nested {
     void foo_nested_method(foo &);
   };
 };
 
 void
 foo::foo_nested::foo_nested_method(foo &f)
 {
   f.foo_method();
 }
 
 which would be legit according to the proposed new rules. 2.96 20000625
 now accepts this code as valid.
 
 Is there a timeframe as to when this might make it into the C++
 standard? i.e. at what point should I report this as a bug against
 the SUN C++ compiler?
 
 Thanks,
 JF
 
 
 Jean-Francois Panisset                                panisset@discreet.com
 Software Engineer
 Discreet Logic