Re: libstdc++/7186: DR179 for std::deque::iterator and const_iterator

Re: libstdc++/7186: DR179 for std::deque::iterator and const_iterator

[B. Kosnik]

The following reply was made to PR libstdc++/7186; it has been noted by GNATS.

From: "B. Kosnik" <bkoz@nabi.net>
To: Paolo Carlini <pcarlini@unitus.it>
Cc: bkoz@gcc.gnu.org, gcc-bugs@gcc.gnu.org, gcc-prs@gcc.gnu.org,
   nobody@gcc.gnu.org, paolo@gcc.gnu.org, pcarlini@unitus.it,
   gcc-gnats@gcc.gnu.org
Subject: Re: libstdc++/7186: DR179 for std::deque::iterator and const_iterator
Date: Wed, 3 Jul 2002 10:11:17 -0700

 > >    Is there a survey of existing practice for this issue?
 > >
 > Agreed, I'm going to take care of this, surveying the existing
 > practice of 3 or 4 implementations I have access too.
 
 Great.
 
 > > Can all but the reverse_iterator bits go in now, since that will
 > > have ABI-impacting effects?
 > >
 > Well, as I mentioned in the PR, decoupling the reverse_iterator case 
 > from the "normal" iterator case seems difficult for the special 
 > std::deque::iterator and const_iterator case (it isn't for 
 > __normal_iterator, used by the other standard containers). Again, I
 > will do my best in the next weeks.
 
 Hmm. 
 
 Thanks for looking into this.
 
 -benjamin

Re: libstdc++/7186: DR179 for std::deque::iterator and const_iterator

[paolo]

Synopsis: DR179 for std::deque::iterator and const_iterator

State-Changed-From-To: feedback->closed
State-Changed-By: paolo
State-Changed-When: Wed Jul 24 23:48:32 2002
State-Changed-Why:
    Yes, Benjamin, it is fixed for 3.3.

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=7186

Re: libstdc++/7186: DR179 for std::deque::iterator and const_iterator

[bkoz]

Synopsis: DR179 for std::deque::iterator and const_iterator

State-Changed-From-To: analyzed->feedback
State-Changed-By: bkoz
State-Changed-When: Wed Jul 24 23:25:44 2002
State-Changed-Why:
    I think you've fixed this, right Paolo?
    
    -benjamin

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=7186

Re: libstdc++/7186: DR179 for std::deque::iterator and const_iterator

[Benjamin Kosnik]

The following reply was made to PR libstdc++/7186; it has been noted by GNATS.

From: Benjamin Kosnik <bkoz@redhat.com>
To: pcarlini@unitus.it
Cc: gcc-gnats@gcc.gnu.org, gcc-prs@gcc.gnu.org, gcc-bugs@gcc.gnu.org
Subject: Re: libstdc++/7186: DR179 for std::deque::iterator and const_iterator
Date: Thu, 4 Jul 2002 09:49:35 -0700

 > - Metrowerks8 is able to compile both (this is the behaviour mandated
 > by the resolution of DR179, according to my reading)
 
 I believe this is what v3 should shoot for as well.
 
 -benjamin

Re: libstdc++/7186: DR179 for std::deque::iterator and const_iterator

[Paolo Carlini]

The following reply was made to PR libstdc++/7186; it has been noted by GNATS.

From: Paolo Carlini <pcarlini@unitus.it>
To: gcc-gnats@gcc.gnu.org,  gcc-prs@gcc.gnu.org,  pcarlini@unitus.it, 
 gcc-bugs@gcc.gnu.org, Benjamin Kosnik <bkoz@redhat.com>
Cc:  
Subject: Re: libstdc++/7186: DR179 for std::deque::iterator and const_iterator
Date: Wed, 03 Jul 2002 23:28:52 +0200

 A little update on the existing practice (as requested by Benjamin):
 - Interestingly, Intel6 (Dinkumware) and Borland5.5.1 (Rogue Wave) show 
 a behaviour identical to that of current mainline v3 that is 6642's 
 testcase is compiled just fine, while 7186's testcase is not.
 - Metrowerks8 is able to compile both (this is the behaviour mandated by 
 the resolution of DR179, according to my reading)
 
 Paolo.
 
 http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=7186 
 
 

Re: libstdc++/7186: DR179 for std::deque::iterator and const_iterator

[Paolo Carlini]

The following reply was made to PR libstdc++/7186; it has been noted by GNATS.

From: Paolo Carlini <pcarlini@unitus.it>
To: bkoz@gcc.gnu.org,  gcc-bugs@gcc.gnu.org,  gcc-prs@gcc.gnu.org, 
 nobody@gcc.gnu.org,  paolo@gcc.gnu.org,  pcarlini@unitus.it, 
 gcc-gnats@gcc.gnu.org
Cc:  
Subject: Re: libstdc++/7186: DR179 for std::deque::iterator and const_iterator
Date: Wed, 03 Jul 2002 18:28:58 +0200

 bkoz@gcc.gnu.org wrote:
 
 >    Is there a survey of existing practice for this issue?
 >
 Agreed, I'm going to take care of this, surveying the existing practice 
 of 3 or 4 implementations I have access too.
 
 > Can all but the reverse_iterator bits go in now, since that will have ABI-impacting effects?
 >
 Well, as I mentioned in the PR, decoupling the reverse_iterator case 
 from the "normal" iterator case seems difficult for the special 
 std::deque::iterator and const_iterator case (it isn't for 
 __normal_iterator, used by the other standard containers). Again, I will 
 do my best in the next weeks.
 
 Paolo.
 
 http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=7186
 
 

Re: libstdc++/7186: DR179 for std::deque::iterator and const_iterator

[bkoz]

Synopsis: DR179 for std::deque::iterator and const_iterator

Responsible-Changed-From-To: unassigned->paolo
Responsible-Changed-By: bkoz
Responsible-Changed-When: Wed Jul  3 09:22:56 2002
Responsible-Changed-Why:
    I'm going to give this to you...
State-Changed-From-To: open->analyzed
State-Changed-By: bkoz
State-Changed-When: Wed Jul  3 09:22:56 2002
State-Changed-Why:
    Is there a survey of existing practice for this issue? Can all but the reverse_iterator bits go in now, since that will have ABI-impacting effects?
    
    -benjamin

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=7186

libstdc++/7186: DR179 for std::deque::iterator and const_iterator

[pcarlini]

>Number:         7186
>Category:       libstdc++
>Synopsis:       DR179 for std::deque::iterator and const_iterator
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Tue Jul 02 12:16:09 PDT 2002
>Closed-Date:
>Last-Modified:
>Originator:     Paolo Carlini
>Release:        3.1 and current 3.1.1 and 3.2
>Organization:
>Environment:
don't care
>Description:
This is a follow up of libstdc++/6642, now fixed in the
mainline, concerning std::deque: the resolution of DR179
<Ready> needs to be implemented for operator- of
std::deque::iterator and std::deque::const_iterator too.
This is a testcase, which currently does not compile at all
whereas it should:
//////////
#include <deque>

int main()
{
  std::deque<int> d;
  std::deque<int>::iterator it = d.begin();
  std::deque<int>::const_iterator cit = d.begin();

  return it - cit;
}
//////////
>How-To-Repeat:

>Fix:
The tricky point is not affecting the behaviour of reverse
iterators about which the status of the relevant DR280
is still <open>.
>Release-Note:
>Audit-Trail:
>Unformatted: