public inbox for libstdc++@gcc.gnu.org
 help / color / mirror / Atom feed
From: Petr Ovtchenkov <ptr@void-ptr.info>
To: Jonathan Wakely <jwakely@redhat.com>
Cc: libstdc++@gcc.gnu.org, gcc-patches@gcc.gnu.org
Subject: Re: [PATCH] libstdc++: istreambuf_iterator keep attached streambuf
Date: Thu, 28 Sep 2017 12:06:00 -0000	[thread overview]
Message-ID: <20170928150643.2f667ec9@void-ptr.info> (raw)
In-Reply-To: <20170928103425.GG4582@redhat.com>

On Thu, 28 Sep 2017 11:34:25 +0100
Jonathan Wakely <jwakely@redhat.com> wrote:

> On 23/09/17 09:54 +0300, Petr Ovtchenkov wrote:
> >istreambuf_iterator should not forget about attached
> >streambuf when it reach EOF.
> >
> >Checks in debug mode has no infuence more on character
> >extraction in istreambuf_iterator increment operators.
> >In this aspect behaviour in debug and non-debug mode
> >is similar now.
> >
> >Test for detached srteambuf in istreambuf_iterator:
> >When istreambuf_iterator reach EOF of istream, it should not
> >forget about attached streambuf.
> From fact "EOF in stream reached" not follow that
> >stream reach end of life and input operation impossible
> >more.
> >---
> > libstdc++-v3/include/bits/streambuf_iterator.h     | 41 +++++++--------
> > .../24_iterators/istreambuf_iterator/3.cc          | 61 ++++++++++++++++++++++
> > 2 files changed, 80 insertions(+), 22 deletions(-)
> > create mode 100644 libstdc++-v3/testsuite/24_iterators/istreambuf_iterator/3.cc
> >
> >diff --git a/libstdc++-v3/include/bits/streambuf_iterator.h
> >b/libstdc++-v3/include/bits/streambuf_iterator.h index f0451b1..45c3d89 100644
> >--- a/libstdc++-v3/include/bits/streambuf_iterator.h
> >+++ b/libstdc++-v3/include/bits/streambuf_iterator.h
> >@@ -136,12 +136,20 @@ _GLIBCXX_BEGIN_NAMESPACE_VERSION
> >       istreambuf_iterator&
> >       operator++()
> >       {
> >-	__glibcxx_requires_cond(!_M_at_eof(),
> >+	__glibcxx_requires_cond(_M_sbuf,
> > 				_M_message(__gnu_debug::__msg_inc_istreambuf)
> > 				._M_iterator(*this));
> > 	if (_M_sbuf)
> > 	  {
> >+#ifdef _GLIBCXX_DEBUG_PEDANTIC
> >+	    int_type _tmp =
> >+#endif
> > 	    _M_sbuf->sbumpc();
> >+#ifdef _GLIBCXX_DEBUG_PEDANTIC
> >+	    __glibcxx_requires_cond(!traits_type::eq_int_type(_tmp,traits_type::eof()),
> >+				    _M_message(__gnu_debug::__msg_inc_istreambuf)
> >+				    ._M_iterator(*this));
> >+#endif
> > 	    _M_c = traits_type::eof();
> > 	  }
> > 	return *this;
> >@@ -151,14 +159,16 @@ _GLIBCXX_BEGIN_NAMESPACE_VERSION
> >       istreambuf_iterator
> >       operator++(int)
> >       {
> >-	__glibcxx_requires_cond(!_M_at_eof(),
> >+        _M_get();
> >+	__glibcxx_requires_cond(_M_sbuf
> >+				&& !traits_type::eq_int_type(_M_c,traits_type::eof()),
> > 				_M_message(__gnu_debug::__msg_inc_istreambuf)
> > 				._M_iterator(*this));
> >
> > 	istreambuf_iterator __old = *this;
> > 	if (_M_sbuf)
> > 	  {
> >-	    __old._M_c = _M_sbuf->sbumpc();
> >+	    _M_sbuf->sbumpc();
> > 	    _M_c = traits_type::eof();
> > 	  }
> > 	return __old;
> >@@ -177,18 +187,9 @@ _GLIBCXX_BEGIN_NAMESPACE_VERSION
> >       _M_get() const
> >       {
> > 	const int_type __eof = traits_type::eof();
> >-	int_type __ret = __eof;
> >-	if (_M_sbuf)
> >-	  {
> >-	    if (!traits_type::eq_int_type(_M_c, __eof))
> >-	      __ret = _M_c;
> >-	    else if (!traits_type::eq_int_type((__ret = _M_sbuf->sgetc()),
> >-					       __eof))
> >-	      _M_c = __ret;
> >-	    else
> >-	      _M_sbuf = 0;
> >-	  }
> >-	return __ret;
> >+	if (_M_sbuf && traits_type::eq_int_type(_M_c, __eof))
> >+          _M_c = _M_sbuf->sgetc();
> >+	return _M_c;
> >       }
> >
> >       bool
> >@@ -339,7 +340,7 @@ _GLIBCXX_BEGIN_NAMESPACE_VERSION
> >       typedef typename __is_iterator_type::streambuf_type  streambuf_type;
> >       typedef typename traits_type::int_type               int_type;
> >
> >-      if (__first._M_sbuf && !__last._M_sbuf)
> >+      if (__first._M_sbuf && (__last == istreambuf_iterator<_CharT>()))
> > 	{
> > 	  streambuf_type* __sb = __first._M_sbuf;
> > 	  int_type __c = __sb->sgetc();
> >@@ -374,7 +375,7 @@ _GLIBCXX_BEGIN_NAMESPACE_VERSION
> >       typedef typename __is_iterator_type::streambuf_type  streambuf_type;
> >       typedef typename traits_type::int_type               int_type;
> >
> >-      if (__first._M_sbuf && !__last._M_sbuf)
> >+      if (__first._M_sbuf && (__last == istreambuf_iterator<_CharT>()))
> > 	{
> > 	  const int_type __ival = traits_type::to_int_type(__val);
> > 	  streambuf_type* __sb = __first._M_sbuf;
> >@@ -395,11 +396,7 @@ _GLIBCXX_BEGIN_NAMESPACE_VERSION
> > 	      else
> > 		__c = __sb->snextc();
> > 	    }
> >-
> >-	  if (!traits_type::eq_int_type(__c, traits_type::eof()))
> >-	    __first._M_c = __c;
> >-	  else
> >-	    __first._M_sbuf = 0;
> >+	  __first._M_c = __c;
> > 	}
> >       return __first;
> >     }
> >diff --git a/libstdc++-v3/testsuite/24_iterators/istreambuf_iterator/3.cc
> >b/libstdc++-v3/testsuite/24_iterators/istreambuf_iterator/3.cc new file mode 100644
> >index 0000000..803ede4
> >--- /dev/null
> >+++ b/libstdc++-v3/testsuite/24_iterators/istreambuf_iterator/3.cc
> >@@ -0,0 +1,61 @@
> >+// { dg-options "-std=gnu++17" }
> >+
> >+// Copyright (C) 2017 Free Software Foundation, Inc.
> >+//
> >+// This file is part of the GNU ISO C++ Library.  This library is free
> >+// software; you can redistribute it and/or modify it under the
> >+// terms of the GNU General Public License as published by the
> >+// Free Software Foundation; either version 3, or (at your option)
> >+// any later version.
> >+
> >+// This library is distributed in the hope that it will be useful,
> >+// but WITHOUT ANY WARRANTY; without even the implied warranty of
> >+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> >+// GNU General Public License for more details.
> >+
> >+// You should have received a copy of the GNU General Public License along
> >+// with this library; see the file COPYING3.  If not see
> >+// <http://www.gnu.org/licenses/>.
> >+
> >+#include <algorithm>
> >+#include <sstream>
> >+#include <iterator>
> >+#include <cstring>
> >+#include <testsuite_hooks.h>
> >+
> >+void test03()
> >+{
> >+  using namespace std;
> >+  bool test __attribute__((unused)) = true;
> >+
> >+  std::stringstream s;
> >+  char b[] = "c2ee3d09-43b3-466d-b490-db35999a22cf";
> >+  char r[] = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx";
> >+  char q[] = "3c4852d6-d47b-4f46-b05e-b5edc1aa440e";
> >+  //          012345678901234567890123456789012345
> >+  //          0         1         2         3
> >+  s << b;
> >+  VERIFY( !s.fail() );
> >+
> >+  istreambuf_iterator<char> i(s);
> >+  copy_n(i, 36, r);
> >+  ++i; // EOF reached
> >+  VERIFY(i == std::istreambuf_iterator<char>());
> >+
> >+  VERIFY(memcmp(b, r, 36) == 0);
> >+
> >+  s << q;
> >+  VERIFY(!s.fail());
> >+
> >+  copy_n(i, 36, r);
> 
> This is undefined behaviour. The end-of-stream iterator value cannot
> be dereferenced.

Within this test istreambuf_iterator in eof state never dereferenced.
The test itself simulate "stop and go" istream usage.
stringstream is convenient for behaviuor illustration, but in "real life"
I can assume socket or tty on this place.

debugmode-dependent behaviour is also issue in this patch;
I don't suggest it as a separate patch because solutions are intersect.

WBR,

--

   - ptr

  reply	other threads:[~2017-09-28 12:06 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-09-23  7:10 Petr Ovtchenkov
2017-09-25 13:46 ` Jonathan Wakely
2017-09-28 10:34 ` Jonathan Wakely
2017-09-28 12:06   ` Petr Ovtchenkov [this message]
2017-09-28 12:38     ` Jonathan Wakely
2017-10-03 20:39       ` Petr Ovtchenkov
2017-10-04  5:04         ` [PATCH v2] " Petr Ovtchenkov
2017-10-06 16:01         ` [PATCH] libstdc++: istreambuf_iterator proxy (was: keep attached streambuf) François Dumont
2017-10-06 18:00           ` Petr Ovtchenkov
2017-10-08 14:59             ` [PATCH] libstdc++: istreambuf_iterator proxy François Dumont
2017-10-09 19:32               ` Petr Ovtchenkov
2017-10-10  5:52               ` Petr Ovtchenkov
2017-10-10 14:21           ` [PATCH] libstdc++: istreambuf_iterator proxy (was: keep attached streambuf) Jonathan Wakely
  -- strict thread matches above, loose matches on Subject: below --
2017-10-13 17:14 Make istreambuf_iterator::_M_sbuf immutable and add debug checks François Dumont
2017-10-23 19:08 ` François Dumont
2017-11-06 21:19   ` François Dumont
2017-11-16  5:52     ` Petr Ovtchenkov
2017-11-16 10:57       ` Jonathan Wakely
2017-11-16 11:46         ` Jonathan Wakely
2017-11-16 12:08           ` Petr Ovtchenkov
2017-11-16 17:40           ` François Dumont
2017-11-16 18:12             ` Petr Ovtchenkov
2017-11-16 21:31               ` François Dumont
2017-09-27 20:16 Make tests less istreambuf_iterator implementation dependent François Dumont
2017-09-28 12:12 ` Jonathan Wakely
2017-09-28 19:59   ` François Dumont
2017-09-28 21:56     ` Jonathan Wakely
2017-10-02  5:43       ` François Dumont
2017-10-03 14:20         ` Jonathan Wakely
2017-10-04 16:21           ` François Dumont
2017-10-04 23:23             ` Jonathan Wakely
2017-11-15 20:52             ` [PATCH 1/4] Revert "2017-10-04 Petr Ovtchenkov <ptr@void-ptr.info>" Petr Ovtchenkov
2017-11-15 20:52               ` [PATCH 2/4] libstdc++: istreambuf_iterator keep attached streambuf Petr Ovtchenkov
2017-11-15 20:52                 ` [PATCH 3/4] libstdc++: avoid character accumulation in istreambuf_iterator Petr Ovtchenkov
2017-11-15 20:52                   ` [PATCH 4/4] libstdc++: immutable _M_sbuf " Petr Ovtchenkov
2017-11-15 21:31                   ` [PATCH 3/4] libstdc++: avoid character accumulation " Paolo Carlini
2017-11-16  5:32                     ` Petr Ovtchenkov
2017-11-16  9:39                       ` Paolo Carlini
2017-11-16 11:03                         ` Petr Ovtchenkov
2017-11-16 11:29                           ` Paolo Carlini
2017-11-16 11:41                             ` Petr Ovtchenkov
2017-11-16 11:51                               ` Paolo Carlini
2017-11-16 10:56               ` [PATCH 1/4] Revert "2017-10-04 Petr Ovtchenkov <ptr@void-ptr.info>" Jonathan Wakely
2017-11-16 11:35                 ` Petr Ovtchenkov
2017-11-16 11:39                   ` Jonathan Wakely
2017-11-16 11:40                     ` Jonathan Wakely
2017-11-16 11:57                     ` Petr Ovtchenkov
2017-08-24 11:27 [PATCH] streambuf_iterator: avoid debug-dependent behaviour Petr Ovtchenkov
2017-08-29 20:02 ` François Dumont
2017-08-30  5:05   ` Petr Ovtchenkov
2017-08-31 20:30     ` François Dumont
2017-09-01  9:10 ` Jonathan Wakely
2017-09-07 21:02   ` François Dumont
2017-09-08  5:47     ` Petr Ovtchenkov
2017-09-08  6:15       ` François Dumont
2017-09-09 20:17       ` François Dumont
2017-09-21  5:46         ` François Dumont
2017-09-28 10:50           ` Jonathan Wakely
2017-09-28 10:58             ` Jonathan Wakely
2017-09-21 18:23   ` Petr Ovtchenkov
2017-09-25  9:34     ` Petr Ovtchenkov

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20170928150643.2f667ec9@void-ptr.info \
    --to=ptr@void-ptr.info \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=jwakely@redhat.com \
    --cc=libstdc++@gcc.gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).