From: Jonathan Wakely <jwakely@redhat.com>
To: Petr Ovtchenkov <ptr@void-ptr.info>
Cc: libstdc++@gcc.gnu.org, gcc-patches@gcc.gnu.org
Subject: Re: [PATCH] libstdc++: istreambuf_iterator keep attached streambuf
Date: Thu, 28 Sep 2017 12:38:00 -0000 [thread overview]
Message-ID: <20170928123806.GN4582@redhat.com> (raw)
In-Reply-To: <20170928150643.2f667ec9@void-ptr.info>
On 28/09/17 15:06 +0300, Petr Ovtchenkov wrote:
>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;
This variable should be removed, we don't need these variables any
more.
>> >+ 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.
That is quite implementation dependent.
The libc++ and VC++ implementations fail this test, because once an
istreambuf_iterator has been detected to reach end-of-stream it
doesn't get "reset" by changes to the streambuf.
The libc++ implementation crashes, because operator== on an
end-of-stream iterator sets its streambuf* to null, and any further
increment or dereference will segfault.
So this is testing something that other implementations don't support,
and isn't justifiable from the standard.
>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.
At the very minimum we should have a comment in the test explaining
how it relies on non-standard, non-portable behaviour.
But I'd prefer to avoid introducing more divergence from other
implementations.
>debugmode-dependent behaviour is also issue in this patch;
>I don't suggest it as a separate patch because solutions are intersect.
next prev parent reply other threads:[~2017-09-28 12:38 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
2017-09-28 12:38 ` Jonathan Wakely [this message]
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=20170928123806.GN4582@redhat.com \
--to=jwakely@redhat.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=libstdc++@gcc.gnu.org \
--cc=ptr@void-ptr.info \
/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).