[Bug c++/12226] New: g++ fails to enforce accessibility requirement for copy constructor

[Bug c++/12226] New: g++ fails to enforce accessibility requirement for copy constructor

[austern at apple dot com]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226

           Summary: g++ fails to enforce accessibility requirement for copy
                    constructor
           Product: gcc
           Version: 3.4
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: c++
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: austern at apple dot com
                CC: gcc-bugs at gcc dot gnu dot org,gdr at integrable-
                    solutions dot net
 GCC build triplet: i686-pc-linux-gnu
  GCC host triplet: i686-pc-linux-gnu
GCC target triplet: i686-pc-linux-gnu

Consider the following code snippet, which is very similar to an example near the end of section 
8.5.3 in the C++ standard.
  struct A { protected: A(const A&); };
  struct B : public A { public: B(const B&); };
  extern B f();
  const A& ra = f();
Even with -W -Wall, this compiles cleanly with no diagnostics.

This is wrong.  This example is ill-formed, and a diagnostic is required.  A's copy constructor is not 
accessible in the initialization of ra, and the standard requires it to be.  From section 8.5.3, 
paragraph 5: "If the initializer expression is an rvalue, with T2 a class type, and cv1 T1 is 
reference-compatible with cv2 T2, the reference is bound in one of the following ways [description 
of the two ways omitted] ... The constructor that would be used to make the copy shall be callable 
whether or not the copy is actually done."

I believe that http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3471 was an incorrect fix, and that it 
disabled some correct diagnostics.  Accessibility checking is required whether or not the copy 
constructor is invoked, because eliding the copy constructor is just an optimization and 
correctness of a program does not depend on how smart a compiler's optimizer happens to be.

[Bug c++/12226] g++ fails to enforce accessibility requirement for copy constructor

[austern at apple dot com]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226


austern at apple dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |accepts-invalid

[Bug c++/12226] g++ fails to enforce accessibility requirement for copy constructor

[austern at apple dot com]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226



------- Additional Comments From austern at apple dot com  2003-09-09 18:02 -------
Oh, one more quote from the standard: clause 12.2, paragraph 1: "Temporaries of class types are 
created in various contexts ... Even when the creation of the temporary object is avoided, all the 
semantic checks must be respected as if the temporary object was created."

[Bug c++/12226] g++ fails to enforce accessibility requirement for copy constructor

[bangerth at dealii dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226



------- Additional Comments From bangerth at dealii dot org  2003-09-09 18:55 -------
Referring to the section you quote: it says that the compiler is free to choose either of 
the two possibilities. The second one is to convert to a temporay of type "cv1 T2", 
which in your case is "const B", just as explained in the example there. But the 
copy constructor of B _is_ accessible, so gcc would be allowed to accept the code. 
 
What do I miss? 
 
W.

[Bug c++/12226] g++ fails to enforce accessibility requirement for copy constructor

[pinskia at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226


pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |hg211 at ural2 dot hszk dot
                   |                            |bme dot hu


------- Additional Comments From pinskia at gcc dot gnu dot org  2003-10-14 16:39 -------
*** Bug 12591 has been marked as a duplicate of this bug. ***

[Bug c++/12226] [3.3/3.4 Regression] g++ fails to enforce accessibility requirement for copy constructor

[pinskia at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226


pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
   Last reconfirmed|0000-00-00 00:00:00         |2003-10-18 00:46:44
               date|                            |
            Summary|g++ fails to enforce        |[3.3/3.4 Regression] g++
                   |accessibility requirement   |fails to enforce
                   |for copy constructor        |accessibility requirement
                   |                            |for copy constructor
   Target Milestone|---                         |3.4


------- Additional Comments From pinskia at gcc dot gnu dot org  2003-10-18 00:46 -------
The example in this PR is not a good example.
Here is a better example for PR 12591:
class foo {
        private:
                foo(const foo &);
        public:
                foo();
};
const foo &bar = foo();

Here is the quote from clause 12.2, paragraph 1 which applies (for temporary objects):
"Example: even if the copy constructor is not called, all the semantic restrictions, such as 
accessibility (clause 11), shall be satisfied."
The part that applies is clause 12.2, paragraph 5, the example.
Since 3.3.3 most likely will not exist, setting target to 3.4.
: Search converges between 2002-10-11-trunk (#101) and 2002-10-12-trunk (#102).

[Bug c++/12226] [3.3/3.4 Regression] g++ fails to enforce accessibility requirement for copy constructor

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2003-12-20 04:08 -------
Mark you caused this regression, could you look into it?

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |mark at codesourcery dot com
                   |dot org                     |
             Status|NEW                         |ASSIGNED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226

[Bug c++/12226] [3.3/3.4 Regression] g++ fails to enforce accessibility requirement for copy constructor

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|normal                      |critical


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226

[Bug c++/12226] [3.3/3.4 Regression] g++ fails to enforce accessibility requirement for copy constructor

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2004-01-04 22:42 -------
Subject: Bug 12226

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	mmitchel@gcc.gnu.org	2004-01-04 22:42:22

Modified files:
	gcc/cp         : ChangeLog call.c parser.c 
	gcc/testsuite  : ChangeLog 
	libstdc++-v3   : ChangeLog 
	libstdc++-v3/testsuite/27_io/basic_filebuf: 4.cc 
	libstdc++-v3/testsuite/27_io/basic_fstream: 4.cc 
	libstdc++-v3/testsuite/27_io/basic_ifstream: 4.cc 
	libstdc++-v3/testsuite/27_io/basic_ios: 4.cc 
	libstdc++-v3/testsuite/27_io/basic_iostream: 4.cc 
	libstdc++-v3/testsuite/27_io/basic_istream: 4.cc 
	libstdc++-v3/testsuite/27_io/basic_istringstream: 4.cc 
	libstdc++-v3/testsuite/27_io/basic_ofstream: 4.cc 
	libstdc++-v3/testsuite/27_io/basic_ostream: 4.cc 
	libstdc++-v3/testsuite/27_io/basic_ostringstream: 4.cc 
	libstdc++-v3/testsuite/27_io/basic_stringbuf: 5.cc 
	libstdc++-v3/testsuite/27_io/basic_stringstream: 4.cc 
Added files:
	gcc/testsuite/g++.dg/init: copy7.C 
	gcc/testsuite/g++.dg/parse: cast1.C 

Log message:
	PR c++/12226
	* call.c (CHECK_COPY_CONSTRUCTOR_P): New macro.
	(reference_binding): Set it when appropriate.
	(build_temp): New function, split out from ...
	(convert_like_real): ... here.  Honor CHECK_COPY_CONSTRUCTOR_P.
	(initialize_reference): Likewise.
	
	PR c++/12226
	* g++.dg/init/copy7.c: New test.
	
	PR c++/12226
	* testsuite/27_io/basic_filebuf/4.cc: Remove use of invalid copy
	constructor.
	* testsuite/27_io/basic_fstream/4.cc: Likewise.
	* testsuite/27_io/basic_ifstream/4.cc: Likewise.
	* testsuite/27_io/basic_ios/4.cc: Likewise.
	* testsuite/27_io/basic_iostream/4.cc: Likewise.
	* testsuite/27_io/basic_istream/4.cc: Likewise.
	* testsuite/27_io/basic_istingstream/4.cc: Likewise.
	* testsuite/27_io/basic_ofstream/4.cc: Likewise.
	* testsuite/27_io/basic_ostream/4.cc: Likewise.
	* testsuite/27_io/basic_ostringstream/4.cc: Likewise.
	* testsuite/27_io/basic_stringbuf/5.cc: Likewise.
	* testsuite/27_io/basic_stringstream/4.cc: Likewise.
	
	PR c++/13536
	* parser.c (cp_parser): Add in_type_id_in_expr_p.
	(cp_parser_new): Initialize it.
	(cp_parser_postfix_expression): Set it.
	(cp_parser_sizeof_operand): Likewise.
	(cp_parser_parameteR_declaration): Do not commit early to tenative
	parsers when in_type_id_in_expr_p is set.
	
	PR c++/13536
	* g++.dg/parse/cast1.C: New test.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/ChangeLog.diff?cvsroot=gcc&r1=1.3848&r2=1.3849
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/call.c.diff?cvsroot=gcc&r1=1.447&r2=1.448
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/parser.c.diff?cvsroot=gcc&r1=1.143&r2=1.144
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&r1=1.3331&r2=1.3332
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/g++.dg/init/copy7.C.diff?cvsroot=gcc&r1=NONE&r2=1.1
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/g++.dg/parse/cast1.C.diff?cvsroot=gcc&r1=NONE&r2=1.1
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc++-v3/ChangeLog.diff?cvsroot=gcc&r1=1.2193&r2=1.2194
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc++-v3/testsuite/27_io/basic_filebuf/4.cc.diff?cvsroot=gcc&r1=1.2&r2=1.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc++-v3/testsuite/27_io/basic_fstream/4.cc.diff?cvsroot=gcc&r1=1.2&r2=1.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc++-v3/testsuite/27_io/basic_ifstream/4.cc.diff?cvsroot=gcc&r1=1.2&r2=1.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc++-v3/testsuite/27_io/basic_ios/4.cc.diff?cvsroot=gcc&r1=1.2&r2=1.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc++-v3/testsuite/27_io/basic_iostream/4.cc.diff?cvsroot=gcc&r1=1.2&r2=1.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc++-v3/testsuite/27_io/basic_istream/4.cc.diff?cvsroot=gcc&r1=1.2&r2=1.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc++-v3/testsuite/27_io/basic_istringstream/4.cc.diff?cvsroot=gcc&r1=1.2&r2=1.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc++-v3/testsuite/27_io/basic_ofstream/4.cc.diff?cvsroot=gcc&r1=1.2&r2=1.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc++-v3/testsuite/27_io/basic_ostream/4.cc.diff?cvsroot=gcc&r1=1.2&r2=1.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc++-v3/testsuite/27_io/basic_ostringstream/4.cc.diff?cvsroot=gcc&r1=1.2&r2=1.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc++-v3/testsuite/27_io/basic_stringbuf/5.cc.diff?cvsroot=gcc&r1=1.2&r2=1.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc++-v3/testsuite/27_io/basic_stringstream/4.cc.diff?cvsroot=gcc&r1=1.2&r2=1.3



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226

[Bug c++/12226] [3.3 Regression] g++ fails to enforce accessibility requirement for copy constructor

[mmitchel at gcc dot gnu dot org]

------- Additional Comments From mmitchel at gcc dot gnu dot org  2004-01-04 22:44 -------
Fixed in GCC 3.4.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED
            Summary|[3.3/3.4 Regression] g++    |[3.3 Regression] g++ fails
                   |fails to enforce            |to enforce accessibility
                   |accessibility requirement   |requirement for copy
                   |for copy constructor        |constructor


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226

[Bug c++/12226] [3.3 Regression] g++ fails to enforce accessibility requirement for copy constructor

[dbaron at dbaron dot org]

------- Additional Comments From dbaron at dbaron dot org  2004-01-05 22:52 -------
Before I go fix Mozilla to deal with this
( http://bugzilla.mozilla.org/show_bug.cgi?id=230118 ), I just want to
double-check that this change was intended to apply to calls to functions taking
reference types in addition to initialization of reference variables.  That is,
the program:

class C {
  public:  C() {}
  private: C(const C&) {}
};
                                                                                
void f(const C& c) { }
                                                                                
int main()
{
  f(C());
  return 0;
}

now yields an error.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dbaron at dbaron dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226

[Bug c++/12226] [3.3 Regression] g++ fails to enforce accessibility requirement for copy constructor

[mark at codesourcery dot com]

------- Additional Comments From mark at codesourcery dot com  2004-01-06 00:48 -------
Subject: Re:  [3.3 Regression] g++ fails to enforce
	accessibility requirement for copy constructor

On Mon, 2004-01-05 at 14:52, dbaron at dbaron dot org wrote:
> ------- Additional Comments From dbaron at dbaron dot org  2004-01-05 22:52 -------
> Before I go fix Mozilla to deal with this
> ( http://bugzilla.mozilla.org/show_bug.cgi?id=230118 ), I just want to
> double-check that this change was intended to apply to calls to functions taking
> reference types in addition to initialization of reference variables.  That is,
> the program:
> 
> class C {
>   public:  C() {}
>   private: C(const C&) {}
> };
>                                                                                 
> void f(const C& c) { }
>                                                                                 
> int main()
> {
>   f(C());
>   return 0;
> }
> 
> now yields an error.

You are correct; that program is invalid.  The copy constructor for C
must be accessible at the point that the rvalue is bound to the
reference, even though G++ will optimize away its use.



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226

[Bug c++/12226] [3.3 Regression] g++ fails to enforce accessibility requirement for copy constructor

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-02-02 22:17 -------
*** Bug 13986 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |bangerth at dealii dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226

[Bug c++/12226] [3.3 Regression] g++ fails to enforce accessibility requirement for copy constructor

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-08-27 08:46 -------
*** Bug 14213 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |pluto at pld-linux dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226

[Bug c++/12226] [3.3 Regression] g++ fails to enforce accessibility requirement for copy constructor

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-08-27 08:47 -------
*** Bug 14463 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |arnej at europe dot yahoo-
                   |                            |inc dot com


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226

[Bug c++/12226] [3.3 Regression] g++ fails to enforce accessibility requirement for copy constructor

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-08-27 08:48 -------
*** Bug 14823 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |robert dot schweikert at
                   |                            |abaqus dot com


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226

[Bug c++/12226] [3.3 Regression] g++ fails to enforce accessibility requirement for copy constructor

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-08-27 08:48 -------
*** Bug 15079 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |david dot lee at teracruz
                   |                            |dot com


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226

[Bug c++/12226] [3.3 Regression] g++ fails to enforce accessibility requirement for copy constructor

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-08-27 08:51 -------
*** Bug 15284 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |carlo at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226

[Bug c++/12226] [3.3 Regression] g++ fails to enforce accessibility requirement for copy constructor

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-08-27 08:45 -------
*** Bug 13986 has been marked as a duplicate of this bug. ***

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226

[Bug c++/12226] [3.3 Regression] g++ fails to enforce accessibility requirement for copy constructor

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-08-27 08:50 -------
*** Bug 16145 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |apl at alum dot mit dot edu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226

[Bug c++/12226] [3.3 Regression] g++ fails to enforce accessibility requirement for copy constructor

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-08-27 09:00 -------
*** Bug 15254 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |hyliviu-gcc at yahoo dot com


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12226