[Bug c++/13478] New: gcc uses wrong constructor to initialize a const reference

[Bug c++/13478] New: gcc uses wrong constructor to initialize a const reference

[erich_guenther at hotmail dot com]

#define FAILS
/* gcc (GCC) 3.3 20030226 (prerelease) (SuSE Linux)
tzst2.cpp: In function `void foo(const SA*)':
tzst2.cpp:29: error: `CA::CA(const CA&)' is private
tzst2.cpp:34: error: within this context
tzst2.cpp:34: error:   initializing temporary from result of `CA::CA(const SA&)
*/

// Data layout only
struct SA
{
};


class CA :protected SA
{
public:
	CA()
	{
	};

	CA(const SA& a)
	{
	};


#ifdef FAILS
	private:
	CA(const CA& a);
#endif
};

void foo(const SA * pa)
{
	const CA& ra=*pa;
}

-- 
           Summary: gcc uses wrong constructor to initialize a const
                    reference
           Product: gcc
           Version: 3.3
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: c++
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: erich_guenther at hotmail dot com
                CC: gcc-bugs at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13478

[Bug c++/13478] gcc uses wrong constructor to initialize a const reference

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2003-12-23 18:13 -------
The error is right you have to an accessorable copy constructor when initing a variable.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |INVALID


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13478

[Bug c++/13478] gcc uses wrong constructor to initialize a const reference

[giovannibajo at libero dot it]

------- Additional Comments From giovannibajo at libero dot it  2003-12-24 02:40 -------
That's not a variable, it's a reference. GCC should not call any constructor 
when binding the variable to that reference.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |UNCONFIRMED
         Resolution|INVALID                     |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13478

[Bug c++/13478] gcc uses wrong constructor to initialize a const reference

[giovannibajo at libero dot it]

------- Additional Comments From giovannibajo at libero dot it  2003-12-24 02:55 -------
Uhm, ok, it must call a copy constructor, but the first one.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13478

[Bug c++/13478] [3.3/3.4 Regression] gcc uses wrong constructor to initialize a const reference

[giovannibajo at libero dot it]

------- Additional Comments From giovannibajo at libero dot it  2004-01-11 12:24 -------
Ok. Given:

------------------------------------
struct A {};
struct B : protected A {
    B() {};
    B(const A& ) {};
private:
    B(const B& ) {};
};

void foo(const A* ap)
{
  const B& br = *ap;
}
------------------------------------

The code should compile, because it should call B(const A&) to construct the 
temporary which is then bound to the const reference. Instead, GCC seems to 
call B(const B&), which is private. I can't test on 3.0/3.1, but it's surely a 
regression since 3.0.4, and can be made a wrong-code one if we make the 
constructor public. I keep it as rejects-valid as it's easier to debug.


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |giovannibajo at libero dot
                   |                            |it
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
           Keywords|                            |rejects-valid
   Last reconfirmed|0000-00-00 00:00:00         |2004-01-11 12:24:53
               date|                            |
            Summary|gcc uses wrong constructor  |[3.3/3.4 Regression] gcc
                   |to initialize a const       |uses wrong constructor to
                   |reference                   |initialize a const reference
   Target Milestone|---                         |3.4.0
            Version|3.3                         |3.4.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13478

[Bug c++/13478] [3.3/3.4 Regression] gcc uses wrong constructor to initialize a const reference

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-01-15 05:36 -------
Failing since at least 2000-12-31.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13478

[Bug c++/13478] [3.3/3.4 Regression] gcc uses wrong constructor to initialize a const reference

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2004-01-16 16:59 -------
Subject: Bug 13478

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	mmitchel@gcc.gnu.org	2004-01-16 16:59:30

Modified files:
	gcc/cp         : ChangeLog call.c 
	gcc/testsuite  : ChangeLog 
Added files:
	gcc/testsuite/g++.dg/init: ref10.C 

Log message:
	PR c++/13478
	* call.c (convert_like_real): Do not perform an additional
	direct-initialization when binding to a reference.
	
	PR c++/13478
	* g++.dg/init/ref10.C: New test.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/ChangeLog.diff?cvsroot=gcc&r1=1.3887&r2=1.3888
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/call.c.diff?cvsroot=gcc&r1=1.449&r2=1.450
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&r1=1.3384&r2=1.3385
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/g++.dg/init/ref10.C.diff?cvsroot=gcc&r1=NONE&r2=1.1



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13478

[Bug c++/13478] [3.3/3.4 Regression] gcc uses wrong constructor to initialize a const reference

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2004-01-16 17:54 -------
Subject: Bug 13478

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	mmitchel@gcc.gnu.org	2004-01-16 17:54:25

Modified files:
	gcc/cp         : ChangeLog call.c 

Log message:
	PR c++/13478
	* call.c (initialize_reference): Pass -1 for inner parameter to
	convert_like_real.
	
	PR c++/13478
	* g++.dg/init/ref10.C: New test.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/ChangeLog.diff?cvsroot=gcc&r1=1.3888&r2=1.3889
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/call.c.diff?cvsroot=gcc&r1=1.450&r2=1.451



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13478

[Bug c++/13478] [3.3 Regression] gcc uses wrong constructor to initialize a const reference

[mmitchel at gcc dot gnu dot org]

------- Additional Comments From mmitchel at gcc dot gnu dot org  2004-01-16 17:57 -------
Fixed in GCC 3.4.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[3.3/3.4 Regression] gcc    |[3.3 Regression] gcc uses
                   |uses wrong constructor to   |wrong constructor to
                   |initialize a const reference|initialize a const reference


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13478

[Bug c++/13478] [3.3 Regression] gcc uses wrong constructor to initialize a const reference

[bangerth at dealii dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|3.4.0                       |3.3.3


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13478

[Bug c++/13478] [3.3 Regression] gcc uses wrong constructor to initialize a const reference

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2004-01-21 07:30 -------
Subject: Bug 13478

CVSROOT:	/cvs/gcc
Module name:	gcc
Branch: 	gcc-3_3-branch
Changes by:	gdr@gcc.gnu.org	2004-01-21 07:30:07

Modified files:
	gcc/cp         : ChangeLog call.c 
Added files:
	gcc/testsuite/g++.dg/init: ref10.C 

Log message:
	Backport from mainline
	2004-01-16  Mark Mitchell  <mark@codesourcery.com>
	
	PR c++/13478
	* call.c (initialize_reference): Pass -1 for inner parameter to
	convert_like_real.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/ChangeLog.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=1.3076.2.240&r2=1.3076.2.241
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/call.c.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=1.341.2.35&r2=1.341.2.36
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/g++.dg/init/ref10.C.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=NONE&r2=1.1.6.1



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13478

[Bug c++/13478] [3.3 Regression] gcc uses wrong constructor to initialize a const reference

[gdr at gcc dot gnu dot org]

------- Additional Comments From gdr at gcc dot gnu dot org  2004-01-21 07:31 -------
Fixed with a backport from mainline.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13478