[Bug c++/38870] New: G++ allows modification of data members for a non lvalue object

[Bug c++/38870] New: G++ allows modification of data members for a non lvalue object

[stefan dot constantin at gmail dot com]

taking into account the following program
-------------------------------------------------
class tester {
public:
        int i;
        void foo() {}
};

int main()
{
        tester().foo();
        tester().i = 10;
        return 0;
}
-------------------------------------------------

The C++ standard (under 3.10.2 lvalues and rvalues) specifies that the result
from tester() is not an lvalue but implementations can invoke member functions
on the rvalue.
In this case tester().foo() is permitted, which is ok, but tester().i =
something should not be permitted since the implementation is only allowed to
invoke member functions.

Thank you,

Stefan Constantin


-- 
           Summary: G++ allows modification of data members for a non lvalue
                    object
           Product: gcc
           Version: 4.3.2
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: stefan dot constantin at gmail dot com
 GCC build triplet: x86_64-redhat-linux
  GCC host triplet: x86_64-redhat-linux
GCC target triplet: x86_64-redhat-linux


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38870

[Bug c++/38870] G++ allows modification of data members for a non lvalue object

[stefan dot constantin at gmail dot com]

------- Comment #1 from stefan dot constantin at gmail dot com  2009-01-16 12:07 -------
Sorry for not mentioning this earlier, 

the sample program provide compiles without error or warning with

g++ -Wall. 
I expected the compiler to error on the line "tester().i = 10;" because such a
construct requires an lvalue.

Thanks


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38870

[Bug c++/38870] G++ allows modification of data members for a non lvalue object

[pinskia at gcc dot gnu dot org]

------- Comment #2 from pinskia at gcc dot gnu dot org  2009-01-16 20:33 -------
Except really it is 5.2.5 that causes this to be invalid code.  Well the C++
standard as written in 1998, it is valid but there is a defect report against
it and it fixed this code to be invalid.

*** This bug has been marked as a duplicate of 15218 ***


-- 

pinskia at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38870