[Bug c++/12041] New: No overload search in evaluation of template parameter default value

[Bug c++/12041] New: No overload search in evaluation of template parameter default value

[raoulgough at yahoo dot co dot uk]

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

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12041

           Summary: No overload search in evaluation of template parameter
                    default value
           Product: gcc
           Version: 3.3.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: c++
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: raoulgough at yahoo dot co dot uk
                CC: gcc-bugs at gcc dot gnu dot org
  GCC host triplet: i686-pc-mingw

How to reproduce. File overload.cc:

-----8<-------
struct A { };
struct B { };

char foo (A const &);
short foo (B const &);

template<typename T, unsigned Size = sizeof(foo(T()))>
struct bar {};

int main () {
  bar<A>();
  bar<B>();
}
-----8<-------

$ g++ -c overload.cc
overload.cc: In function `int main()':
overload.cc:11: error: could not convert `A()' to `const B&'
overload.cc:5: error: in passing argument 1 of `short int foo(const B&)'
overload.cc:11: error: template argument 2 is invalid

I think that this is *different* to bug 6424 and core language issue 339 
because it does not involve any function template parameter deduction (and the 
error message is different, too). It seems like g++ is only considering the 
most recently seen version of foo(), since reversing the declaration order of 
the two overloads produces a transposed error message:

short foo (B const &);
char foo (A const &);

// ...

overload.cc: In function `int main()':
overload.cc:12: error: could not convert `B()' to `const A&'
overload.cc:5: error: in passing argument 1 of `char foo(const A&)'
overload.cc:12: error: template argument 2 is invalid

[Bug c++/12041] No overload search in evaluation of template parameter default value

[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=12041


pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
           Keywords|                            |rejects-valid
         Resolution|                            |FIXED
   Target Milestone|---                         |3.4


------- Additional Comments From pinskia at gcc dot gnu dot org  2003-08-23 13:26 -------
I can reproduce it in 3.3.1 but on the mainline (20030823), this is already fixed, since this is not a 
regression, I am closing this as fixed.