[Bug c++/50595] New: template overload resolution insufficiently sensitive to name dependency?

[Bug c++/50595] New: template overload resolution insufficiently sensitive to name dependency?

[zackw at panix dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50595

             Bug #: 50595
           Summary: template overload resolution insufficiently sensitive
                    to name dependency?
    Classification: Unclassified
           Product: gcc
           Version: 4.6.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: zackw@panix.com


Created attachment 25401
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=25401
test case

Per http://stackoverflow.com/questions/7630806 the expected output of the
attached test case should be

f(char): 1
f(int):  T(1)
f(int):  t
f(char): 1
f(char): T(1)
f(char): t

since the second and third calls to f() in g() have arguments that are
template-dependent, so resolution of *those* function calls (but not the first
call) should be deferred until the point of instantiation (h()), at which time
both f(int) and f(char) are visible. What actually happens with g++ 4.6 is

f(char): 1
f(char): T(1)
f(char): t
f(char): 1
f(char): T(1)
f(char): t

I'm not 100% convinced by the argument above, but I am convinced enough to see
what y'all think.

[Bug c++/50595] template overload resolution insufficiently sensitive to name dependency?

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50595

--- Comment #1 from Paolo Carlini <paolo.carlini at oracle dot com> 2011-10-03 10:04:17 UTC ---
Without having looked at the code, ICC behaves exactly like GCC.

[Bug c++/50595] template overload resolution insufficiently sensitive to name dependency?

[zackw at panix dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=50595

Zack Weinberg <zackw at panix dot com> changed:

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

--- Comment #2 from Zack Weinberg <zackw at panix dot com> ---
Per http://www.open-std.org/jtc1/sc22/wg21/docs/cwg_defects.html#197 GCC's
behavior is correct.