clang does not implement DR472

[991901f3-cc14-4404-b340-165691b62a58]

Bugzilla Link	16228
Version	unspecified
OS	All
CC	@DougGregor,@zygoloid

Extended Description

Per http://www.open-std.org/jtc1/sc22/wg21/docs/cwg_active.html#472 we should not allow the following code:

class B {
public:
  int m;
};

class S: private B {
  friend class N;
};
class N: private S {
  void f() {
    B* p = this;    // OK because class S satisfies the fourth condition
                    // above: B is a base class of N accessible in f() because
                    // B is an accessible base class of S and S is an accessible
                    // base class of N.
  }
};

class N2: protected B { };

class P2: public N2 {
  void f2(N2* n2p) {
    B* bp = n2p;    // error: invented member would be protected and naming
                    // class N2 not the same as or derived from the referencing
                    // class P2
    n2p->m = 0;     // error (cf 11.4 [class.protected]) for the same reason
  }
};

GCC correctly raises a diagnostic.

[991901f3-cc14-4404-b340-165691b62a58]

Sorry about that! I interpreted "Status: drafting" as tacit agreement that there was consensus on how to handle this case.

[ec04fc15-fa35-46f2-80e1-5d271f2ef708]

DR472 is still open, and the current wording on the core issues list is incorrect. We don't yet know what the revised rule will be, but I expect it'll end up mirroring 11.4/1's wording to the extent possible.

(FWIW, I've started to track Clang's state on various DRs at http://clang.llvm.org/cxx_dr_status.html, but I've not got as far as this one yet)

[arsenm]

clang and gcc error today:

/tmp/issue16602.cpp:25:10: error: 'm' is a protected member of 'B'
   25 |     n2p->m = 0;     // error (cf 11.4 [class.protected]) for the same reason
      |          ^
/tmp/issue16602.cpp:18:11: note: constrained by protected inheritance here
   18 | class N2: protected B { };
      |           ^~~~~~~~~~~
/tmp/issue16602.cpp:3:7: note: member is declared here
    3 |   int m;
      |       ^
1 error generated.

[shafik]

CC @Endilll

[Endilll]

Yeah, I'll get back to C++ DRs at some point.

[llvmbot]

@llvm/issue-subscribers-clang-frontend

[Endilll]

We don't issue a diagnostic for the key test case, so this is not resolved yet.