[PATCH] D49158: [clang-tidy] Fixing segfault when there's no IdentifierInfo

[Aaron Ballman via Phabricator via cfe-commits]

aaron.ballman added a comment.

In https://reviews.llvm.org/D49158#1172178, @juliehockett wrote:

> In https://reviews.llvm.org/D49158#1159882, @hokein wrote:
>
> > In https://reviews.llvm.org/D49158#1158327, @JonasToth wrote:
> >
> > > Is there a way to add a test, that would trigger the old segfault and show that it does not happen anymore with this fix?
> >
> >
> > +1, we should have a minimal test case for this fix, https://bugs.llvm.org/show_bug.cgi?id=36150 provides a case, but we need to reduce it (getting rid of the STD header).
>
>
> Does anyone know of a case where the base would not be a CXXRecordDecl that doesn't involve std::functional? (for background, I could only reproduce the error in the test case in https://bugs.llvm.org/show_bug.cgi?id=36150 on a Mac)


Perhaps something like:

  template <typename Ty>
  struct S : Ty {
    S(Ty Fn) : Ty(Fn) {}
  };
  
  void f() {
    S s{[](){}};
  }

Not certain if this is what you're looking for, but it is a trick used in <functional> implementations sometimes.


https://reviews.llvm.org/D49158