murrayc added a comment. In http://reviews.llvm.org/D20857#449080, @alexfh wrote:
> Looks like a useful check to have. I'm not sure though, that it has anything > to do with "modernize". I'd suggest adding a new "bugprone" module (should be > added by http://reviews.llvm.org/D18821, hopefully soon) and moving the check > there. Fair enough. My logic is that this is a problem that can only be fixed properly in C++11 and that the best/correct/common way to do this has changed from C++98 to C++11. It's not just a nice use of new syntax (such as auto), it's also fixes bugs, but it's still use of new syntax. ================ Comment at: test/clang-tidy/modernize-explicit-operator-bool-void-pointer.cpp:6 @@ +5,3 @@ + operator const void *() const { + // CHECK-MESSAGES: :[[@LINE-1]]:3: warning: implicit operator const void* declaration should probably be explicit operator bool [modernize-explicit-operator-bool] + return reinterpret_cast<void *>(something != 0); ---------------- alexfh wrote: > From the first glance, this doesn't look like an easy mistake to make. Have > you actually seen this pattern in real code? Yes, in glibmm and gtkmm, which I maintain. This commit is from me though the idea wasn't: https://git.gnome.org/browse/gtkmm/commit/gtk/src/treerowreference.hg?id=c182608593e2d4799f523580a0532fbc68d296b2 We later used a typedef to make that clearer: https://git.gnome.org/browse/gtkmm/commit/gtk/src/treerowreference.hg?id=7dff74cca47827d6e34bc8f239674bf044ddedaa There's lots of mention of this in StackOverflow, though not always so clearly. For instance: https://stackoverflow.com/questions/9134888/is-using-void-instead-of-bool-an-advisable-practice http://reviews.llvm.org/D20857 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
