Re: r263299 - Add fix-it for format-security warnings.

I think r263584 does what you suggest. Let me know if not. > On Mar 11, 2016, at 4:53 PM, Bob Wilson via cfe-commits > wrote: > > OK. I will do that. > >> On Mar 11, 2016, at 4:15 PM, David Blaikie > > wrote: >> >> >> >> On Fri, Mar 11, 2016 at 4:14 PM, Bob Wilson

Re: r263299 - Add fix-it for format-security warnings.

OK. I will do that. > On Mar 11, 2016, at 4:15 PM, David Blaikie wrote: > > > > On Fri, Mar 11, 2016 at 4:14 PM, Bob Wilson via cfe-commits > mailto:cfe-commits@lists.llvm.org>> wrote: > I’m not sure how to interpret that. It says: "Clang must recover from errors > as if the fix-it had been

Re: r263299 - Add fix-it for format-security warnings.

On Fri, Mar 11, 2016 at 4:14 PM, Bob Wilson via cfe-commits < cfe-commits@lists.llvm.org> wrote: > I’m not sure how to interpret that. It says: "Clang must recover from > errors as if the fix-it had been applied.” I suppose that format-security > could be an error if you’re building with -Werror,

Re: r263299 - Add fix-it for format-security warnings.

I’m not sure how to interpret that. It says: "Clang must recover from errors as if the fix-it had been applied.” I suppose that format-security could be an error if you’re building with -Werror, but I had been interpreting that to mean an error that would block further compilation. Can someone c

Re: r263299 - Add fix-it for format-security warnings.

I think http://clang.llvm.org/docs/InternalsManual.html#fix-it-hints says that if a fixit is on a warning, then clang should process the code as if the fixit had been applied. That's not the case here, so I think the fixit should be on a note instead. On Fri, Mar 11, 2016 at 4:55 PM, Bob Wilson vi