[PATCH] D18475: [clang-tidy] Add more detection rules for redundant c_str calls.

[Etienne Bergeron via cfe-commits]

etienneb added a comment.

In http://reviews.llvm.org/D18475#383520, @Eugene.Zelenko wrote:

> Thank you for enhancement!
>
> I think will be good idea to extend check for any function call that receives .c_str(), since it mat be in other classes could have methods which receives both C++ or C strings. See also PR24870.


I was thinking the same when I started that patch. But, there is multiple cases where it will bring a false-positive.
As an example:

  string& insert (size_t pos, const string& str, size_t subpos, size_t sublen = npos);
  string& insert (size_t pos, const char* s, size_t n);

Even if the overloaded name/types are compatible, the semantic is not the same. You can't simply remove the .c_str() here.

I agree that a more general cases could exists, but I would like to be able to turn it off over a large code base when trying to keep false positives as low as posible.


http://reviews.llvm.org/D18475