sidney added a comment. In http://reviews.llvm.org/D17043#390188, @alexfh wrote:
> What's the status of this patch? Do you still want to continue working on it > or are you fine with the warn_unused_result/nodiscard-based solution? I'm still interested in working on this, but I was waiting for y'all (maybe Richard?). IMHO, the current diagnostic for `warn_unused_result` is confusing (see the Google search and examples above <http://reviews.llvm.org/D17043#348569>). Adding a parameter sounds great as long as it supports both canned answers (e.g. `nodiscard(nosideeffects)`), to avoid having the same message in many places, and literal strings (`e.g. nodiscard("recv() returns the length of the received message, its return value should always be used.")`). If that won't happen and you all don't have a better idea, I can keep working on this patch. http://reviews.llvm.org/D17043 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
