[PATCH] D71963: clang-tidy doc: Add the severities description

Kim Viggedal via Phabricator via cfe-commits cfe-commits at lists.llvm.org
Mon Dec 30 06:03:29 PST 2019


vingeldal added a comment.

@whisperity I think you misunderstood my comment. I was not trying to give a more correct description of the current definition of style-level severity in CodeChecker. I was trying to propose **new** definitions of the different severity levels that this patch propose for clang-tidy.
So the claim that my suggestion is not "true" is invalid since that isn't a matter of true or false, but merely a matter of how one defines the style level.

My reason for proposing these new definitions is that I think my proposals highlight how I think that the levels could be more distinctly different in their definition instead. I also think these severity levels need more distinct definitions to be more useful than they are problematic. I'm afraid that definitions which people may interpret differently are at significant risk of causing more confusion than clarification.
If severity levels must be exactly like they are currently defined in CodeChecker then IMO that is a rather strong reason not to introduce them in clang-tidy and just keep that stuff in CodeChecker.

> A `low` diagnostics (and everything "above", assuming a(n at least) partial ordering on severities) should mean the coding construct is problematic: there is a chance -- perhaps once in a lifetime, perhaps not -- that doing this will cause a real error. A `style` thing should mean //"Hey, whatever you have written, is pretty much A-Ok! and works, congrats for writing valid (Obj)?C(++)?, but please realise that we are writing DERP/C++-42 and not C89 and please move your loop variable inside the loop's statement"//. No **real** "game-breaking" issue should ever arise from deciding on fixing or ignoring a `style` check's output.

If we are to only distinguish severity level based on the probability that violating the rule will cause an error then we should support every choice of severity level with data, something which may be quite hard, for most contributors writing a new check, to actually achieve.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D71963/new/

https://reviews.llvm.org/D71963





More information about the cfe-commits mailing list