In http://reviews.llvm.org/D7639#193443, @LegalizeAdulthood wrote:

> In http://reviews.llvm.org/D7639#192668, @danielmarjamaki wrote:
>
> > sorry but I am personally skeptic about this checker.
> >
> > why is the void removed?
> >
> > it does not cause any wrong behaviour to keep it.
> >
> > the void is not likely added there by mistake, is it? the developer 
> > probably wrote it by intention and this checker thinks that the developer 
> > intentions are wrong..
> >
> > how about moving it to clang-modernize?
>
>
> It is a readability check, it isn't designed to detect "wrong" behavior.  
> (void) is a C-ism and is a holdover from C-style coding.  It is completely 
> redundant and unnecessary in C++.  If a developer wants to keep unnecessary 
> and redundant tokens in their code, then they can turn this check off or not 
> run it.


Agree. Also, we want to migrate clang-modernize transformations to clang-tidy 
some day (it's an extremely low priority task though). Wrt this check, it might 
fit better in a new module named "legacy". Not extremely important and 
definitely not necessary for this patch.

What's necessary, is to address my comments 
(http://reviews.llvm.org/D7639?id=28202#inline-85173 and 
http://reviews.llvm.org/D7639?id=28202#inline-85176).


http://reviews.llvm.org/D7639

EMAIL PREFERENCES
  http://reviews.llvm.org/settings/panel/emailpreferences/



_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits

Reply via email to