lebedev.ri added inline comments.
================
Comment at: clang-tidy/cppcoreguidelines/AvoidGotoCheck.cpp:21
+AST_MATCHER(GotoStmt, isForwardJumping) {
+ return Node.getLocStart() < Node.getLabel()->getLocStart();
+}
----------------
JonasToth wrote:
> lebedev.ri wrote:
> > Hm, on a second thought, i think this will have false-positive if the label
> > and the goto are on the same line, like
> > ```
> > goto label; ; label: ;
> > ```
> > I wonder we could **easily** compare accounting for the position in the
> > line, or it is not worth the extra complexity.
> Iam not sure what you mean. Is the concern that the `goto` does not actually
> skip something?
> If yes it is still worth it, because its not necessary to have the `goto`
> then.
>
> The `<` on the locations should always be sensefull because AFAIK its at byte
> level (or something similar).
>
> Your example is diagnosed correctly.
> The < on the locations should always be sensefull because AFAIK its at byte
> level (or something similar).
Ah, great then. I had some recollections of having some issues with that
before, but it makes sense that it would just work.
Repository:
rCTE Clang Tools Extra
https://reviews.llvm.org/D41815
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
