Prazek added a comment.
So the problem I got is that every time I want to check if there is already a
feature in clang-tidy/static-analyzer that solves my issue, I either have to
deal with static-analyzer command line, which is horrible, or I have to modify
and recompile the source code.
The case I was looking at is a check to find
int *p = g():
*p = 42;
if(!p) {g(): }
I wanted to find null checks after the pointer was dereferenced, and indeed I
found it in deadcode or unreachable code analysis in static-analyzer.
I am not sure what you are afraid of - as long as this feature is not turned by
default everything should be fine. If someone would turn it on then as long as
he doesn't see any false positives then everything is still fine, and it can
always disable flag.
AFAIK there is a way to specify in cl::opt that the option is hidden, which
could probably solve your concerns?
https://reviews.llvm.org/D28729
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
