gribozavr added inline comments.
================
Comment at: clang-tools-extra/clang-tidy/modernize/MakeSmartPtrCheck.cpp:311
+ }
+ }
+ if (CEArg->isStdInitListInitialization())
----------------
So `Foo(Bar{1, 2})` is not problematic, it is just that we can't tell it apart
from `Foo({1,2})`?
================
Comment at: clang-tools-extra/test/clang-tidy/modernize-make-unique.cpp:457
// CHECK-MESSAGES: :[[@LINE-1]]:28: warning: use std::make_unique instead
- // CHECK-FIXES: std::unique_ptr<J> PJ2 = std::make_unique<J>(E{1, 2}, 1);
+ // CHECK-FIXES: std::unique_ptr<J> PJ2 = std::unique_ptr<J>(new J(E{1, 2},
1));
PJ2.reset(new J(E{1, 2}, 1));
----------------
hokein wrote:
> gribozavr wrote:
> > Why do we print the warning if the fixit is not fixing anything?
> We may still want to inform users, these cases can still be converted to
> `make_unique`, but the check currently is not smart enough to give correct
> fixes.
Sorry I don't understand -- so what the file used to say,
`std::make_unique<J>(E{1, 2}, 1)`, is correct, and this patch introduces a
regression? Or is some other syntax necessary?
Please add FIXMEs anyway.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D62736/new/
https://reviews.llvm.org/D62736
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
