https://github.com/upsj requested changes to this pull request.
I am wondering whether this new interface to resolveForwardingParameters is
generic enough - what if you have a function that - through multiple forwarding
layers chipping away at parameters from the front - passes some of the
arguments to a class initializer list, but some others to regular parameters?
Wouldn't it make more sense to use a `SmallVector<variant<ParmVarDecl,
FieldDecl>>` and resolve them individually?
Also, this looks like it would fail if we are also initializing base class
fields, e.g.
```
struct A {
int a;
};
struct B : A {
int b;
};
auto b = std::make_unique<B>(1, 2);
```
I think something like this would also be a good test case to add.
https://github.com/llvm/llvm-project/pull/176635
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
