[libcxx-commits] [PATCH] D59099: Integer and pointer types of 'midpoint' from P0811

[Marshall Clow via Phabricator via libcxx-commits]

mclow.lists marked 5 inline comments as done.
mclow.lists added inline comments.


================
Comment at: libcxx/include/numeric:533
+{
+    using _Up = std::make_unsigned_t<remove_cv_t<_Tp>>;
+
----------------
EricWF wrote:
> Does `const` or `volatile` occur for by-value template parameters?
I can remove that now that I realize that users are not supposed to explicitly list the template parameters.



================
Comment at: libcxx/include/numeric:551
+enable_if_t<is_pointer_v<_TPtr>, _TPtr>
+midpoint(_TPtr __a, _TPtr __b) noexcept
+{
----------------
EricWF wrote:
> Why the SFINAE?
> 
> Just declare it `midpoint(_Tp*, _Tp*)` ?
Because of this bad thing:
    midpoint<int>(0, 0)
returns a `int *`

Yes, I know that users are not supposed to do that, but this is easy prevention.


================
Comment at: libcxx/test/std/numerics/numeric.ops/numeric.ops.midpoint/midpoint.integer.pass.cpp:28
+    
+    assert(std::midpoint<T>(0,0)   == T(0));
+    assert(std::midpoint<T>(0,2)   == T(1));
----------------
EricWF wrote:
> How about cases where the sum is odd? AKA `0, 3`, `3, 0`, `4, -1`, and `-1, 4`).
I have more down below.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D59099/new/

https://reviews.llvm.org/D59099