On 3/10/25 11:11 PM, Howard Hinnant wrote:
> On Mar 10, 2025, at 6:52 PM, Tom Honermann via Std-Proposals <std-proposals_at_[hidden]> wrote:
>> If it also adds constexprto the declarations of the time_point related functions
>
> Nik’s proposal does *not* add constexpr to the declarations of the time_point releated functions. … That would be redundant:
>
> https://eel.is/c++draft/time.point
Thank you. I had looked to see if time_point functions were already
declared constexpr, but I apparently failed that simple task :)
>
> One can already do plenty of time_point and calendrical compuations at compile time in C++20:
>
> #include <chrono>
>
> int
> main()
> {
> using namespace std::chrono;
> auto constexpr my_epoch = sys_days{January/1/2000} + 0h + 0min + 0s;
> static_assert(my_epoch - sys_days{1970y/1/1} == 946684800s);
> }

👍

Tom.

>
> Howard
>