Vasyl Gello

Good! and I will fix this: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=984682

> > Hm, I forgot about it. Adding it to my todolist. > > It's all good, I'll try and dig into it. I found an issue! The `std::chrono::system_clock::time_point` type...

And replacing `std::chrono::system_clock::time_point` with `std::chrono::time_point` solves the conversion: ``` $ ./test Expected timepoint repr: 2-05-12 Converted timepoint repr: 2-05-12 Date-conv timepoint repr: 2-05-12 ```

I am just testing it, and specified "double" everywhere to avoid implicit conversions resulting to loss of bitness. I will push more commits if the build that has just finished...

OK, I pushed another commit that solves the dialog issue:  > I really don't think we need to specify `double` everywhere. Why do you think that is necessary? The...

> I really don't think it needs to be used everywhere though, only as needed OK, then let's define what's needed beyond `CDateTime` so I can redo that commit. Can...

After another series of plying with reproducer I understand we have two options: 1. Use `std::chrono::time_point`. It has only microsecond precision but does not distort microseconds: ``` Start of 'long...

I think we reached the point we can start the actual review of this PR. First of all, I fixed the dialog issue. I decided to declare `m_time` as a...

Since you mentioned @HowardHinnant: Howard, can you please tag the new release? 3.0.0 has no fix for USE_OS_TZDB which you did in master :)

What I see from debug log is that the RebitTV script receives some unintialized reply from server (?) and throws a memory access error. Can you reproduce once more with...