Martin Moene

@trapexit Can you have a look if above change solves compiling with clang 3.4? Wasn't able to test it myself via [godbolt](https://godbolt.org/z/7e9YcP).

@trapexit Only now special-cased the type traits that are missing when using clang 3.4. Would appreciate it if you have a look if above change (take 2) solves compiling with...

@trapexit Fixed overlooked second template argument of `std11::is_asssignable`.

@trapexit I'm not very adept at Travis' usage and the various distributions one can or must use to certain effects, or deducing the information of interest from a log. That...

Hi @bibhas, I've not worked on this yet, and I do not expect it will happen any time soon I'm afraid :(

[](https://en.cppreference.com/w/cpp/header/type_traits) is a C++11 header. The first GCC compiler to support C++11 seems to be [gcc 4.3/4.8](https://en.cppreference.com/w/cpp/11) (see bottom of page). Doesn't seem like a good idea to me to...

Hi Arthur, Thanks for your reaction. While issuing, I already felt some reluctance. It was simply inspired on `std::span`'s constructors taking a container, however it's meaning with `ring_span` might be...

Or perhaps assume initially empty buffer: ``` return nonstd::ring_span(&arr[0], &arr[0] + N, &arr[0], 0); // ^ v return nonstd::ring_span(arr.begin(), arr.end(), arr.begin(), 0); ```

Added [another demo at godbolt](https://godbolt.org/z/cxfveE88s). Relevant talks: - [How to Write Well-Behaved Value Wrappers - Simon Brand [C++ on Sea 2019]](https://youtu.be/sQcPte0JNmE) - [C++ Function Templates: How Do They Really Work?...

Thanks for your analysis, Indeed `nonstd::optional` may have been intended to be a literal type under C++11/14, but clearly is not so now (as tests would have revealed :(. To...