On Monday, 12 August 2019 09:58:39 CEST Lars Knoll wrote: > I’d personally be favour of using newer version of gcc/clang, but I’m not > sure we gain a lot with it, as Apple clang is then probably the limiting > factor. But we could upgrade that to Apple Clang 11 as well. Yeah, I was looking through https://en.cppreference.com/w/cpp/compiler_support for good c++20 features we might chose to opt-in on. At first it looked like there was a few promising options:
explicit(bool): We use some ugly patterns in places to simulate this. Requires gcc 9, clang 9, MSVC 19.22 char8_t: This has potential to break a bunch of stuff if we don't handle it explicitly, as per previous discussion it might be hard to handle with ifdefs. Requires gcc 9, clang 7, MSVC 19.22 Integrating feature-test macros: Could make >c++17 feature detection more elegant. Requires gcc 5, clang 3.4, MSVC 19.20 But then I noticed, they are all still missing from Apple clang 11 :( So, yes. Apple clang is the new minimum setter. What a world. 'Allan _______________________________________________ Development mailing list [email protected] https://lists.qt-project.org/listinfo/development
