Stephan T. Lavavej

Neil omitted an important bit of info - this involves a very recent change to MSVC's STL, https://github.com/microsoft/STL/pull/4591 . To reproduce it, you'll either have to build and consume our...

Correct. To fix this, implement [CWG-2518](https://cplusplus.github.io/CWG/issues/2518.html) in Cling (if that's the tool that's emitting this error), allowing `static_assert(false)` to work in templates. MSVC and Clang have already implemented this Core...

It shipped in Clang 17, which is the version that microsoft/STL currently requires. https://clang.llvm.org/cxx_dr_status.html documents this. I *believe* that https://github.com/llvm/llvm-project/commit/00e2098bf49f0ed45b3b8c22894cd3ac9a530e0f implemented it, but there's a slight chance there were followup...

I don't actually know what the G694476FC error code is (the only codes I recognize are MSVC Cnnnn, IntelliSense Ennnn, and Clang named warnings/errors), but the "Unexpected size" is coming...

Thanks for noticing this! We did a bit of programmer-archaeology - what happened here is that `_STRINGIZE` is the classic macro we've used, `_CRT_STRINGIZE` was introduced during the UCRT split,...

Several different thoughts: * I don't think that, in principle, STL algorithms should be required to tolerate elements whose values change during the algorithm call, or predicates whose answer changes...

We talked about this at the weekly maintainer meeting and have no strong feelings about this - specifically, we're not inclined to remove the debug checks from `clamp` if they...

While I agree that the DevCom issue should be resolved as wontfix, we currently don't have a way to use a different test matrix for a subset of libcxx tests...

I think I'm still seeing this with the 17.11 Preview 1 toolset update.

* Pushed a source-conflict-free merge with `main` to pick up #4271. * Fully reduced and reported VSO-1949451 "EDG concepts rejects `std::expected` trying to propagate triviality of assignment operations". * Recorded...