Stephan T. Lavavej comments

Results 354 comments of


                                            Stephan T. Lavavej

trafficstars

STL: Non-standard non-reserved identifiers available via standard headers

As an aside, the Standard Library Modules are the first way to avoid this problem.

comparison of identical zoned_time objects fails if created in different dll contexts

Thanks for the self-contained test case! We talked about this at the weekly maintainer meeting; @dmitrykobets-msft noted that the C++ Standard doesn't recognize the existence of DLLs so this is...

<variant>: `std::variant` template param order determines if destructor is constant expression

> I have actually reported this compiler bug in october 2023 but its still unfixed What's the VS Developer Community bug number/link? > what are your thoughts on a fix...

<variant>: `std::variant` template param order determines if destructor is constant expression

Thanks, I've marked our internal mirror of that bug as "STL-requested" and noted the `std::variant` impact.

Should `std::error_code::message` respect the locale set by the user?

@jovibor Yep, agreed. Marking as "bug" but not resolving as a duplicate, until we review the pending PR.

`<cassert>`: assert's `_wassert` is shadowable

This should be reported to the VS Developer Community as it needs to be fixed in the UCRT. (I'd do it myself, but external customer reports are actually treated with...

<regex>: Should regex_match() use the match_results allocator?

@muellerj2 I am somewhat uncomfortable from a risk perspective with the idea of chaining together two significant changes into a single mega-fix, especially when the Standard doesn't speak clearly about...

<complex>: Remove complex<NonFloating>

Yes; in addition to the bincompat breaks, source compat breaks are acceptable when they have a purpose and are worth the disruption they cause.

`<concepts>`: `std::swap` for arrays uses `memcpy`

I would say that we're pretty confused about what "core headers" are supposed to be. Originally they were supposed to be free of runtime dependencies (e.g. the core of ``...

std::swap of arrays, why is there no specialization for trivial types

@monamimani , note that we can't simply use `memcpy`/`memmove` as they don't swap, they overwrite (so we'd need temporary storage). @frederick-vs-ja , I believe that the vectorized `swap_ranges` implementation is...