On May 20, 2025, Jonathan Wakely wrote:
> On Tue, 20 May 2025, 05:00 Alexandre Oliva, wrote:
> OK
Thanks
> Maybe we should just define it in terms of std::stod, so it always exists
> but might not support the full accuracy of long double.
That sounds desirable to me, FWIW, but from past exper
On Tue, 20 May 2025, 05:00 Alexandre Oliva, wrote:
>
> basic_string.h doesn't define the non-w string version of std::stold
> when certain conditions aren't met, and then a couple of tests fail to
> compile.
>
> Guard the portions of the tests that depend on std::stold with the
> conditions for i
