You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In general, we currently do not robustly handle custom arithmetic-like types. Note that we expect this use case to be very uncommon, so this is not a high-priority bug.
Until this is completely addressed, we should make sure to at least fail cleanly if custom arithmetic-like types are used. As an example, we currently use _STD nextafter on a duration::rep that could potentially be a custom arithmetic-like type, in which case it would fail cleanly.
I'm not sure there's much we can do until the Standard clarifies what the handwavy "a class emulating an arithmetic type" is supposed to mean and how the Standard Library is supposed to interact with such a thing.
Observed by @statementreply.
In general, we currently do not robustly handle custom arithmetic-like types. Note that we expect this use case to be very uncommon, so this is not a high-priority bug.
Until this is completely addressed, we should make sure to at least fail cleanly if custom arithmetic-like types are used. As an example, we currently use
_STD nextafter
on aduration::rep
that could potentially be a custom arithmetic-like type, in which case it would fail cleanly.STL/stl/inc/chrono
Lines 3243 to 3246 in 65eb507
The text was updated successfully, but these errors were encountered: