US334 27.05.10 [time.duration.io] Incorrect operator<< for floating-point durations LWG 3317

[wg21bot]

operator<< for floating-point durations always produces output with six digits after the decimal point, and doesn’t use the stream’s locale either.

Proposed change:
Rewrite the specification to not rely on to_string() for floating-point formatting.

[tahonermann]

SG16 plans to discuss this comment Tuesday morning:

• http://wiki.edg.com/bin/view/Wg21belfast/SG16

[HowardHinnant]

This would be consistent with the example implementation's behavior but needs to continue to respect width and alignment, which probably means using ostringstream like complex does. Piggy-backing on fmt for floating point might be a good way to go, not sure at this point. Or modeling after the complex streaming operator might be better.

[HowardHinnant]

Proposed wording:

Modify 27.5.10 I/O [time.duration.io]:

Effects: Forms a basic_string<charT, traits> from d.count() using
to_string if charT is char, or to_wstring if charT is wchar_t. Appends
the units suffix described below to the basic_string. Inserts the
resulting basic_string into os. [Note: This specification ensures that
the result of this streaming operation will obey the width and
alignment properties of the stream. —end note]

Inserts the duration d onto the stream os as if it were implemented
as follows:

basic_ostringstream<charT, traits> s;
s.flags(os.flags());
s.imbue(os.getloc());
s.precision(os.precision());
s << d.count() << units_suffix;
return os << s.str();

The units_suffix depends on the type `Period::type` as follows: ...

[Dani-Hub]

The associated issue is LWG 3317

[tahonermann]

SG16 in Belfast: Recommend to accept with modification to specify behavior in terms of ostringstream.

Howard's last comment reflects SG16's guidance.

Removing the SG16 label. Since the recommended resolution results in a clear behavioral change, I think this needs to go back to LEWG. Adding the LEWG label.

SG16 minutes are available at:

• http://wiki.edg.com/bin/view/Wg21belfast/SG16NBUS334

[tituswinters]

LEWG in Belfast: Forward LWG issue 3317 to LWG for C++20.

[JeffGarland]

LWG issue moved to ready as of 2019-11-07