cppdraft_translate/cppdraft/time/traits.md

[time.traits]

# 30 Time library [[time]](./#time)

## 30.4 Time-related traits [time.traits]

### [30.4.1](#is.fp) treat_as_floating_point [[time.traits.is.fp]](time.traits.is.fp)

[🔗](#lib:treat_as_floating_point)

`template<class Rep> struct treat_as_floating_point : is_floating_point<Rep> { };
`

[1](#is.fp-1)

[#](http://github.com/Eelis/draft/tree/9adde4bc1c62ec234483e63ea3b70a59724c745a/source/time.tex#L1071)

The duration template uses the treat_as_floating_point trait to
help determine if a duration object can be converted to anotherduration with a different tick period[.](#is.fp-1.sentence-1)

Iftreat_as_floating_point_v<Rep> is true, then implicit conversions
are allowed among durations[.](#is.fp-1.sentence-2)

Otherwise, the implicit convertibility
depends on the tick periods of the durations[.](#is.fp-1.sentence-3)

[*Note [1](#is.fp-note-1)*:

The intention of this trait is to indicate whether a given class behaves like a floating-point
type, and thus allows division of one value by another with acceptable loss of precision[.](#is.fp-1.sentence-4)

Iftreat_as_floating_point_v<Rep> is false, Rep will be treated as
if it behaved like an integral type for the purpose of these conversions[.](#is.fp-1.sentence-5)

— *end note*]

### [30.4.2](#duration.values) duration_values [[time.traits.duration.values]](time.traits.duration.values)

[🔗](#lib:duration_values)

`template<class Rep>
  struct duration_values {
  public:
    static constexpr Rep zero() noexcept;
    static constexpr Rep min() noexcept;
    static constexpr Rep max() noexcept;
  };
`

[1](#duration.values-1)

[#](http://github.com/Eelis/draft/tree/9adde4bc1c62ec234483e63ea3b70a59724c745a/source/time.tex#L1098)

The duration template uses the duration_values trait to
construct special values of the duration's representation (Rep)[.](#duration.values-1.sentence-1)

This is
done because the representation can be a class type with behavior that
requires some other implementation to return these special values[.](#duration.values-1.sentence-2)

In that case,
the author of that class type should specialize duration_values to
return the indicated values[.](#duration.values-1.sentence-3)

[🔗](#lib:zero,duration_values)

`static constexpr Rep zero() noexcept;
`

[2](#duration.values-2)

[#](http://github.com/Eelis/draft/tree/9adde4bc1c62ec234483e63ea3b70a59724c745a/source/time.tex#L1112)

*Returns*: Rep(0)[.](#duration.values-2.sentence-1)

[*Note [1](#duration.values-note-1)*:

Rep(0) is specified instead ofRep() because Rep() can have some other meaning, such as an
uninitialized value[.](#duration.values-2.sentence-2)

— *end note*]

[3](#duration.values-3)

[#](http://github.com/Eelis/draft/tree/9adde4bc1c62ec234483e63ea3b70a59724c745a/source/time.tex#L1121)

*Remarks*: The value returned shall be the additive identity[.](#duration.values-3.sentence-1)

[🔗](#lib:min,duration_values)

`static constexpr Rep min() noexcept;
`

[4](#duration.values-4)

[#](http://github.com/Eelis/draft/tree/9adde4bc1c62ec234483e63ea3b70a59724c745a/source/time.tex#L1132)

*Returns*: numeric_limits<Rep>​::​lowest()[.](#duration.values-4.sentence-1)

[5](#duration.values-5)

[#](http://github.com/Eelis/draft/tree/9adde4bc1c62ec234483e63ea3b70a59724c745a/source/time.tex#L1136)

*Remarks*: The value returned shall compare less than or equal to zero()[.](#duration.values-5.sentence-1)

[🔗](#lib:max,duration_values)

`static constexpr Rep max() noexcept;
`

[6](#duration.values-6)

[#](http://github.com/Eelis/draft/tree/9adde4bc1c62ec234483e63ea3b70a59724c745a/source/time.tex#L1147)

*Returns*: numeric_limits<Rep>​::​max()[.](#duration.values-6.sentence-1)

[7](#duration.values-7)

[#](http://github.com/Eelis/draft/tree/9adde4bc1c62ec234483e63ea3b70a59724c745a/source/time.tex#L1151)

*Remarks*: The value returned shall compare greater than zero()[.](#duration.values-7.sentence-1)

### [30.4.3](#specializations) Specializations of common_type [[time.traits.specializations]](time.traits.specializations)

[🔗](#lib:common_type)

`template<class Rep1, class Period1, class Rep2, class Period2>
  struct common_type<chrono::duration<Rep1, Period1>, chrono::duration<Rep2, Period2>> {
    using type = chrono::duration<common_type_t<Rep1, Rep2>, see below>;
  };
`

[1](#specializations-1)

[#](http://github.com/Eelis/draft/tree/9adde4bc1c62ec234483e63ea3b70a59724c745a/source/time.tex#L1166)

The period of the duration indicated by this specialization ofcommon_type is the greatest common divisor of Period1 andPeriod2[.](#specializations-1.sentence-1)

[*Note [1](#specializations-note-1)*:

This can be computed by forming a ratio of the
greatest common divisor of Period1​::​num and Period2​::​num and the
least common multiple of Period1​::​den and Period2​::​den[.](#specializations-1.sentence-2)

— *end note*]

[2](#specializations-2)

[#](http://github.com/Eelis/draft/tree/9adde4bc1c62ec234483e63ea3b70a59724c745a/source/time.tex#L1176)

[*Note [2](#specializations-note-2)*:

The typedef name type is a synonym for theduration with the largest tick period possible where bothduration arguments will convert to it without requiring a division
operation[.](#specializations-2.sentence-1)

The representation of this type is intended to be able to hold any
value resulting from this conversion with no truncation error, although
floating-point durations can have round-off errors[.](#specializations-2.sentence-2)

— *end note*]

[🔗](#lib:common_type,duration)

`template<class Clock, class Duration1, class Duration2>
  struct common_type<chrono::time_point<Clock, Duration1>, chrono::time_point<Clock, Duration2>> {
    using type = chrono::time_point<Clock, common_type_t<Duration1, Duration2>>;
  };
`

[3](#specializations-3)

[#](http://github.com/Eelis/draft/tree/9adde4bc1c62ec234483e63ea3b70a59724c745a/source/time.tex#L1194)

The common type of two time_point types is a time_point with the same
clock as the two types and the common type of their two durations[.](#specializations-3.sentence-1)

### [30.4.4](#is.clock) Class template is_clock [[time.traits.is.clock]](time.traits.is.clock)

[🔗](#lib:is_clock)

`template<class T> struct is_clock;
`

[1](#is.clock-1)

[#](http://github.com/Eelis/draft/tree/9adde4bc1c62ec234483e63ea3b70a59724c745a/source/time.tex#L1205)

is_clock is a [*Cpp17UnaryTypeTrait*](meta.rqmts#:Cpp17UnaryTypeTrait "21.3.2 Requirements [meta.rqmts]") ([[meta.rqmts]](meta.rqmts "21.3.2 Requirements"))
with a base characteristic of true_type if T meets the [*Cpp17Clock*](time.clock.req#:Cpp17Clock "30.3 Cpp17Clock requirements [time.clock.req]") requirements ([[time.clock.req]](time.clock.req "30.3 Cpp17Clock requirements")),
otherwise false_type[.](#is.clock-1.sentence-1)

For the purposes of the specification of this trait,
the extent to which an implementation determines
that a type cannot meet the [*Cpp17Clock*](time.clock.req#:Cpp17Clock "30.3 Cpp17Clock requirements [time.clock.req]") requirements is unspecified,
except that as a minimum
a type T shall not qualify as a [*Cpp17Clock*](time.clock.req#:Cpp17Clock "30.3 Cpp17Clock requirements [time.clock.req]") unless it meets all of the following conditions:

- [(1.1)](#is.clock-1.1)

the [*qualified-id*](expr.prim.id.qual#nt:qualified-id "7.5.5.3 Qualified names [expr.prim.id.qual]")*s*T​::​rep,T​::​period,T​::​duration, andT​::​time_point are valid and each denotes a type ([[temp.deduct]](temp.deduct "13.10.3 Template argument deduction")),

- [(1.2)](#is.clock-1.2)

the expressionT​::​is_steady is well-formed when treated as an [unevaluated operand](expr.context#def:unevaluated_operand "7.2.3 Context dependence [expr.context]"),

- [(1.3)](#is.clock-1.3)

the expressionT​::​now() is well-formed when treated as an unevaluated operand[.](#is.clock-1.sentence-2)

[2](#is.clock-2)

[#](http://github.com/Eelis/draft/tree/9adde4bc1c62ec234483e63ea3b70a59724c745a/source/time.tex#L1231)

The behavior of a program that adds specializations for is_clock is undefined[.](#is.clock-2.sentence-1)