Compares two values using 3-way comparison and produces a result of typestd::strong_ordering.

Lett andu be expressions andT andU denotedecltype((t)) anddecltype((u)) respectively,std::strong_order(t, u) isexpression-equivalent to:

• Ifstd::is_same_v<std::decay_t<T>,std::decay_t<U>> istrue:
  • std::strong_ordering(strong_order(t, u)), if it is a well-formed expression with overload resolution performed in a context that does not include a declaration ofstd::strong_order,
  • otherwise, ifT is a floating-point type:
    • ifstd::numeric_limits<T>::is_iec559 istrue, performs the ISO/IEC/IEEE 60559totalOrder comparison of floating-point values and returns that result as a value of typestd::strong_ordering (note: this comparison can distinguish between the positive and negative zero and between the NaNs with different representations),
    • otherwise, yields a value of typestd::strong_ordering that is consistent with the ordering observed byT's comparison operators,
  • otherwise,std::strong_ordering(std::compare_three_way()(t, u)) if it is well-formed.
• In all other cases, the expression is ill-formed, which can result insubstitution failure when it appears in the immediate context of a template instantiation.

Customization point objects

The namestd::strong_order denotes acustomization point object, which is a constfunction object of aliteralsemiregular class type. SeeCustomizationPointObject for details.

[edit]Strict total order of IEEE floating-point types

Letx andy be values of same IEEE floating-point type, andtotal_order_less(x, y) be the boolean result indicating ifx precedesy in the strict total order defined bytotalOrder in ISO/IEC/IEEE 60559.

(total_order_less(x, y)|| total_order_less(y, x))==false if and only ifx andy have the same bit pattern.

• if neitherx nory is NaN:
  • ifx< y, thentotal_order_less(x, y)==true;
  • ifx> y, thentotal_order_less(x, y)==false;
  • ifx== y,
    • ifx is negative zero andy is positive zero,total_order_less(x, y)==true,
    • ifx is not zero andx's exponent field is less thany's, thentotal_order_less(x, y)==(x>0) (only meaningful for decimal floating-point number);
• if eitherx ory is NaN:
  • ifx is negative NaN andy is not negative NaN, thentotal_order_less(x, y)==true,
  • ifx is not positive NaN andy is positive NaN, thentotal_order_less(x, y)==true,
  • if bothx andy are NaNs with the same sign andx's mantissa field is less thany's, thentotal_order_less(x, y)==!std::signbit(x).

[edit]Example

[edit]See also

• In other languages

• Deutsch
• Español
• 日本語
• Русский
• 中文