Integer traits
Unlike the standard C++ library where random variates precision is largely
implementation-defined, the <etf> library recognizes the need to control
the trade-off between speed and quality and requires thus the specification of
the number of random digits W to be used for random variates generation.
This additional flexibility brings the need for a generic way to select
integers able to hold the generated W random bits, a need that is addressed
by the integer_trait class:
template<std::size_t W>
struct integer_traits;
Template parameters
| Parameter | Description |
W | minimum number of digits (in bits) of the member types; if no signed integer or no unsigned integer can be found that can hold W bits, a static assertion is triggered (note that the availability of such integers for W>64 is platform-dependent) |
Member types
| Type | Description |
int_fast_t | A fast signed integer type with at least W bits |
int_least_t | A small signed integer type with at least W bits |
uint_fast_t | A fast unsigned integer type with at least W bits |
uint_least_t | A small unsigned integer type with at least W bits |