nlohmann::detail::dtoa_impl Namespace Reference

implements the Grisu2 algorithm for binary to decimal floating-point conversion. More...

Classes
struct	boundaries
struct	cached_power
struct	diyfp

Functions
	__attribute__ ((__nonnull__(1))) inline void grisu2(char *buf
	v = buf * 10^decimal_exponent len is the length of the buffer (number of decimal digits) The buffer must be large enough, i.e. More...
template<typename FloatType >
	__attribute__ ((__nonnull__(1))) void grisu2(char *buf
	v = buf * 10^decimal_exponent len is the length of the buffer (number of decimal digits) The buffer must be large enough, i.e. More...
return	append_exponent (buf, n - 1)
const diyfp	c_minus_k (cached.f, cached.e)
template<typename FloatType >
boundaries	compute_boundaries (FloatType value)
	Compute the (normalized) diyfp representing the input number 'value' and its boundaries. More...
int	find_largest_pow10 (const std::uint32_t n, std::uint32_t &pow10)
	For n != 0, returns k, such that pow10 := 10^(k-1) <= n < 10^k. More...
cached_power	get_cached_power_for_binary_exponent (int e)
	For a normalized diyfp w = f * 2^e, this function returns a (normalized) cached power-of-ten c = f_c * 2^e_c, such that the exponent of the product w * c satisfies (Definition 3.2 from [1]) More...
	grisu2 (buf, len, decimal_exponent, w.minus, w.w, w.plus)
	grisu2_digit_gen (buf, len, decimal_exponent, M_minus, w, M_plus)
void	grisu2_digit_gen (char *buffer, int &length, int &decimal_exponent, diyfp M_minus, diyfp w, diyfp M_plus)
	Generates V = buffer * 10^decimal_exponent, such that M- <= V <= M+. More...
void	grisu2_round (char *buf, int len, std::uint64_t dist, std::uint64_t delta, std::uint64_t rest, std::uint64_t ten_k)
static_cast< bool >(e< 1000) ? void(0)	if (e< 0)
	if (k<=n &&n<=max_exp)
	if (k==1)
template<typename Target , typename Source >
Target	reinterpret_bits (const Source source)
	static_cast< bool > (static_cast< bool >(value > 0) ? void(0) :__assert_fail("value > 0", __builtin_FILE(), __builtin_LINE(), __extension__ __PRETTY_FUNCTION__))

Variables
*	buf = static_cast<char>('0' + k / 10)
static_cast< bool >(m_plus.e==v.e) ? void(0) const cached_power	cached = get_cached_power_for_binary_exponent(m_plus.e)
int int &	decimal_exponent = -cached.k
int	e
	else
auto	k = static_cast<std::uint32_t>(e)
constexpr int	kAlpha = -60
constexpr int	kGamma = -32
int &	len
int int diyfp	m_minus
const diyfp	M_minus (w_minus.f+1, w_minus.e)
int int diyfp diyfp diyfp	m_plus
const diyfp	M_plus (w_plus.f - 1, w_plus.e)
int int int int	max_exp
int int int	min_exp
const int	n = len + decimal_exponent
int int diyfp diyfp	v
int int FloatType	value
const diyfp	w = diyfp::mul(v, c_minus_k)
const diyfp	w_minus = diyfp::mul(m_minus, c_minus_k)
const diyfp	w_plus = diyfp::mul(m_plus, c_minus_k)

Detailed Description

implements the Grisu2 algorithm for binary to decimal floating-point conversion.

This implementation is a slightly modified version of the reference implementation which may be obtained from http://florian.loitsch.com/publications (bench.tar.gz).

The code is distributed under the MIT license, Copyright (c) 2009 Florian Loitsch.

For a detailed description of the algorithm see:

[1] Loitsch, "Printing Floating-Point Numbers Quickly and Accurately with Integers", Proceedings of the ACM SIGPLAN 2010 Conference on Programming Language Design and Implementation, PLDI 2010 [2] Burger, Dybvig, "Printing Floating-Point Numbers Quickly and Accurately", Proceedings of the ACM SIGPLAN 1996 Conference on Programming Language Design and Implementation, PLDI 1996

Function Documentation

__attribute__() [1/2]

nlohmann::detail::dtoa_impl::__attribute__

(

(__nonnull__(1))

)

v = buf * 10^decimal_exponent len is the length of the buffer (number of decimal digits) The buffer must be large enough, i.e.

prettify v = buf * 10^decimal_exponent

appends a decimal representation of e to buf

>= max_digits10.

Returns

a pointer to the element following the exponent.

Precondition

-1000 < e < 1000

If v is in the range [10^min_exp, 10^max_exp) it will be printed in fixed-point notation. Otherwise it will be printed in exponential notation.

Precondition

min_exp < 0

max_exp > 0

__attribute__() [2/2]

template<typename FloatType >

nlohmann::detail::dtoa_impl::__attribute__

(

(__nonnull__(1))

)

v = buf * 10^decimal_exponent len is the length of the buffer (number of decimal digits) The buffer must be large enough, i.e.

>= max_digits10.

compute_boundaries()

template<typename FloatType >

boundaries nlohmann::detail::dtoa_impl::compute_boundaries

(

FloatType

value

)

Compute the (normalized) diyfp representing the input number 'value' and its boundaries.

Precondition

value must be finite and positive

find_largest_pow10()

int nlohmann::detail::dtoa_impl::find_largest_pow10	(	const std::uint32_t	n,
		std::uint32_t &	pow10
	)

For n != 0, returns k, such that pow10 := 10^(k-1) <= n < 10^k.

For n == 0, returns 1 and sets pow10 := 1.

get_cached_power_for_binary_exponent()

cached_power nlohmann::detail::dtoa_impl::get_cached_power_for_binary_exponent

(

int

e

)

For a normalized diyfp w = f * 2^e, this function returns a (normalized) cached power-of-ten c = f_c * 2^e_c, such that the exponent of the product w * c satisfies (Definition 3.2 from [1])

 alpha <= e_c + e + q <= gamma.

grisu2_digit_gen()

void nlohmann::detail::dtoa_impl::grisu2_digit_gen	(	char *	buffer,
		int &	length,
		int &	decimal_exponent,
		diyfp	M_minus,
		diyfp	w,
		diyfp	M_plus
	)

Generates V = buffer * 10^decimal_exponent, such that M- <= V <= M+.

M- and M+ must be normalized and share the same exponent -60 <= e <= -32.

References nlohmann::detail::array.

Variable Documentation

e

int nlohmann::detail::dtoa_impl::e

Initial value:

{
    (static_cast <bool> (e > -1000) ? void (0) : __assert_fail ("e > -1000", __builtin_FILE (), __builtin_LINE (), __extension__ __PRETTY_FUNCTION__))

else

nlohmann::detail::dtoa_impl::else

Initial value:

{
        *buf++ = '+'

m_plus

int int diyfp diyfp diyfp nlohmann::detail::dtoa_impl::m_plus

Initial value:

{
    (static_cast <bool> (m_plus.e == m_minus.e) ? void (0) : __assert_fail ("m_plus.e == m_minus.e", __builtin_FILE (), __builtin_LINE (), __extension__ __PRETTY_FUNCTION__))

max_exp

int int int int nlohmann::detail::dtoa_impl::max_exp

Initial value:

{
    (static_cast <bool> (min_exp < 0) ? void (0) : __assert_fail ("min_exp < 0", __builtin_FILE (), __builtin_LINE (), __extension__ __PRETTY_FUNCTION__))

value

int int FloatType nlohmann::detail::dtoa_impl::value

Initial value:

{
    static_assert(diyfp::kPrecision >= std::numeric_limits<FloatType>::digits + 3,
                  "internal error: not enough precision")