mirror of
https://github.com/fastfloat/fast_float.git
synced 2026-07-31 00:36:39 +08:00
# alignment next step performance: original now perform my program test for 133704ms and my fork give more significant performance improve than before: 105259ms.
This commit is contained in:
parent
eef7133edc
commit
cd4aafd2f6
@ -288,7 +288,8 @@ enum class parse_error : uint_fast8_t {
|
||||
|
||||
template <typename UC> struct parsed_number_string_t {
|
||||
am_mant_t mantissa;
|
||||
|
||||
am_pow_t exponent;
|
||||
UC const *lastmatch;
|
||||
#ifndef FASTFLOAT_ONLY_POSITIVE_C_NUMBER_WO_INF_NAN
|
||||
bool negative;
|
||||
#endif
|
||||
@ -296,12 +297,9 @@ template <typename UC> struct parsed_number_string_t {
|
||||
bool too_many_digits;
|
||||
parse_error error;
|
||||
|
||||
am_pow_t exponent;
|
||||
|
||||
// contains the range of the significant digits
|
||||
span<UC const> integer; // non-nullable
|
||||
span<UC const> fraction; // nullable
|
||||
UC const *lastmatch;
|
||||
};
|
||||
|
||||
using byte_span = span<char const>;
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user