mirrors/coreutils
1
0
Fork 0
mirror of https://github.com/uutils/coreutils.git synced 2025-08-04 11:00:14 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 7

uucore: num_parser: Optimize bigdecimal create when exponent is 0

Browse source 
Makes creating float number without an exponent part quite a bit
faster. Saves about 9% speed in sort -g.
This commit is contained in:
Nicolas Boichat 2025-06-03 15:13:43 +02:00
parent 3b18316337
commit 7457a76f40
1 changed files with 17 additions and 12 deletions
Download patch file Download diff file Expand all files Collapse all files

29
src/uucore/src/lib/features/parser/num_parser.rs
View file

@ -67,15 +67,15 @@ impl Base {
&self,
str: &'a str,
digits: Option<BigUint>,
) -> (Option<BigUint>, u64, &'a str) {
) -> (Option<BigUint>, i64, &'a str) {
let mut digits: Option<BigUint> = digits;
let mut count: u64 = 0;
let mut count: i64 = 0;
let mut rest = str;
// Doing operations on BigUint is really expensive, so we do as much as we
// can on u64, then add them to the BigUint.
let mut digits_tmp: u64 = 0;
let mut count_tmp: u64 = 0;
let mut count_tmp: i64 = 0;
let mut mul_tmp: u64 = 1;
while let Some(d) = rest.chars().next().and_then(|c| self.digit(c)) {
(digits_tmp, count_tmp, mul_tmp) = (
@ -290,7 +290,7 @@ impl ExtendedParser for ExtendedBigDecimal {
}
}
fn parse_digits(base: Base, str: &str, fractional: bool) -> (Option<BigUint>, u64, &str) {
fn parse_digits(base: Base, str: &str, fractional: bool) -> (Option<BigUint>, i64, &str) {
// Parse the integral part of the number
let (digits, rest) = base.parse_digits(str);
@ -472,7 +472,7 @@ fn construct_extended_big_decimal<'a>(
digits: BigUint,
negative: bool,
base: Base,
scale: u64,
scale: i64,
exponent: BigInt,
) -> Result<ExtendedBigDecimal, ExtendedParserError<'a, ExtendedBigDecimal>> {
if digits == BigUint::zero() {
@ -490,14 +490,19 @@ fn construct_extended_big_decimal<'a>(
let bd = if scale == 0 && exponent.is_zero() {
BigDecimal::from_bigint(signed_digits, 0)
} else if base == Base::Decimal {
let new_scale = -exponent + scale;
// BigDecimal "only" supports i64 scale.
// Note that new_scale is a negative exponent: large positive value causes an underflow, large negative values an overflow.
if let Some(new_scale) = new_scale.to_i64() {
BigDecimal::from_bigint(signed_digits, new_scale)
if exponent.is_zero() {
// Optimization: Converting scale to Bigint and back is relatively slow.
BigDecimal::from_bigint(signed_digits, scale)
} else {
return Err(make_error(new_scale.is_negative(), negative));
let new_scale = -exponent + scale;
// BigDecimal "only" supports i64 scale.
// Note that new_scale is a negative exponent: large positive value causes an underflow, large negative values an overflow.
if let Some(new_scale) = new_scale.to_i64() {
BigDecimal::from_bigint(signed_digits, new_scale)
} else {
return Err(make_error(new_scale.is_negative(), negative));
}
}
} else if base == Base::Hexadecimal {
// pow "only" supports u32 values, just error out if given more than 2**32 fractional digits.