[英]Compute `n * 10^p` as accurately as `f64::from_str` does?
I've got two values: n: f64
and p: i32
, and I need to compute n * 10^p
.我有两个值:
n: f64
和p: i32
,我需要计算n * 10^p
。
I tried two methods:我尝试了两种方法:
f64::powi
f64::powi
format!()
and f64::from_str
format!()
和f64::from_str
The latter is more accurate (see output below) but obviously inefficient.后者更准确(参见下面的 output)但显然效率低下。 Is there a way to get the same accuracy without going through a string conversion?
有没有办法在不经过字符串转换的情况下获得相同的准确性? Here's my code:
这是我的代码:
fn main() {
let f1 = |n: f64, e: i32| n * 10f64.powi(e);
let f2 = |n: f64, e: i32| format!("{}e{}", n, e).parse::<f64>().unwrap();
for &n in &[1.1, 2.2, 3.3, 4.4] {
for &e in &[-2, 2] {
println!("{} {}", f1(n, e), f2(n, e));
}
}
}
Output: Output:
0.011000000000000001 0.011
110.00000000000001 110
0.022000000000000002 0.022
220.00000000000003 220
0.033 0.033
330 330
0.044000000000000004 0.044
440.00000000000006 440
As always, there is a crate for it: rust_decimal It is not exactly, what you want, but it adds the needed precision to your task without going the way of formatting.与往常一样,它有一个板条箱: rust_decimal它不完全是您想要的,但它为您的任务增加了所需的精度,而无需进行格式化。 Maybe give it a try, it is unfortunately not possible to use it on the playground.
也许试一试,不幸的是不能在操场上使用它。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.