| commit | 4affffc2e8292af18557e4afc2ccb4840d52bfda | [log] [tgz] |
|---|---|---|
| author | Scott Lobdell <slobdell@google.com> | Tue Feb 23 11:55:06 2021 -0800 |
| committer | Scott Lobdell <slobdell@google.com> | Tue Feb 23 11:55:06 2021 -0800 |
| tree | 20577b756f45c418a4746c9051bd5418266d8b69 | |
| parent | 59b437524da63c364898632c2d0e2dca86bb128d [diff] | |
| parent | 0b345a46228ab1c88b22de407a416f51be6a1ab3 [diff] |
Merge SP1A.210222.001 Change-Id: Ifa1fe62dd1b81a542e84d3859a9834d89eb88ce4
Pure Rust implementation of Ryū, an algorithm to quickly convert floating point numbers to decimal strings.
The PLDI'18 paper Ryū: fast float-to-string conversion by Ulf Adams includes a complete correctness proof of the algorithm. The paper is available under the creative commons CC-BY-SA license.
This Rust implementation is a line-by-line port of Ulf Adams' implementation in C, https://github.com/ulfjack/ryu.
Requirements: this crate supports any compiler version back to rustc 1.31; it uses nothing from the Rust standard library so is usable from no_std crates.
[dependencies] ryu = "1.0"
fn main() { let mut buffer = ryu::Buffer::new(); let printed = buffer.format(1.234); assert_eq!(printed, "1.234"); }
You can run upstream's benchmarks with:
$ git clone https://github.com/ulfjack/ryu c-ryu $ cd c-ryu $ bazel run -c opt //ryu/benchmark
And the same benchmark against our implementation with:
$ git clone https://github.com/dtolnay/ryu rust-ryu $ cd rust-ryu $ cargo run --example upstream_benchmark --release
These benchmarks measure the average time to print a 32-bit float and average time to print a 64-bit float, where the inputs are distributed as uniform random bit patterns 32 and 64 bits wide.
The upstream C code, the unsafe direct Rust port, and the safe pretty Rust API all perform the same, taking around 21 nanoseconds to format a 32-bit float and 31 nanoseconds to format a 64-bit float.
There is also a Rust-specific benchmark comparing this implementation to the standard library which you can run with:
$ cargo bench
The benchmark shows Ryū approximately 4-10x faster than the standard library across a range of f32 and f64 inputs. Measurements are in nanoseconds per iteration; smaller is better.
| type=f32 | 0.0 | 0.1234 | 2.718281828459045 | f32::MAX |
|---|---|---|---|---|
| RYU | 3ns | 28ns | 23ns | 22ns |
| STD | 40ns | 106ns | 128ns | 110ns |
| type=f64 | 0.0 | 0.1234 | 2.718281828459045 | f64::MAX |
|---|---|---|---|---|
| RYU | 3ns | 50ns | 35ns | 32ns |
| STD | 39ns | 105ns | 128ns | 202ns |
This library tends to produce more human-readable output than the standard library's to_string, which never uses scientific notation. Here are two examples:
Both libraries print short decimals such as 0.0000123 without scientific notation.