| 1 |
use criterion::{black_box, criterion_group, criterion_main, Criterion}; |
| 2 |
use unicode_segmentation; |
| 3 |
|
| 4 |
use std::fs; |
| 5 |
use unicode_segmentation::UnicodeSegmentation; |
| 6 |
|
| 7 |
fn graphemes(c: &mut Criterion, lang: &str, path: &str) { |
| 8 |
let text = fs::read_to_string(path).unwrap(); |
| 9 |
|
| 10 |
c.bench_function(&format!("graphemes_{}", lang), |bench| { |
| 11 |
bench.iter(|| { |
| 12 |
for g in UnicodeSegmentation::graphemes(black_box(&*text), true) { |
| 13 |
black_box(g); |
| 14 |
} |
| 15 |
}) |
| 16 |
}); |
| 17 |
} |
| 18 |
|
| 19 |
fn graphemes_arabic(c: &mut Criterion) { |
| 20 |
graphemes(c, "arabic", "benches/texts/arabic.txt"); |
| 21 |
} |
| 22 |
|
| 23 |
fn graphemes_english(c: &mut Criterion) { |
| 24 |
graphemes(c, "english", "benches/texts/english.txt"); |
| 25 |
} |
| 26 |
|
| 27 |
fn graphemes_hindi(c: &mut Criterion) { |
| 28 |
graphemes(c, "hindi", "benches/texts/hindi.txt"); |
| 29 |
} |
| 30 |
|
| 31 |
fn graphemes_japanese(c: &mut Criterion) { |
| 32 |
graphemes(c, "japanese", "benches/texts/japanese.txt"); |
| 33 |
} |
| 34 |
|
| 35 |
fn graphemes_korean(c: &mut Criterion) { |
| 36 |
graphemes(c, "korean", "benches/texts/korean.txt"); |
| 37 |
} |
| 38 |
|
| 39 |
fn graphemes_mandarin(c: &mut Criterion) { |
| 40 |
graphemes(c, "mandarin", "benches/texts/mandarin.txt"); |
| 41 |
} |
| 42 |
|
| 43 |
fn graphemes_russian(c: &mut Criterion) { |
| 44 |
graphemes(c, "russian", "benches/texts/russian.txt"); |
| 45 |
} |
| 46 |
|
| 47 |
fn graphemes_source_code(c: &mut Criterion) { |
| 48 |
graphemes(c, "source_code", "benches/texts/source_code.txt"); |
| 49 |
} |
| 50 |
|
| 51 |
criterion_group!( |
| 52 |
benches, |
| 53 |
graphemes_arabic, |
| 54 |
graphemes_english, |
| 55 |
graphemes_hindi, |
| 56 |
graphemes_japanese, |
| 57 |
graphemes_korean, |
| 58 |
graphemes_mandarin, |
| 59 |
graphemes_russian, |
| 60 |
graphemes_source_code, |
| 61 |
); |
| 62 |
|
| 63 |
criterion_main!(benches); |