From 52dba3b87428901693ba8e7f65197f8aa3cbf86c Mon Sep 17 00:00:00 2001
From: Tim McNamara <tim@dragonfly.co.nz>
Date: Wed, 2 Jun 2021 14:15:10 +1200
Subject: [PATCH] Benchmark other methods mentioned in README

The word boundary extensions to &str/String behave in a
very similiar, but not identical manner to .graphemes().
For example, Mandarin to slow(ish) on .graphemes() but
fast(ish) on .word_boundaries() whereas languages with
whitespace-delimited words tend to have the same per-
formance characteristics with the latter methods.

As the library develops, it would be worthwhile to
monitor the speed of the rest of the documented API.
---
 Cargo.toml               |  8 +++++
 benches/unicode_words.rs | 64 ++++++++++++++++++++++++++++++++++++++++
 benches/word_bounds.rs   | 64 ++++++++++++++++++++++++++++++++++++++++
 3 files changed, 136 insertions(+)
 create mode 100644 benches/unicode_words.rs
 create mode 100644 benches/word_bounds.rs

diff --git a/Cargo.toml b/Cargo.toml
index bf237cf..981394e 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -28,4 +28,12 @@ bencher = "0.1"
 
 [[bench]]
 name = "graphemes"
+harness = false
+
+[[bench]]
+name = "unicode_words"
+harness = false
+
+[[bench]]
+name = "word_bounds"
 harness = false
\ No newline at end of file
diff --git a/benches/unicode_words.rs b/benches/unicode_words.rs
new file mode 100644
index 0000000..731e325
--- /dev/null
+++ b/benches/unicode_words.rs
@@ -0,0 +1,64 @@
+#[macro_use]
+extern crate bencher;
+extern crate unicode_segmentation;
+
+use bencher::Bencher;
+use unicode_segmentation::UnicodeSegmentation;
+use std::fs;
+
+fn unicode_words(bench: &mut Bencher, path: &str) {
+    let text = fs::read_to_string(path).unwrap();
+    bench.iter(|| {
+        for w in text.unicode_words() {
+            bencher::black_box(w);
+        }
+    });
+
+    bench.bytes = text.len() as u64;
+}
+
+fn unicode_words_arabic(bench: &mut Bencher) {
+    unicode_words(bench, "benches/texts/arabic.txt");
+}
+
+fn unicode_words_english(bench: &mut Bencher) {
+    unicode_words(bench, "benches/texts/english.txt");
+}
+
+fn unicode_words_hindi(bench: &mut Bencher) {
+    unicode_words(bench, "benches/texts/hindi.txt");
+}
+
+fn unicode_words_japanese(bench: &mut Bencher) {
+    unicode_words(bench, "benches/texts/japanese.txt");
+}
+
+fn unicode_words_korean(bench: &mut Bencher) {
+    unicode_words(bench, "benches/texts/korean.txt");
+}
+
+fn unicode_words_mandarin(bench: &mut Bencher) {
+    unicode_words(bench, "benches/texts/mandarin.txt");
+}
+
+fn unicode_words_russian(bench: &mut Bencher) {
+    unicode_words(bench, "benches/texts/russian.txt");
+}
+
+fn unicode_words_source_code(bench: &mut Bencher) {
+    unicode_words(bench, "benches/texts/source_code.txt");
+}
+
+benchmark_group!(
+    benches,
+    unicode_words_arabic,
+    unicode_words_english,
+    unicode_words_hindi,
+    unicode_words_japanese,
+    unicode_words_korean,
+    unicode_words_mandarin,
+    unicode_words_russian,
+    unicode_words_source_code,
+);
+
+benchmark_main!(benches);
diff --git a/benches/word_bounds.rs b/benches/word_bounds.rs
new file mode 100644
index 0000000..035f57e
--- /dev/null
+++ b/benches/word_bounds.rs
@@ -0,0 +1,64 @@
+#[macro_use]
+extern crate bencher;
+extern crate unicode_segmentation;
+
+use bencher::Bencher;
+use unicode_segmentation::UnicodeSegmentation;
+use std::fs;
+
+fn word_bounds(bench: &mut Bencher, path: &str) {
+    let text = fs::read_to_string(path).unwrap();
+    bench.iter(|| {
+        for w in text.split_word_bounds() {
+            bencher::black_box(w);
+        }
+    });
+
+    bench.bytes = text.len() as u64;
+}
+
+fn word_bounds_arabic(bench: &mut Bencher) {
+    word_bounds(bench, "benches/texts/arabic.txt");
+}
+
+fn word_bounds_english(bench: &mut Bencher) {
+    word_bounds(bench, "benches/texts/english.txt");
+}
+
+fn word_bounds_hindi(bench: &mut Bencher) {
+    word_bounds(bench, "benches/texts/hindi.txt");
+}
+
+fn word_bounds_japanese(bench: &mut Bencher) {
+    word_bounds(bench, "benches/texts/japanese.txt");
+}
+
+fn word_bounds_korean(bench: &mut Bencher) {
+    word_bounds(bench, "benches/texts/korean.txt");
+}
+
+fn word_bounds_mandarin(bench: &mut Bencher) {
+    word_bounds(bench, "benches/texts/mandarin.txt");
+}
+
+fn word_bounds_russian(bench: &mut Bencher) {
+    word_bounds(bench, "benches/texts/russian.txt");
+}
+
+fn word_bounds_source_code(bench: &mut Bencher) {
+    word_bounds(bench, "benches/texts/source_code.txt");
+}
+
+benchmark_group!(
+    benches,
+    word_bounds_arabic,
+    word_bounds_english,
+    word_bounds_hindi,
+    word_bounds_japanese,
+    word_bounds_korean,
+    word_bounds_mandarin,
+    word_bounds_russian,
+    word_bounds_source_code,
+);
+
+benchmark_main!(benches);