Treat emoji presentation sequences as fullwidth

.github/workflows/rust.yml

@@ -7,7 +7,11 @@ on:
     branches: [ "master" ]
 
 env:
+  CARGO_INCREMENTAL: 0
   CARGO_TERM_COLOR: always
+  RUST_BACKTRACE: 1
+  RUSTFLAGS: -D warnings
+  RUSTDOCFLAGS: -D warnings
 
 jobs:
   build:
@@ -18,10 +22,19 @@ jobs:
       run: cargo build --verbose
     - name: Run tests
       run: cargo test --verbose
+    - name: Build docs
+      run: cargo doc
+    - name: Check formatting
+      run: cargo fmt --check
+    - name: Check clippy
+      run: cargo clippy --lib --tests
   regen:
     runs-on: ubuntu-latest
     steps:
     - uses: actions/checkout@v3
+    - uses: actions/setup-python@v5
+      with:
+        python-version: '3.12'
     - name: Regen
       run: cd scripts && python3 unicode.py
     - name: Diff

.gitignore

@@ -3,3 +3,4 @@ Cargo.lock
 scripts/tmp
 scripts/*.txt
 scripts/*.rs
+bench_data/*

Cargo.toml

@@ -2,20 +2,23 @@
 
 name = "unicode-width"
 version = "0.1.11"
-authors = ["kwantam <kwantam@gmail.com>", "Manish Goregaokar <manishsmail@gmail.com>"]
-
+authors = [
+    "kwantam <kwantam@gmail.com>",
+    "Manish Goregaokar <manishsmail@gmail.com>",
+]
 homepage = "https://github.com/unicode-rs/unicode-width"
 repository = "https://github.com/unicode-rs/unicode-width"
 documentation = "https://unicode-rs.github.io/unicode-width"
 license = "MIT/Apache-2.0"
 keywords = ["text", "width", "unicode"]
 readme = "README.md"
+edition = "2021"
 description = """
 Determine displayed width of `char` and `str` types
 according to Unicode Standard Annex #11 rules.
 """
 
-exclude = [ "target/*", "Cargo.lock" ]
+exclude = ["target/*", "Cargo.lock"]
 
 [dependencies]
 std = { version = "1.0", package = "rustc-std-workspace-std", optional = true }
@@ -27,7 +30,6 @@ unicode-normalization = "0.1.23"
 
 [features]
 default = []
-bench = []
 rustc-dep-of-std = ['std', 'core', 'compiler_builtins']
 
 # Legacy, now a no-op

benches/benches.rs

@@ -0,0 +1,113 @@
+// Copyright 2012-2015 The Rust Project Developers. See the COPYRIGHT
+// file at the top-level directory of this distribution and at
+// http://rust-lang.org/COPYRIGHT.
+//
+// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
+// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
+// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
+// option. This file may not be copied, modified, or distributed
+// except according to those terms.
+#![feature(test)]
+
+extern crate test;
+
+use std::iter;
+
+use test::Bencher;
+
+use unicode_width::{UnicodeWidthChar, UnicodeWidthStr};
+
+#[bench]
+fn cargo(b: &mut Bencher) {
+    let string = iter::repeat('a').take(4096).collect::<String>();
+
+    b.iter(|| {
+        for c in string.chars() {
+            test::black_box(UnicodeWidthChar::width(c));
+        }
+    });
+}
+
+#[bench]
+fn stdlib(b: &mut Bencher) {
+    let string = iter::repeat('a').take(4096).collect::<String>();
+
+    b.iter(|| {
+        for c in string.chars() {
+            test::black_box(c.width());
+        }
+    });
+}
+
+#[bench]
+fn simple_if(b: &mut Bencher) {
+    let string = iter::repeat('a').take(4096).collect::<String>();
+
+    b.iter(|| {
+        for c in string.chars() {
+            test::black_box(simple_width_if(c));
+        }
+    });
+}
+
+#[bench]
+fn simple_match(b: &mut Bencher) {
+    let string = iter::repeat('a').take(4096).collect::<String>();
+
+    b.iter(|| {
+        for c in string.chars() {
+            test::black_box(simple_width_match(c));
+        }
+    });
+}
+
+#[inline]
+fn simple_width_if(c: char) -> Option<usize> {
+    let cu = c as u32;
+    if cu < 127 {
+        if cu > 31 {
+            Some(1)
+        } else if cu == 0 {
+            Some(0)
+        } else {
+            None
+        }
+    } else {
+        UnicodeWidthChar::width(c)
+    }
+}
+
+#[inline]
+fn simple_width_match(c: char) -> Option<usize> {
+    match c as u32 {
+        cu if cu == 0 => Some(0),
+        cu if cu < 0x20 => None,
+        cu if cu < 0x7f => Some(1),
+        _ => UnicodeWidthChar::width(c),
+    }
+}
+
+#[bench]
+fn enwik8(b: &mut Bencher) {
+    // To benchmark, download & unzip `enwik8` from https://data.deepai.org/enwik8.zip
+    let data_path = "bench_data/enwik8";
+    let string = std::fs::read_to_string(data_path).unwrap_or_default();
+    b.iter(|| test::black_box(UnicodeWidthStr::width(string.as_str())));
+}
+
+#[bench]
+fn jawiki(b: &mut Bencher) {
+    // To benchmark, download & extract `jawiki-20240201-pages-articles-multistream-index.txt` from
+    // https://dumps.wikimedia.org/jawiki/20240201/jawiki-20240201-pages-articles-multistream-index.txt.bz2
+    let data_path = "bench_data/jawiki-20240201-pages-articles-multistream-index.txt";
+    let string = std::fs::read_to_string(data_path).unwrap_or_default();
+    b.iter(|| test::black_box(UnicodeWidthStr::width(string.as_str())));
+}
+
+#[bench]
+fn emoji(b: &mut Bencher) {
+    // To benchmark, download emoji-style.txt from https://www.unicode.org/emoji/charts/emoji-style.txt
+    let data_path = "bench_data/emoji-style.txt";
+    let string = std::fs::read_to_string(data_path).unwrap_or_default();
+    b.iter(|| test::black_box(UnicodeWidthStr::width(string.as_str())));
+}