1// Copyright 2012-2015 The Rust Project Developers. See the COPYRIGHT
2// file at the top-level directory of this distribution and at
3// http://rust-lang.org/COPYRIGHT.
4//
5// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
6// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
7// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
8// option. This file may not be copied, modified, or distributed
9// except according to those terms.
10
11//! Iterators which split strings on Grapheme Cluster, Word or Sentence boundaries, according
12//! to the [Unicode Standard Annex #29](http://www.unicode.org/reports/tr29/) rules.
13//!
14//! ```rust
15//! extern crate unicode_segmentation;
16//!
17//! use unicode_segmentation::UnicodeSegmentation;
18//!
19//! fn main() {
20//! let s = "a̐éö̲\r\n";
21//! let g = UnicodeSegmentation::graphemes(s, true).collect::<Vec<&str>>();
22//! let b: &[_] = &["a̐", "é", "ö̲", "\r\n"];
23//! assert_eq!(g, b);
24//!
25//! let s = "The quick (\"brown\") fox can't jump 32.3 feet, right?";
26//! let w = s.unicode_words().collect::<Vec<&str>>();
27//! let b: &[_] = &["The", "quick", "brown", "fox", "can't", "jump", "32.3", "feet", "right"];
28//! assert_eq!(w, b);
29//!
30//! let s = "The quick (\"brown\") fox";
31//! let w = s.split_word_bounds().collect::<Vec<&str>>();
32//! let b: &[_] = &["The", " ", "quick", " ", "(", "\"", "brown", "\"", ")", " ", "fox"];
33//! assert_eq!(w, b);
34//! }
35//! ```
36//!
37//! # no_std
38//!
39//! unicode-segmentation does not depend on libstd, so it can be used in crates
40//! with the `#![no_std]` attribute.
41//!
42//! # crates.io
43//!
44//! You can use this package in your project by adding the following
45//! to your `Cargo.toml`:
46//!
47//! ```toml
48//! [dependencies]
49//! unicode-segmentation = "1.9.0"
50//! ```
51
52#![deny(missing_docs, unsafe_code)]
53#![doc(
54 html_logo_url = "https://unicode-rs.github.io/unicode-rs_sm.png",
55 html_favicon_url = "https://unicode-rs.github.io/unicode-rs_sm.png"
56)]
57#![no_std]
58
59#[cfg(test)]
60#[macro_use]
61extern crate std;
62
63#[cfg(test)]
64#[macro_use]
65extern crate quickcheck;
66
67pub use grapheme::{GraphemeCursor, GraphemeIncomplete};
68pub use grapheme::{GraphemeIndices, Graphemes};
69pub use sentence::{USentenceBoundIndices, USentenceBounds, UnicodeSentences};
70pub use tables::UNICODE_VERSION;
71pub use word::{UWordBoundIndices, UWordBounds, UnicodeWordIndices, UnicodeWords};
72
73mod grapheme;
74#[rustfmt::skip]
75mod tables;
76mod sentence;
77mod word;
78
79#[cfg(test)]
80mod test;
81#[cfg(test)]
82mod testdata;
83
84/// Methods for segmenting strings according to
85/// [Unicode Standard Annex #29](http://www.unicode.org/reports/tr29/).
86pub trait UnicodeSegmentation {
87 /// Returns an iterator over the [grapheme clusters][graphemes] of `self`.
88 ///
89 /// [graphemes]: http://www.unicode.org/reports/tr29/#Grapheme_Cluster_Boundaries
90 ///
91 /// If `is_extended` is true, the iterator is over the
92 /// *extended grapheme clusters*;
93 /// otherwise, the iterator is over the *legacy grapheme clusters*.
94 /// [UAX#29](http://www.unicode.org/reports/tr29/#Grapheme_Cluster_Boundaries)
95 /// recommends extended grapheme cluster boundaries for general processing.
96 ///
97 /// # Examples
98 ///
99 /// ```
100 /// # use self::unicode_segmentation::UnicodeSegmentation;
101 /// let gr1 = UnicodeSegmentation::graphemes("a\u{310}e\u{301}o\u{308}\u{332}", true)
102 /// .collect::<Vec<&str>>();
103 /// let b: &[_] = &["a\u{310}", "e\u{301}", "o\u{308}\u{332}"];
104 ///
105 /// assert_eq!(&gr1[..], b);
106 ///
107 /// let gr2 = UnicodeSegmentation::graphemes("a\r\nb🇷🇺🇸🇹", true).collect::<Vec<&str>>();
108 /// let b: &[_] = &["a", "\r\n", "b", "🇷🇺", "🇸🇹"];
109 ///
110 /// assert_eq!(&gr2[..], b);
111 /// ```
112 fn graphemes<'a>(&'a self, is_extended: bool) -> Graphemes<'a>;
113
114 /// Returns an iterator over the grapheme clusters of `self` and their
115 /// byte offsets. See `graphemes()` for more information.
116 ///
117 /// # Examples
118 ///
119 /// ```
120 /// # use self::unicode_segmentation::UnicodeSegmentation;
121 /// let gr_inds = UnicodeSegmentation::grapheme_indices("a̐éö̲\r\n", true)
122 /// .collect::<Vec<(usize, &str)>>();
123 /// let b: &[_] = &[(0, "a̐"), (3, "é"), (6, "ö̲"), (11, "\r\n")];
124 ///
125 /// assert_eq!(&gr_inds[..], b);
126 /// ```
127 fn grapheme_indices<'a>(&'a self, is_extended: bool) -> GraphemeIndices<'a>;
128
129 /// Returns an iterator over the words of `self`, separated on
130 /// [UAX#29 word boundaries](http://www.unicode.org/reports/tr29/#Word_Boundaries).
131 ///
132 /// Here, "words" are just those substrings which, after splitting on
133 /// UAX#29 word boundaries, contain any alphanumeric characters. That is, the
134 /// substring must contain at least one character with the
135 /// [Alphabetic](http://unicode.org/reports/tr44/#Alphabetic)
136 /// property, or with
137 /// [General_Category=Number](http://unicode.org/reports/tr44/#General_Category_Values).
138 ///
139 /// # Example
140 ///
141 /// ```
142 /// # use self::unicode_segmentation::UnicodeSegmentation;
143 /// let uws = "The quick (\"brown\") fox can't jump 32.3 feet, right?";
144 /// let uw1 = uws.unicode_words().collect::<Vec<&str>>();
145 /// let b: &[_] = &["The", "quick", "brown", "fox", "can't", "jump", "32.3", "feet", "right"];
146 ///
147 /// assert_eq!(&uw1[..], b);
148 /// ```
149 fn unicode_words<'a>(&'a self) -> UnicodeWords<'a>;
150
151 /// Returns an iterator over the words of `self`, separated on
152 /// [UAX#29 word boundaries](http://www.unicode.org/reports/tr29/#Word_Boundaries), and their
153 /// offsets.
154 ///
155 /// Here, "words" are just those substrings which, after splitting on
156 /// UAX#29 word boundaries, contain any alphanumeric characters. That is, the
157 /// substring must contain at least one character with the
158 /// [Alphabetic](http://unicode.org/reports/tr44/#Alphabetic)
159 /// property, or with
160 /// [General_Category=Number](http://unicode.org/reports/tr44/#General_Category_Values).
161 ///
162 /// # Example
163 ///
164 /// ```
165 /// # use self::unicode_segmentation::UnicodeSegmentation;
166 /// let uwis = "The quick (\"brown\") fox can't jump 32.3 feet, right?";
167 /// let uwi1 = uwis.unicode_word_indices().collect::<Vec<(usize, &str)>>();
168 /// let b: &[_] = &[(0, "The"), (4, "quick"), (12, "brown"), (20, "fox"), (24, "can't"),
169 /// (30, "jump"), (35, "32.3"), (40, "feet"), (46, "right")];
170 ///
171 /// assert_eq!(&uwi1[..], b);
172 /// ```
173 fn unicode_word_indices<'a>(&'a self) -> UnicodeWordIndices<'a>;
174
175 /// Returns an iterator over substrings of `self` separated on
176 /// [UAX#29 word boundaries](http://www.unicode.org/reports/tr29/#Word_Boundaries).
177 ///
178 /// The concatenation of the substrings returned by this function is just the original string.
179 ///
180 /// # Example
181 ///
182 /// ```
183 /// # use self::unicode_segmentation::UnicodeSegmentation;
184 /// let swu1 = "The quick (\"brown\") fox".split_word_bounds().collect::<Vec<&str>>();
185 /// let b: &[_] = &["The", " ", "quick", " ", "(", "\"", "brown", "\"", ")", " ", "fox"];
186 ///
187 /// assert_eq!(&swu1[..], b);
188 /// ```
189 fn split_word_bounds<'a>(&'a self) -> UWordBounds<'a>;
190
191 /// Returns an iterator over substrings of `self`, split on UAX#29 word boundaries,
192 /// and their offsets. See `split_word_bounds()` for more information.
193 ///
194 /// # Example
195 ///
196 /// ```
197 /// # use self::unicode_segmentation::UnicodeSegmentation;
198 /// let swi1 = "Brr, it's 29.3°F!".split_word_bound_indices().collect::<Vec<(usize, &str)>>();
199 /// let b: &[_] = &[(0, "Brr"), (3, ","), (4, " "), (5, "it's"), (9, " "), (10, "29.3"),
200 /// (14, "°"), (16, "F"), (17, "!")];
201 ///
202 /// assert_eq!(&swi1[..], b);
203 /// ```
204 fn split_word_bound_indices<'a>(&'a self) -> UWordBoundIndices<'a>;
205
206 /// Returns an iterator over substrings of `self` separated on
207 /// [UAX#29 sentence boundaries](http://www.unicode.org/reports/tr29/#Sentence_Boundaries).
208 ///
209 /// Here, "sentences" are just those substrings which, after splitting on
210 /// UAX#29 sentence boundaries, contain any alphanumeric characters. That is, the
211 /// substring must contain at least one character with the
212 /// [Alphabetic](http://unicode.org/reports/tr44/#Alphabetic)
213 /// property, or with
214 /// [General_Category=Number](http://unicode.org/reports/tr44/#General_Category_Values).
215 ///
216 /// # Example
217 ///
218 /// ```
219 /// # use self::unicode_segmentation::UnicodeSegmentation;
220 /// let uss = "Mr. Fox jumped. [...] The dog was too lazy.";
221 /// let us1 = uss.unicode_sentences().collect::<Vec<&str>>();
222 /// let b: &[_] = &["Mr. ", "Fox jumped. ", "The dog was too lazy."];
223 ///
224 /// assert_eq!(&us1[..], b);
225 /// ```
226 fn unicode_sentences<'a>(&'a self) -> UnicodeSentences<'a>;
227
228 /// Returns an iterator over substrings of `self` separated on
229 /// [UAX#29 sentence boundaries](http://www.unicode.org/reports/tr29/#Sentence_Boundaries).
230 ///
231 /// The concatenation of the substrings returned by this function is just the original string.
232 ///
233 /// # Example
234 ///
235 /// ```
236 /// # use self::unicode_segmentation::UnicodeSegmentation;
237 /// let ssbs = "Mr. Fox jumped. [...] The dog was too lazy.";
238 /// let ssb1 = ssbs.split_sentence_bounds().collect::<Vec<&str>>();
239 /// let b: &[_] = &["Mr. ", "Fox jumped. ", "[...] ", "The dog was too lazy."];
240 ///
241 /// assert_eq!(&ssb1[..], b);
242 /// ```
243 fn split_sentence_bounds<'a>(&'a self) -> USentenceBounds<'a>;
244
245 /// Returns an iterator over substrings of `self`, split on UAX#29 sentence boundaries,
246 /// and their offsets. See `split_sentence_bounds()` for more information.
247 ///
248 /// # Example
249 ///
250 /// ```
251 /// # use self::unicode_segmentation::UnicodeSegmentation;
252 /// let ssis = "Mr. Fox jumped. [...] The dog was too lazy.";
253 /// let ssi1 = ssis.split_sentence_bound_indices().collect::<Vec<(usize, &str)>>();
254 /// let b: &[_] = &[(0, "Mr. "), (4, "Fox jumped. "), (16, "[...] "),
255 /// (22, "The dog was too lazy.")];
256 ///
257 /// assert_eq!(&ssi1[..], b);
258 /// ```
259 fn split_sentence_bound_indices<'a>(&'a self) -> USentenceBoundIndices<'a>;
260}
261
262impl UnicodeSegmentation for str {
263 #[inline]
264 fn graphemes(&self, is_extended: bool) -> Graphemes {
265 grapheme::new_graphemes(self, is_extended)
266 }
267
268 #[inline]
269 fn grapheme_indices(&self, is_extended: bool) -> GraphemeIndices {
270 grapheme::new_grapheme_indices(self, is_extended)
271 }
272
273 #[inline]
274 fn unicode_words(&self) -> UnicodeWords {
275 word::new_unicode_words(self)
276 }
277
278 #[inline]
279 fn unicode_word_indices(&self) -> UnicodeWordIndices {
280 word::new_unicode_word_indices(self)
281 }
282
283 #[inline]
284 fn split_word_bounds(&self) -> UWordBounds {
285 word::new_word_bounds(self)
286 }
287
288 #[inline]
289 fn split_word_bound_indices(&self) -> UWordBoundIndices {
290 word::new_word_bound_indices(self)
291 }
292
293 #[inline]
294 fn unicode_sentences(&self) -> UnicodeSentences {
295 sentence::new_unicode_sentences(self)
296 }
297
298 #[inline]
299 fn split_sentence_bounds(&self) -> USentenceBounds {
300 sentence::new_sentence_bounds(self)
301 }
302
303 #[inline]
304 fn split_sentence_bound_indices(&self) -> USentenceBoundIndices {
305 sentence::new_sentence_bound_indices(self)
306 }
307}
308