/rust/registry/src/index.crates.io-6f17d22bba15001f/unicode-width-0.1.13/src/lib.rs

Source (jump to first uncovered line)
// 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.

//! Determine displayed width of `char` and `str` types according to
//! [Unicode Standard Annex #11](http://www.unicode.org/reports/tr11/)
//! and other portions of the Unicode standard.
//! See the [Rules for determining width](#rules-for-determining-width) section
//! for the exact rules.
//!
//! This crate is `#![no_std]`.
//!
//! ```rust
//! use unicode_width::UnicodeWidthStr;
//!
//! let teststr = "Ｈｅｌｌｏ, ｗｏｒｌｄ!";
//! let width = UnicodeWidthStr::width(teststr);
//! println!("{}", teststr);
//! println!("The above string is {} columns wide.", width);
//! let width = teststr.width_cjk();
//! println!("The above string is {} columns wide (CJK).", width);
//! ```
//!
//! # Rules for determining width
//!
//! This crate currently uses the following rules to determine the width of a
//! character or string, in order of decreasing precedence. These may be tweaked in the future.
//!
//! 1. [Emoji presentation sequences] have width 2.
//! 2. Outside of an East Asian context, [text presentation sequences] have width 1
//!    if their base character:
//!    - Has the [`Emoji_Presentation`] property, and
//!    - Is not in the [Enclosed Ideographic Supplement] block.
//! 3. The sequence `"\r\n"` has width 1.
//! 4. [Lisu tone letter] combinations consisting of a character in the range `'\u{A4F8}'..='\u{A4FB}'`
//!    followed by a character in the range `'\u{A4FC}'..='\u{A4FD}'` have width 1.
//! 5. In an East Asian context only, `<`, `=`, or `>` have width 2 when followed by [`'\u{0338}'` COMBINING LONG SOLIDUS OVERLAY].
//! 6. [`'\u{115F}'` HANGUL CHOSEONG FILLER](https://util.unicode.org/UnicodeJsps/character.jsp?a=115F) has width 2.
//! 7. The following have width 0:
//!    - [Characters](https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5Cp%7BDefault_Ignorable_Code_Point%7D)
//!       with the [`Default_Ignorable_Code_Point`] property.
//!    - [Characters](https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5Cp%7BGrapheme_Extend%7D)
//!       with the [`Grapheme_Extend`] property.
//!    - The following 8 characters, all of which have NFD decompositions consisting of two [`Grapheme_Extend`] characters:
//!      - [`'\u{0CC0}'` KANNADA VOWEL SIGN II](https://util.unicode.org/UnicodeJsps/character.jsp?a=0CC0),
//!      - [`'\u{0CC7}'` KANNADA VOWEL SIGN EE](https://util.unicode.org/UnicodeJsps/character.jsp?a=0CC7),
//!      - [`'\u{0CC8}'` KANNADA VOWEL SIGN AI](https://util.unicode.org/UnicodeJsps/character.jsp?a=0CC8),
//!      - [`'\u{0CCA}'` KANNADA VOWEL SIGN O](https://util.unicode.org/UnicodeJsps/character.jsp?a=0CCA),
//!      - [`'\u{0CCB}'` KANNADA VOWEL SIGN OO](https://util.unicode.org/UnicodeJsps/character.jsp?a=0CCB),
//!      - [`'\u{1B3B}'` BALINESE VOWEL SIGN RA REPA TEDUNG](https://util.unicode.org/UnicodeJsps/character.jsp?a=1B3B),
//!      - [`'\u{1B3D}'` BALINESE VOWEL SIGN LA LENGA TEDUNG](https://util.unicode.org/UnicodeJsps/character.jsp?a=1B3D), and
//!      - [`'\u{1B43}'` BALINESE VOWEL SIGN PEPET TEDUNG](https://util.unicode.org/UnicodeJsps/character.jsp?a=1B43).
//!    - [Characters](https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5Cp%7BHangul_Syllable_Type%3DV%7D%5Cp%7BHangul_Syllable_Type%3DT%7D)
//!       with a [`Hangul_Syllable_Type`] of `Vowel_Jamo` (`V`) or `Trailing_Jamo` (`T`).
//!    - The following [`Prepended_Concatenation_Mark`]s:
//!      - [`'\u{0605}'` NUMBER MARK ABOVE](https://util.unicode.org/UnicodeJsps/character.jsp?a=0605),
//!      - [`'\u{070F}'` SYRIAC ABBREVIATION MARK](https://util.unicode.org/UnicodeJsps/character.jsp?a=070F),
//!      - [`'\u{0890}'` POUND MARK ABOVE](https://util.unicode.org/UnicodeJsps/character.jsp?a=0890),
//!      - [`'\u{0891}'` PIASTRE MARK ABOVE](https://util.unicode.org/UnicodeJsps/character.jsp?a=0891), and
//!      - [`'\u{08E2}'` DISPUTED END OF AYAH](https://util.unicode.org/UnicodeJsps/character.jsp?a=08E2).
//!    - [`'\u{A8FA}'` DEVANAGARI CARET](https://util.unicode.org/UnicodeJsps/character.jsp?a=A8FA).
//! 8. [Characters](https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5Cp%7BEast_Asian_Width%3DF%7D%5Cp%7BEast_Asian_Width%3DW%7D)
//!    with an [`East_Asian_Width`] of [`Fullwidth`] or [`Wide`] have width 2.
//! 9. Characters fulfilling all of the following conditions have width 2 in an East Asian context, and width 1 otherwise:
//!    - Has an [`East_Asian_Width`] of [`Ambiguous`], or
//!      has a canonical decomposition to an [`Ambiguous`] character followed by [`'\u{0338}'` COMBINING LONG SOLIDUS OVERLAY], or
//!      is [`'\u{0387}'` GREEK ANO TELEIA](https://util.unicode.org/UnicodeJsps/character.jsp?a=0387), and
//!    - Does not have a [`General_Category`] of `Modifier_Symbol`, and
//!    - Does not have a [`Script`] of `Latin`, `Greek`, or `Cyrillic`, or is a Roman numeral in the range `'\u{2160}'..='\u{217F}'`.
//! 10. All other characters have width 1.
//!
//! [`'\u{0338}'` COMBINING LONG SOLIDUS OVERLAY]: https://util.unicode.org/UnicodeJsps/character.jsp?a=0338
//!
//! [`Default_Ignorable_Code_Point`]: https://www.unicode.org/versions/Unicode15.0.0/ch05.pdf#G40095
//! [`East_Asian_Width`]: https://www.unicode.org/reports/tr11/#ED1
//! [`Emoji_Presentation`]: https://unicode.org/reports/tr51/#def_emoji_presentation
//! [`General_Category`]: https://www.unicode.org/versions/Unicode15.0.0/ch04.pdf#G124142
//! [`Grapheme_Extend`]: https://www.unicode.org/versions/Unicode15.0.0/ch03.pdf#G52443
//! [`Hangul_Syllable_Type`]: https://www.unicode.org/versions/Unicode15.0.0/ch03.pdf#G45593
//! [`Prepended_Concatenation_Mark`]: https://www.unicode.org/versions/Unicode15.0.0/ch23.pdf#G37908
//! [`Script`]: https://www.unicode.org/reports/tr24/#Script
//!
//! [`Fullwidth`]: https://www.unicode.org/reports/tr11/#ED2
//! [`Wide`]: https://www.unicode.org/reports/tr11/#ED4
//! [`Ambiguous`]: https://www.unicode.org/reports/tr11/#ED6
//!
//! [Emoji presentation sequences]: https://unicode.org/reports/tr51/#def_emoji_presentation_sequence
//! [text presentation sequences]: https://unicode.org/reports/tr51/#def_text_presentation_sequence
//!
//! [Enclosed Ideographic Supplement]: https://unicode.org/charts/nameslist/n_1F200.html
//!
//! [Lisu tone letter]: https://www.unicode.org/versions/Unicode15.0.0/ch18.pdf#G42078
//!
//! ## Canonical equivalence
//!
//! Canonically equivalent strings are assigned the same width (CJK and non-CJK).

#![forbid(unsafe_code)]
#![deny(missing_docs)]
#![doc(
    html_logo_url = "https://unicode-rs.github.io/unicode-rs_sm.png",
    html_favicon_url = "https://unicode-rs.github.io/unicode-rs_sm.png"
)]
#![no_std]

use tables::charwidth as cw;
pub use tables::UNICODE_VERSION;

mod tables;

/// Methods for determining displayed width of Unicode characters.
pub trait UnicodeWidthChar {
    /// Returns the character's displayed width in columns, or `None` if the
    /// character is a control character.
    ///
    /// This function treats characters in the Ambiguous category according
    /// to [Unicode Standard Annex #11](http://www.unicode.org/reports/tr11/)
    /// as 1 column wide. This is consistent with the recommendations for non-CJK
    /// contexts, or when the context cannot be reliably determined.
    fn width(self) -> Option<usize>;

    /// Returns the character's displayed width in columns, or `None` if the
    /// character is a control character.
    ///
    /// This function treats characters in the Ambiguous category according
    /// to [Unicode Standard Annex #11](http://www.unicode.org/reports/tr11/)
    /// as 2 columns wide. This is consistent with the recommendations for
    /// CJK contexts.
    fn width_cjk(self) -> Option<usize>;
}

impl UnicodeWidthChar for char {
    #[inline]
    fn width(self) -> Option<usize> {
        single_char_width(self, false)
    }

    #[inline]
    fn width_cjk(self) -> Option<usize> {
        single_char_width(self, true)
    }
}

/// Returns the [UAX #11](https://www.unicode.org/reports/tr11/) based width of `c`, or
/// `None` if `c` is a control character.
/// If `is_cjk == true`, ambiguous width characters are treated as double width; otherwise,
/// they're treated as single width.
#[inline]
fn single_char_width(c: char, is_cjk: bool) -> Option<usize> {
    if c < '\u{7F}' {
        if c >= '\u{20}' {
            // U+0020 to U+007F (exclusive) are single-width ASCII codepoints
            Some(1)
        } else {
            // U+0001 to U+0020 (exclusive) are control codes
            None
        }
    } else if c >= '\u{A0}' {
        // No characters >= U+00A0 are control codes, so we can consult the lookup tables
        Some(cw::lookup_width(c, is_cjk))
    } else {
        // U+007F to U+00A0 (exclusive) are control codes
        None
    }
}

/// Methods for determining displayed width of Unicode strings.
pub trait UnicodeWidthStr {
    /// Returns the string's displayed width in columns.
    ///
    /// This function treats characters in the Ambiguous category according
    /// to [Unicode Standard Annex #11](http://www.unicode.org/reports/tr11/)
    /// as 1 column wide. This is consistent with the recommendations for
    /// non-CJK contexts, or when the context cannot be reliably determined.
    fn width(&self) -> usize;

    /// Returns the string's displayed width in columns.
    ///
    /// This function treats characters in the Ambiguous category according
    /// to [Unicode Standard Annex #11](http://www.unicode.org/reports/tr11/)
    /// as 2 column wide. This is consistent with the recommendations for
    /// CJK contexts.
    fn width_cjk(&self) -> usize;
}

impl UnicodeWidthStr for str {
    #[inline]
    fn width(&self) -> usize {
        str_width(self, false)
    }

    #[inline]
    fn width_cjk(&self) -> usize {
        str_width(self, true)
    }
}

#[derive(Clone, Copy, Debug, Default, PartialEq, Eq)]
enum NextCharInfo {
    #[default]
    Default,
    /// `'\n'`
    LineFeed,
    /// '\u{0338}'
    /// For preserving canonical equivalence with CJK
    CombiningLongSolidusOverlay,
    /// `'\u{A4FC}'..='\u{A4FD}'`
    /// <https://www.unicode.org/versions/Unicode15.0.0/ch18.pdf#G42078>
    TrailingLisuToneLetter,
    /// `'\u{FE0E}'`
    Vs15,
    /// `'\u{FE0F}'`
    Vs16,
}

fn str_width(s: &str, is_cjk: bool) -> usize {
    s.chars()
        .rfold((0, NextCharInfo::Default), |(sum, next_info), c| {
            let (add, info) = width_in_str(c, is_cjk, next_info);
            (sum + add, info)
        })
        .0
}

/// Returns the [UAX #11](https://www.unicode.org/reports/tr11/) based width of `c`.
/// If `is_cjk == true`, ambiguous width characters are treated as double width; otherwise,
/// they're treated as single width.
#[inline]
fn width_in_str(c: char, is_cjk: bool, next_info: NextCharInfo) -> (usize, NextCharInfo) {
    if (is_cjk
        && next_info == NextCharInfo::CombiningLongSolidusOverlay
        && matches!(c, '<' | '=' | '>'))
        || (next_info == NextCharInfo::Vs16 && cw::starts_emoji_presentation_seq(c))
    {
        (2, NextCharInfo::Default)
    } else if c <= '\u{A0}' {
        match c {
            '\n' => (1, NextCharInfo::LineFeed),
            '\r' if next_info == NextCharInfo::LineFeed => (0, NextCharInfo::Default),
            _ => (1, NextCharInfo::Default),
        }
    } else {
        match (c, next_info) {
            ('\u{A4F8}'..='\u{A4FB}', NextCharInfo::TrailingLisuToneLetter) => {
                (0, NextCharInfo::Default)
            }
            ('\u{0338}', _) => (0, NextCharInfo::CombiningLongSolidusOverlay),
            ('\u{A4FC}'..='\u{A4FD}', _) => (1, NextCharInfo::TrailingLisuToneLetter),
            ('\u{FE0E}', _) => (0, NextCharInfo::Vs15),
            ('\u{FE0F}', _) => (0, NextCharInfo::Vs16),
            (_, NextCharInfo::Vs15)
                if !is_cjk && cw::starts_non_ideographic_text_presentation_seq(c) =>
            {
                (1, NextCharInfo::Default)
            }
            _ => (cw::lookup_width(c, is_cjk), NextCharInfo::Default),
        }
    }
}

Coverage Report

Created: 2024-10-16 07:58

Line	Count	Source (jump to first uncovered line)
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		//! Determine displayed width of `char` and `str` types according to
12		//! [Unicode Standard Annex #11](http://www.unicode.org/reports/tr11/)
13		//! and other portions of the Unicode standard.
14		//! See the [Rules for determining width](#rules-for-determining-width) section
15		//! for the exact rules.
16		//!
17		//! This crate is `#![no_std]`.
18		//!
19		//! ```rust
20		//! use unicode_width::UnicodeWidthStr;
21		//!
22		//! let teststr = "Ｈｅｌｌｏ, ｗｏｒｌｄ!";
23		//! let width = UnicodeWidthStr::width(teststr);
24		//! println!("{}", teststr);
25		//! println!("The above string is {} columns wide.", width);
26		//! let width = teststr.width_cjk();
27		//! println!("The above string is {} columns wide (CJK).", width);
28		//! ```
29		//!
30		//! # Rules for determining width
31		//!
32		//! This crate currently uses the following rules to determine the width of a
33		//! character or string, in order of decreasing precedence. These may be tweaked in the future.
34		//!
35		//! 1. [Emoji presentation sequences] have width 2.
36		//! 2. Outside of an East Asian context, [text presentation sequences] have width 1
37		//! if their base character:
38		//! - Has the [`Emoji_Presentation`] property, and
39		//! - Is not in the [Enclosed Ideographic Supplement] block.
40		//! 3. The sequence `"\r\n"` has width 1.
41		//! 4. [Lisu tone letter] combinations consisting of a character in the range `'\u{A4F8}'..='\u{A4FB}'`
42		//! followed by a character in the range `'\u{A4FC}'..='\u{A4FD}'` have width 1.
43		//! 5. In an East Asian context only, `<`, `=`, or `>` have width 2 when followed by [`'\u{0338}'` COMBINING LONG SOLIDUS OVERLAY].
44		//! 6. [`'\u{115F}'` HANGUL CHOSEONG FILLER](https://util.unicode.org/UnicodeJsps/character.jsp?a=115F) has width 2.
45		//! 7. The following have width 0:
46		//! - [Characters](https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5Cp%7BDefault_Ignorable_Code_Point%7D)
47		//! with the [`Default_Ignorable_Code_Point`] property.
48		//! - [Characters](https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5Cp%7BGrapheme_Extend%7D)
49		//! with the [`Grapheme_Extend`] property.
50		//! - The following 8 characters, all of which have NFD decompositions consisting of two [`Grapheme_Extend`] characters:
51		//! - [`'\u{0CC0}'` KANNADA VOWEL SIGN II](https://util.unicode.org/UnicodeJsps/character.jsp?a=0CC0),
52		//! - [`'\u{0CC7}'` KANNADA VOWEL SIGN EE](https://util.unicode.org/UnicodeJsps/character.jsp?a=0CC7),
53		//! - [`'\u{0CC8}'` KANNADA VOWEL SIGN AI](https://util.unicode.org/UnicodeJsps/character.jsp?a=0CC8),
54		//! - [`'\u{0CCA}'` KANNADA VOWEL SIGN O](https://util.unicode.org/UnicodeJsps/character.jsp?a=0CCA),
55		//! - [`'\u{0CCB}'` KANNADA VOWEL SIGN OO](https://util.unicode.org/UnicodeJsps/character.jsp?a=0CCB),
56		//! - [`'\u{1B3B}'` BALINESE VOWEL SIGN RA REPA TEDUNG](https://util.unicode.org/UnicodeJsps/character.jsp?a=1B3B),
57		//! - [`'\u{1B3D}'` BALINESE VOWEL SIGN LA LENGA TEDUNG](https://util.unicode.org/UnicodeJsps/character.jsp?a=1B3D), and
58		//! - [`'\u{1B43}'` BALINESE VOWEL SIGN PEPET TEDUNG](https://util.unicode.org/UnicodeJsps/character.jsp?a=1B43).
59		//! - [Characters](https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5Cp%7BHangul_Syllable_Type%3DV%7D%5Cp%7BHangul_Syllable_Type%3DT%7D)
60		//! with a [`Hangul_Syllable_Type`] of `Vowel_Jamo` (`V`) or `Trailing_Jamo` (`T`).
61		//! - The following [`Prepended_Concatenation_Mark`]s:
62		//! - [`'\u{0605}'` NUMBER MARK ABOVE](https://util.unicode.org/UnicodeJsps/character.jsp?a=0605),
63		//! - [`'\u{070F}'` SYRIAC ABBREVIATION MARK](https://util.unicode.org/UnicodeJsps/character.jsp?a=070F),
64		//! - [`'\u{0890}'` POUND MARK ABOVE](https://util.unicode.org/UnicodeJsps/character.jsp?a=0890),
65		//! - [`'\u{0891}'` PIASTRE MARK ABOVE](https://util.unicode.org/UnicodeJsps/character.jsp?a=0891), and
66		//! - [`'\u{08E2}'` DISPUTED END OF AYAH](https://util.unicode.org/UnicodeJsps/character.jsp?a=08E2).
67		//! - [`'\u{A8FA}'` DEVANAGARI CARET](https://util.unicode.org/UnicodeJsps/character.jsp?a=A8FA).
68		//! 8. [Characters](https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5Cp%7BEast_Asian_Width%3DF%7D%5Cp%7BEast_Asian_Width%3DW%7D)
69		//! with an [`East_Asian_Width`] of [`Fullwidth`] or [`Wide`] have width 2.
70		//! 9. Characters fulfilling all of the following conditions have width 2 in an East Asian context, and width 1 otherwise:
71		//! - Has an [`East_Asian_Width`] of [`Ambiguous`], or
72		//! has a canonical decomposition to an [`Ambiguous`] character followed by [`'\u{0338}'` COMBINING LONG SOLIDUS OVERLAY], or
73		//! is [`'\u{0387}'` GREEK ANO TELEIA](https://util.unicode.org/UnicodeJsps/character.jsp?a=0387), and
74		//! - Does not have a [`General_Category`] of `Modifier_Symbol`, and
75		//! - Does not have a [`Script`] of `Latin`, `Greek`, or `Cyrillic`, or is a Roman numeral in the range `'\u{2160}'..='\u{217F}'`.
76		//! 10. All other characters have width 1.
77		//!
78		//! [`'\u{0338}'` COMBINING LONG SOLIDUS OVERLAY]: https://util.unicode.org/UnicodeJsps/character.jsp?a=0338
79		//!
80		//! [`Default_Ignorable_Code_Point`]: https://www.unicode.org/versions/Unicode15.0.0/ch05.pdf#G40095
81		//! [`East_Asian_Width`]: https://www.unicode.org/reports/tr11/#ED1
82		//! [`Emoji_Presentation`]: https://unicode.org/reports/tr51/#def_emoji_presentation
83		//! [`General_Category`]: https://www.unicode.org/versions/Unicode15.0.0/ch04.pdf#G124142
84		//! [`Grapheme_Extend`]: https://www.unicode.org/versions/Unicode15.0.0/ch03.pdf#G52443
85		//! [`Hangul_Syllable_Type`]: https://www.unicode.org/versions/Unicode15.0.0/ch03.pdf#G45593
86		//! [`Prepended_Concatenation_Mark`]: https://www.unicode.org/versions/Unicode15.0.0/ch23.pdf#G37908
87		//! [`Script`]: https://www.unicode.org/reports/tr24/#Script
88		//!
89		//! [`Fullwidth`]: https://www.unicode.org/reports/tr11/#ED2
90		//! [`Wide`]: https://www.unicode.org/reports/tr11/#ED4
91		//! [`Ambiguous`]: https://www.unicode.org/reports/tr11/#ED6
92		//!
93		//! [Emoji presentation sequences]: https://unicode.org/reports/tr51/#def_emoji_presentation_sequence
94		//! [text presentation sequences]: https://unicode.org/reports/tr51/#def_text_presentation_sequence
95		//!
96		//! [Enclosed Ideographic Supplement]: https://unicode.org/charts/nameslist/n_1F200.html
97		//!
98		//! [Lisu tone letter]: https://www.unicode.org/versions/Unicode15.0.0/ch18.pdf#G42078
99		//!
100		//! ## Canonical equivalence
101		//!
102		//! Canonically equivalent strings are assigned the same width (CJK and non-CJK).
103
104		#![forbid(unsafe_code)]
105		#![deny(missing_docs)]
106		#![doc(
107		html_logo_url = "https://unicode-rs.github.io/unicode-rs_sm.png",
108		html_favicon_url = "https://unicode-rs.github.io/unicode-rs_sm.png"
109		)]
110		#![no_std]
111
112		use tables::charwidth as cw;
113		pub use tables::UNICODE_VERSION;
114
115		mod tables;
116
117		/// Methods for determining displayed width of Unicode characters.
118		pub trait UnicodeWidthChar {
119		/// Returns the character's displayed width in columns, or `None` if the
120		/// character is a control character.
121		///
122		/// This function treats characters in the Ambiguous category according
123		/// to [Unicode Standard Annex #11](http://www.unicode.org/reports/tr11/)
124		/// as 1 column wide. This is consistent with the recommendations for non-CJK
125		/// contexts, or when the context cannot be reliably determined.
126		fn width(self) -> Option<usize>;
127
128		/// Returns the character's displayed width in columns, or `None` if the
129		/// character is a control character.
130		///
131		/// This function treats characters in the Ambiguous category according
132		/// to [Unicode Standard Annex #11](http://www.unicode.org/reports/tr11/)
133		/// as 2 columns wide. This is consistent with the recommendations for
134		/// CJK contexts.
135		fn width_cjk(self) -> Option<usize>;
136		}
137
138		impl UnicodeWidthChar for char {
139		#[inline]
140	0	fn width(self) -> Option<usize> {
141	0	single_char_width(self, false)
142	0	}
143
144		#[inline]
145	0	fn width_cjk(self) -> Option<usize> {
146	0	single_char_width(self, true)
147	0	}
148		}
149
150		/// Returns the [UAX #11](https://www.unicode.org/reports/tr11/) based width of `c`, or
151		/// `None` if `c` is a control character.
152		/// If `is_cjk == true`, ambiguous width characters are treated as double width; otherwise,
153		/// they're treated as single width.
154		#[inline]
155	0	fn single_char_width(c: char, is_cjk: bool) -> Option<usize> {
156	0	if c < '\u{7F}' {
157	0	if c >= '\u{20}' {
158		// U+0020 to U+007F (exclusive) are single-width ASCII codepoints
159	0	Some(1)
160		} else {
161		// U+0001 to U+0020 (exclusive) are control codes
162	0	None
163		}
164	0	} else if c >= '\u{A0}' {
165		// No characters >= U+00A0 are control codes, so we can consult the lookup tables
166	0	Some(cw::lookup_width(c, is_cjk))
167		} else {
168		// U+007F to U+00A0 (exclusive) are control codes
169	0	None
170		}
171	0	}
172
173		/// Methods for determining displayed width of Unicode strings.
174		pub trait UnicodeWidthStr {
175		/// Returns the string's displayed width in columns.
176		///
177		/// This function treats characters in the Ambiguous category according
178		/// to [Unicode Standard Annex #11](http://www.unicode.org/reports/tr11/)
179		/// as 1 column wide. This is consistent with the recommendations for
180		/// non-CJK contexts, or when the context cannot be reliably determined.
181		fn width(&self) -> usize;
182
183		/// Returns the string's displayed width in columns.
184		///
185		/// This function treats characters in the Ambiguous category according
186		/// to [Unicode Standard Annex #11](http://www.unicode.org/reports/tr11/)
187		/// as 2 column wide. This is consistent with the recommendations for
188		/// CJK contexts.
189		fn width_cjk(&self) -> usize;
190		}
191
192		impl UnicodeWidthStr for str {
193		#[inline]
194	0	fn width(&self) -> usize {
195	0	str_width(self, false)
196	0	} Unexecuted instantiation: <str as unicode_width::UnicodeWidthStr>::width Unexecuted instantiation: <str as unicode_width::UnicodeWidthStr>::width
197
198		#[inline]
199	0	fn width_cjk(&self) -> usize {
200	0	str_width(self, true)
201	0	}
202		}
203
204		#[derive(Clone, Copy, Debug, Default, PartialEq, Eq)]
205		enum NextCharInfo {
206		#[default]
207		Default,
208		/// `'\n'`
209		LineFeed,
210		/// '\u{0338}'
211		/// For preserving canonical equivalence with CJK
212		CombiningLongSolidusOverlay,
213		/// `'\u{A4FC}'..='\u{A4FD}'`
214		/// <https://www.unicode.org/versions/Unicode15.0.0/ch18.pdf#G42078>
215		TrailingLisuToneLetter,
216		/// `'\u{FE0E}'`
217		Vs15,
218		/// `'\u{FE0F}'`
219		Vs16,
220		}
221
222	0	fn str_width(s: &str, is_cjk: bool) -> usize {
223	0	s.chars()
224	0	.rfold((0, NextCharInfo::Default), \|(sum, next_info), c\| {
225	0	let (add, info) = width_in_str(c, is_cjk, next_info);
226	0	(sum + add, info)
227	0	})
228	0	.0
229	0	}
230
231		/// Returns the [UAX #11](https://www.unicode.org/reports/tr11/) based width of `c`.
232		/// If `is_cjk == true`, ambiguous width characters are treated as double width; otherwise,
233		/// they're treated as single width.
234		#[inline]
235	0	fn width_in_str(c: char, is_cjk: bool, next_info: NextCharInfo) -> (usize, NextCharInfo) {
236	0	if (is_cjk
237	0	&& next_info == NextCharInfo::CombiningLongSolidusOverlay
238	0	&& matches!(c, '<' \| '=' \| '>'))
239	0	\|\| (next_info == NextCharInfo::Vs16 && cw::starts_emoji_presentation_seq(c))
240		{
241	0	(2, NextCharInfo::Default)
242	0	} else if c <= '\u{A0}' {
243	0	match c {
244	0	'\n' => (1, NextCharInfo::LineFeed),
245	0	'\r' if next_info == NextCharInfo::LineFeed => (0, NextCharInfo::Default),
246	0	_ => (1, NextCharInfo::Default),
247		}
248		} else {
249	0	match (c, next_info) {
250	0	('\u{A4F8}'..='\u{A4FB}', NextCharInfo::TrailingLisuToneLetter) => {
251	0	(0, NextCharInfo::Default)
252		}
253	0	('\u{0338}', _) => (0, NextCharInfo::CombiningLongSolidusOverlay),
254	0	('\u{A4FC}'..='\u{A4FD}', _) => (1, NextCharInfo::TrailingLisuToneLetter),
255	0	('\u{FE0E}', _) => (0, NextCharInfo::Vs15),
256	0	('\u{FE0F}', _) => (0, NextCharInfo::Vs16),
257		(_, NextCharInfo::Vs15)
258	0	if !is_cjk && cw::starts_non_ideographic_text_presentation_seq(c) =>
259	0	{
260	0	(1, NextCharInfo::Default)
261		}
262	0	_ => (cw::lookup_width(c, is_cjk), NextCharInfo::Default),
263		}
264		}
265	0	}