xutf 1.3.0

Permissive UTF-8/16/32 transcoding, comparison and BOM detection with SIMD ASCII fast paths
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232

use xutf::{
	Utf8, Utf16, Utf32, graphemes_str, width_str, wrap, wrap_measured, wrap_measured_str, wrap_str,
};

fn lines(input: &str, width: usize) -> Vec<&str> {
	wrap_str(input, width).collect()
}

#[test]
fn reference_behavior() {
	assert_eq!(lines("hello world", 5), ["hello", "world"]);
	assert_eq!(lines("foo barbazqux", 6), ["foo", "barbaz", "qux"]);
	assert_eq!(lines("   longword", 4), ["", "long", "word"]);
	assert_eq!(lines("a b c", 1), ["a", "b", "c"]);
	assert_eq!(lines("aa   bb", 2), ["aa", "bb"]);
	assert_eq!(lines("ab   ", 10), ["ab"]);
	assert_eq!(lines("", 10), [""]);
	assert_eq!(lines("a\n\nb", 10), ["a", "", "b"]);
	assert_eq!(lines("a\n", 10), ["a", ""]);
	assert_eq!(lines("\n", 10), ["", ""]);
	assert_eq!(lines("a\r\nb", 10), ["a", "b"]);
	assert_eq!(lines("a\rb", 10), ["a", "b"]);
	assert_eq!(lines("  hi", 10), ["  hi"]);
	assert_eq!(lines("a\n  b", 10), ["a", "  b"]);
	assert_eq!(lines("a\tb", 1), ["a\t", "b"]);
}

#[test]
fn clusters_are_never_split() {
	assert_eq!(lines("πŸ‘¨β€πŸ‘©β€πŸ‘§", 1), ["πŸ‘¨β€πŸ‘©β€πŸ‘§"]);
	assert_eq!(lines("η•Œη•Œη•Œ", 4), ["η•Œη•Œ", "η•Œ"]);
	assert_eq!(lines("a\u{301}b", 1), ["a\u{301}", "b"]);
	assert_eq!(lines("πŸ‡ΊπŸ‡ΈπŸ‡¨πŸ‡¦", 2), ["πŸ‡ΊπŸ‡Έ", "πŸ‡¨πŸ‡¦"]);
}

#[test]
fn fuzzed_lines_obey_width_and_content_invariants() {
	const PIECES: &[&str] =
		&["a", "b", " ", " ", "\n", "\r\n", "η•Œ", "θͺž", "πŸ˜€", "πŸ‘¨β€πŸ‘©β€πŸ‘§", "e\u{301}", "πŸ‡ΊπŸ‡Έ"];
	let mut state = 0x8e6d_5a43_91c2_7b0du64;

	for case in 0..1_000 {
		state = state.wrapping_mul(6364136223846793005).wrapping_add(1);
		let piece_count = 1 + (state as usize % 80);
		let mut input = String::new();
		for _ in 0..piece_count {
			state = state.wrapping_mul(6364136223846793005).wrapping_add(1);
			input.push_str(PIECES[(state as usize) % PIECES.len()]);
		}
		state = state.wrapping_mul(6364136223846793005).wrapping_add(1);
		let max_width = state as usize % 25;
		let output: Vec<&str> = wrap_str(&input, max_width).collect();

		assert!(!output.is_empty(), "case {case}");
		for (index, line) in output.iter().enumerate() {
			assert!(
				width_str(line) <= max_width || graphemes_str(line).count() == 1,
				"case {case}, width {max_width}, line {line:?}"
			);
			assert!(!line.ends_with(' '), "case {case}, line {line:?}");
			if index > 0 && line.starts_with(' ') {
				assert!(
					input.contains("\n ") || input.contains("\r "),
					"only hard breaks may preserve leading spaces: case {case}, line {line:?}"
				);
			}
		}

		let source: String = input
			.chars()
			.filter(|c| !matches!(c, ' ' | '\n' | '\r'))
			.collect();
		let wrapped: String = output
			.iter()
			.flat_map(|line| line.chars())
			.filter(|c| *c != ' ')
			.collect();
		assert_eq!(wrapped, source, "case {case}");
	}
}

#[test]
fn zero_width_and_long_words_always_progress() {
	assert_eq!(lines("aaa", 0), ["a", "a", "a"]);

	let input = "a".repeat(10_000);
	let mut count = 0;
	for line in wrap_str(&input, 8) {
		assert_eq!(line.len(), 8);
		count += 1;
		assert!(count <= 1_250);
	}
	assert_eq!(count, 1_250);
}

fn assert_encoding_parity(input: &str, max_width: usize) {
	let expected: Vec<String> = wrap_str(input, max_width).map(str::to_owned).collect();

	let utf16: Vec<u16> = input.encode_utf16().collect();
	let native16: Vec<String> = wrap::<Utf16<false>>(&utf16, max_width)
		.map(|line| String::from_utf16(line).unwrap())
		.collect();
	assert_eq!(native16, expected);

	let utf32: Vec<u32> = input.chars().map(u32::from).collect();
	let native32: Vec<String> = wrap::<Utf32<false>>(&utf32, max_width)
		.map(|line| line.iter().map(|&cp| char::from_u32(cp).unwrap()).collect())
		.collect();
	assert_eq!(native32, expected);

	let foreign16: Vec<u16> = utf16.iter().map(|unit| unit.swap_bytes()).collect();
	let swapped16: Vec<String> = wrap::<Utf16<true>>(&foreign16, max_width)
		.map(|line| {
			let native: Vec<u16> = line.iter().map(|unit| unit.swap_bytes()).collect();
			String::from_utf16(&native).unwrap()
		})
		.collect();
	assert_eq!(swapped16, expected);
}

#[test]
fn encodings_have_identical_wrap_boundaries() {
	let cases = [
		("hello world", 5),
		("foo barbazqux", 6),
		("   longword", 4),
		("a b c", 1),
		("aa   bb", 2),
		("ab   ", 10),
		("", 10),
		("a\n\nb", 10),
		("a\n", 10),
		("\n", 10),
		("a\r\nb", 10),
		("a\rb", 10),
		("  hi", 10),
		("a\n  b", 10),
		("πŸ‘¨β€πŸ‘©β€πŸ‘§", 1),
		("η•Œη•Œη•Œ", 4),
		("a\u{301}b", 1),
		("πŸ‡ΊπŸ‡ΈπŸ‡¨πŸ‡¦", 2),
	];
	for (input, max_width) in cases {
		assert_encoding_parity(input, max_width);
	}
}

#[test]
fn malformed_sequences_are_never_split_by_space_trimming() {
	// 0x96 is a two-byte lead, so permissive decoding swallows the following
	// 0x20: it is a sequence tail, not a trailing space, and the cluster must
	// survive intact.
	let input = [0x86u8, 0x25, 0x96, 0x20, 0xec];
	let lines: Vec<&[u8]> = wrap::<Utf8>(&input, 0).collect();
	assert_eq!(lines, [&input[..4], &input[4..]]);
	assert_eq!(lines.concat(), input);
}

#[test]
fn simple_ascii_prose_matches_textwrap_shape() {
	// This comparison is deliberately narrow: textwrap has different word
	// separators and penalties, but coincides for single-space ASCII prose
	// with no overlong words.
	let text = "the quick brown fox jumps over the lazy dog";
	for width in [10, 20] {
		let actual: Vec<&str> = wrap_str(text, width).collect();
		let baseline = textwrap::wrap(text, width);
		let expected: Vec<&str> = baseline.iter().map(|line| line.as_ref()).collect();
		assert_eq!(actual, expected);
	}
}

#[test]
fn measured_lines_match_plain_wrap_and_report_source_positions() {
	let corpus = [
		"",
		"plain ascii paragraph with several words",
		"  leading and trailing spaces   ",
		"aa   bb      cc",
		"a\r\nb\rc\n\nd",
		"η•Œ η•Œη•Œ ι•·γ„ε˜θͺž",
		"πŸ‘¨β€πŸ‘©β€πŸ‘§ family πŸ‡ΊπŸ‡Έ flag",
		"supercalifragilisticexpialidocious",
		"\n",
		"a\n",
	];

	for input in corpus {
		for max_width in [0, 1, 4, 10, 80] {
			let measured: Vec<_> = wrap_measured_str(input, max_width).collect();
			let plain: Vec<_> = wrap_str(input, max_width).collect();
			assert_eq!(measured.len(), plain.len(), "{input:?}, width {max_width}");

			let mut previous_at = None;
			for (line, expected) in measured.iter().zip(&plain) {
				assert_eq!(line.as_str(), *expected, "{input:?}, width {max_width}");
				assert_eq!(
					&input[line.at..line.at + line.as_str().len()],
					line.as_str(),
					"{input:?}, width {max_width}",
				);
				assert_eq!(line.width, width_str(line.as_str()));
				if let Some(at) = previous_at {
					assert!(line.at > at, "{input:?}, width {max_width}");
				}
				previous_at = Some(line.at);
			}
		}
	}
}

#[test]
fn measured_utf16_offsets_are_code_unit_offsets() {
	let input: Vec<u16> = "a πŸ˜€ b\nη•Œη•Œ z".encode_utf16().collect();
	let lines: Vec<_> = wrap_measured::<Utf16<false>>(&input, 4).collect();
	let actual: Vec<_> = lines
		.iter()
		.map(|line| {
			assert_eq!(&input[line.at..line.at + line.units.len()], line.units);
			let text = String::from_utf16(line.units).unwrap();
			assert_eq!(line.width, width_str(&text));
			(line.at, line.width, text)
		})
		.collect();

	assert_eq!(actual, [
		(0, 4, "a πŸ˜€".to_owned()),
		(5, 1, "b".to_owned()),
		(7, 4, "η•Œη•Œ".to_owned()),
		(10, 1, "z".to_owned()),
	],);
}