aboutsummaryrefslogtreecommitdiff
path: root/crates/ra_ide_db/src/line_index.rs
blob: c7c744fce1ef8db144f1d07ab1bd6cb03b99539d (plain)
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
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
//! `LineIndex` maps flat `TextSize` offsets into `(Line, Column)`
//! representation.
use std::iter;

use ra_syntax::{TextRange, TextSize};
use rustc_hash::FxHashMap;
use superslice::Ext;

#[derive(Clone, Debug, PartialEq, Eq)]
pub struct LineIndex {
    /// Offset the the beginning of each line, zero-based
    pub(crate) newlines: Vec<TextSize>,
    /// List of non-ASCII characters on each line
    pub(crate) utf16_lines: FxHashMap<u32, Vec<Utf16Char>>,
}

#[derive(Clone, Copy, Debug, PartialEq, Eq, Hash)]
pub struct LineCol {
    /// Zero-based
    pub line: u32,
    /// Zero-based
    pub col_utf16: u32,
}

#[derive(Clone, Debug, Hash, PartialEq, Eq)]
pub(crate) struct Utf16Char {
    /// Start offset of a character inside a line, zero-based
    pub(crate) start: TextSize,
    /// End offset of a character inside a line, zero-based
    pub(crate) end: TextSize,
}

impl Utf16Char {
    /// Returns the length in 8-bit UTF-8 code units.
    fn len(&self) -> TextSize {
        self.end - self.start
    }

    /// Returns the length in 16-bit UTF-16 code units.
    fn len_utf16(&self) -> usize {
        if self.len() == TextSize::from(4) {
            2
        } else {
            1
        }
    }
}

impl LineIndex {
    pub fn new(text: &str) -> LineIndex {
        let mut utf16_lines = FxHashMap::default();
        let mut utf16_chars = Vec::new();

        let mut newlines = vec![0.into()];
        let mut curr_row = 0.into();
        let mut curr_col = 0.into();
        let mut line = 0;
        for c in text.chars() {
            let c_len = TextSize::of(c);
            curr_row += c_len;
            if c == '\n' {
                newlines.push(curr_row);

                // Save any utf-16 characters seen in the previous line
                if !utf16_chars.is_empty() {
                    utf16_lines.insert(line, utf16_chars);
                    utf16_chars = Vec::new();
                }

                // Prepare for processing the next line
                curr_col = 0.into();
                line += 1;
                continue;
            }

            if !c.is_ascii() {
                utf16_chars.push(Utf16Char { start: curr_col, end: curr_col + c_len });
            }

            curr_col += c_len;
        }

        // Save any utf-16 characters seen in the last line
        if !utf16_chars.is_empty() {
            utf16_lines.insert(line, utf16_chars);
        }

        LineIndex { newlines, utf16_lines }
    }

    pub fn line_col(&self, offset: TextSize) -> LineCol {
        let line = self.newlines.upper_bound(&offset) - 1;
        let line_start_offset = self.newlines[line];
        let col = offset - line_start_offset;

        LineCol { line: line as u32, col_utf16: self.utf8_to_utf16_col(line as u32, col) as u32 }
    }

    pub fn offset(&self, line_col: LineCol) -> TextSize {
        //FIXME: return Result
        let col = self.utf16_to_utf8_col(line_col.line, line_col.col_utf16);
        self.newlines[line_col.line as usize] + col
    }

    pub fn lines(&self, range: TextRange) -> impl Iterator<Item = TextRange> + '_ {
        let lo = self.newlines.lower_bound(&range.start());
        let hi = self.newlines.upper_bound(&range.end());
        let all = iter::once(range.start())
            .chain(self.newlines[lo..hi].iter().copied())
            .chain(iter::once(range.end()));

        all.clone()
            .zip(all.skip(1))
            .map(|(lo, hi)| TextRange::new(lo, hi))
            .filter(|it| !it.is_empty())
    }

    fn utf8_to_utf16_col(&self, line: u32, col: TextSize) -> usize {
        let mut res: usize = col.into();
        if let Some(utf16_chars) = self.utf16_lines.get(&line) {
            for c in utf16_chars {
                if c.end <= col {
                    res -= usize::from(c.len()) - c.len_utf16();
                } else {
                    // From here on, all utf16 characters come *after* the character we are mapping,
                    // so we don't need to take them into account
                    break;
                }
            }
        }
        res
    }

    fn utf16_to_utf8_col(&self, line: u32, mut col: u32) -> TextSize {
        if let Some(utf16_chars) = self.utf16_lines.get(&line) {
            for c in utf16_chars {
                if col > u32::from(c.start) {
                    col += u32::from(c.len()) - c.len_utf16() as u32;
                } else {
                    // From here on, all utf16 characters come *after* the character we are mapping,
                    // so we don't need to take them into account
                    break;
                }
            }
        }

        col.into()
    }
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn test_line_index() {
        let text = "hello\nworld";
        let index = LineIndex::new(text);
        assert_eq!(index.line_col(0.into()), LineCol { line: 0, col_utf16: 0 });
        assert_eq!(index.line_col(1.into()), LineCol { line: 0, col_utf16: 1 });
        assert_eq!(index.line_col(5.into()), LineCol { line: 0, col_utf16: 5 });
        assert_eq!(index.line_col(6.into()), LineCol { line: 1, col_utf16: 0 });
        assert_eq!(index.line_col(7.into()), LineCol { line: 1, col_utf16: 1 });
        assert_eq!(index.line_col(8.into()), LineCol { line: 1, col_utf16: 2 });
        assert_eq!(index.line_col(10.into()), LineCol { line: 1, col_utf16: 4 });
        assert_eq!(index.line_col(11.into()), LineCol { line: 1, col_utf16: 5 });
        assert_eq!(index.line_col(12.into()), LineCol { line: 1, col_utf16: 6 });

        let text = "\nhello\nworld";
        let index = LineIndex::new(text);
        assert_eq!(index.line_col(0.into()), LineCol { line: 0, col_utf16: 0 });
        assert_eq!(index.line_col(1.into()), LineCol { line: 1, col_utf16: 0 });
        assert_eq!(index.line_col(2.into()), LineCol { line: 1, col_utf16: 1 });
        assert_eq!(index.line_col(6.into()), LineCol { line: 1, col_utf16: 5 });
        assert_eq!(index.line_col(7.into()), LineCol { line: 2, col_utf16: 0 });
    }

    #[test]
    fn test_char_len() {
        assert_eq!('メ'.len_utf8(), 3);
        assert_eq!('メ'.len_utf16(), 1);
    }

    #[test]
    fn test_empty_index() {
        let col_index = LineIndex::new(
            "
const C: char = 'x';
",
        );
        assert_eq!(col_index.utf16_lines.len(), 0);
    }

    #[test]
    fn test_single_char() {
        let col_index = LineIndex::new(
            "
const C: char = 'メ';
",
        );

        assert_eq!(col_index.utf16_lines.len(), 1);
        assert_eq!(col_index.utf16_lines[&1].len(), 1);
        assert_eq!(col_index.utf16_lines[&1][0], Utf16Char { start: 17.into(), end: 20.into() });

        // UTF-8 to UTF-16, no changes
        assert_eq!(col_index.utf8_to_utf16_col(1, 15.into()), 15);

        // UTF-8 to UTF-16
        assert_eq!(col_index.utf8_to_utf16_col(1, 22.into()), 20);

        // UTF-16 to UTF-8, no changes
        assert_eq!(col_index.utf16_to_utf8_col(1, 15), TextSize::from(15));

        // UTF-16 to UTF-8
        assert_eq!(col_index.utf16_to_utf8_col(1, 19), TextSize::from(21));

        let col_index = LineIndex::new("a𐐏b");
        assert_eq!(col_index.utf16_to_utf8_col(0, 3), TextSize::from(5));
    }

    #[test]
    fn test_string() {
        let col_index = LineIndex::new(
            "
const C: char = \"メ メ\";
",
        );

        assert_eq!(col_index.utf16_lines.len(), 1);
        assert_eq!(col_index.utf16_lines[&1].len(), 2);
        assert_eq!(col_index.utf16_lines[&1][0], Utf16Char { start: 17.into(), end: 20.into() });
        assert_eq!(col_index.utf16_lines[&1][1], Utf16Char { start: 21.into(), end: 24.into() });

        // UTF-8 to UTF-16
        assert_eq!(col_index.utf8_to_utf16_col(1, 15.into()), 15);

        assert_eq!(col_index.utf8_to_utf16_col(1, 21.into()), 19);
        assert_eq!(col_index.utf8_to_utf16_col(1, 25.into()), 21);

        assert!(col_index.utf8_to_utf16_col(2, 15.into()) == 15);

        // UTF-16 to UTF-8
        assert_eq!(col_index.utf16_to_utf8_col(1, 15), TextSize::from(15));

        // メ UTF-8: 0xE3 0x83 0xA1, UTF-16: 0x30E1
        assert_eq!(col_index.utf16_to_utf8_col(1, 17), TextSize::from(17)); // first メ at 17..20
        assert_eq!(col_index.utf16_to_utf8_col(1, 18), TextSize::from(20)); // space
        assert_eq!(col_index.utf16_to_utf8_col(1, 19), TextSize::from(21)); // second メ at 21..24

        assert_eq!(col_index.utf16_to_utf8_col(2, 15), TextSize::from(15));
    }

    #[test]
    fn test_splitlines() {
        fn r(lo: u32, hi: u32) -> TextRange {
            TextRange::new(lo.into(), hi.into())
        }

        let text = "a\nbb\nccc\n";
        let line_index = LineIndex::new(text);

        let actual = line_index.lines(r(0, 9)).collect::<Vec<_>>();
        let expected = vec![r(0, 2), r(2, 5), r(5, 9)];
        assert_eq!(actual, expected);

        let text = "";
        let line_index = LineIndex::new(text);

        let actual = line_index.lines(r(0, 0)).collect::<Vec<_>>();
        let expected = vec![];
        assert_eq!(actual, expected);

        let text = "\n";
        let line_index = LineIndex::new(text);

        let actual = line_index.lines(r(0, 1)).collect::<Vec<_>>();
        let expected = vec![r(0, 1)];
        assert_eq!(actual, expected)
    }
}