diff options
Diffstat (limited to 'crates/ra_ide/src/line_index.rs')
-rw-r--r-- | crates/ra_ide/src/line_index.rs | 283 |
1 files changed, 0 insertions, 283 deletions
diff --git a/crates/ra_ide/src/line_index.rs b/crates/ra_ide/src/line_index.rs deleted file mode 100644 index 710890d27..000000000 --- a/crates/ra_ide/src/line_index.rs +++ /dev/null | |||
@@ -1,283 +0,0 @@ | |||
1 | //! FIXME: write short doc here | ||
2 | |||
3 | use crate::TextUnit; | ||
4 | use rustc_hash::FxHashMap; | ||
5 | use superslice::Ext; | ||
6 | |||
7 | #[derive(Clone, Debug, PartialEq, Eq)] | ||
8 | pub struct LineIndex { | ||
9 | pub(crate) newlines: Vec<TextUnit>, | ||
10 | pub(crate) utf16_lines: FxHashMap<u32, Vec<Utf16Char>>, | ||
11 | } | ||
12 | |||
13 | #[derive(Clone, Copy, Debug, PartialEq, Eq, Hash)] | ||
14 | pub struct LineCol { | ||
15 | /// Zero-based | ||
16 | pub line: u32, | ||
17 | /// Zero-based | ||
18 | pub col_utf16: u32, | ||
19 | } | ||
20 | |||
21 | #[derive(Clone, Debug, Hash, PartialEq, Eq)] | ||
22 | pub(crate) struct Utf16Char { | ||
23 | pub(crate) start: TextUnit, | ||
24 | pub(crate) end: TextUnit, | ||
25 | } | ||
26 | |||
27 | impl Utf16Char { | ||
28 | fn len(&self) -> TextUnit { | ||
29 | self.end - self.start | ||
30 | } | ||
31 | } | ||
32 | |||
33 | impl LineIndex { | ||
34 | pub fn new(text: &str) -> LineIndex { | ||
35 | let mut utf16_lines = FxHashMap::default(); | ||
36 | let mut utf16_chars = Vec::new(); | ||
37 | |||
38 | let mut newlines = vec![0.into()]; | ||
39 | let mut curr_row = 0.into(); | ||
40 | let mut curr_col = 0.into(); | ||
41 | let mut line = 0; | ||
42 | for c in text.chars() { | ||
43 | curr_row += TextUnit::of_char(c); | ||
44 | if c == '\n' { | ||
45 | newlines.push(curr_row); | ||
46 | |||
47 | // Save any utf-16 characters seen in the previous line | ||
48 | if !utf16_chars.is_empty() { | ||
49 | utf16_lines.insert(line, utf16_chars); | ||
50 | utf16_chars = Vec::new(); | ||
51 | } | ||
52 | |||
53 | // Prepare for processing the next line | ||
54 | curr_col = 0.into(); | ||
55 | line += 1; | ||
56 | continue; | ||
57 | } | ||
58 | |||
59 | let char_len = TextUnit::of_char(c); | ||
60 | if char_len.to_usize() > 1 { | ||
61 | utf16_chars.push(Utf16Char { start: curr_col, end: curr_col + char_len }); | ||
62 | } | ||
63 | |||
64 | curr_col += char_len; | ||
65 | } | ||
66 | |||
67 | // Save any utf-16 characters seen in the last line | ||
68 | if !utf16_chars.is_empty() { | ||
69 | utf16_lines.insert(line, utf16_chars); | ||
70 | } | ||
71 | |||
72 | LineIndex { newlines, utf16_lines } | ||
73 | } | ||
74 | |||
75 | pub fn line_col(&self, offset: TextUnit) -> LineCol { | ||
76 | let line = self.newlines.upper_bound(&offset) - 1; | ||
77 | let line_start_offset = self.newlines[line]; | ||
78 | let col = offset - line_start_offset; | ||
79 | |||
80 | LineCol { line: line as u32, col_utf16: self.utf8_to_utf16_col(line as u32, col) as u32 } | ||
81 | } | ||
82 | |||
83 | pub fn offset(&self, line_col: LineCol) -> TextUnit { | ||
84 | //FIXME: return Result | ||
85 | let col = self.utf16_to_utf8_col(line_col.line, line_col.col_utf16); | ||
86 | self.newlines[line_col.line as usize] + col | ||
87 | } | ||
88 | |||
89 | fn utf8_to_utf16_col(&self, line: u32, mut col: TextUnit) -> usize { | ||
90 | if let Some(utf16_chars) = self.utf16_lines.get(&line) { | ||
91 | let mut correction = TextUnit::from_usize(0); | ||
92 | for c in utf16_chars { | ||
93 | if col >= c.end { | ||
94 | correction += c.len() - TextUnit::from_usize(1); | ||
95 | } else { | ||
96 | // From here on, all utf16 characters come *after* the character we are mapping, | ||
97 | // so we don't need to take them into account | ||
98 | break; | ||
99 | } | ||
100 | } | ||
101 | |||
102 | col -= correction; | ||
103 | } | ||
104 | |||
105 | col.to_usize() | ||
106 | } | ||
107 | |||
108 | fn utf16_to_utf8_col(&self, line: u32, col: u32) -> TextUnit { | ||
109 | let mut col: TextUnit = col.into(); | ||
110 | if let Some(utf16_chars) = self.utf16_lines.get(&line) { | ||
111 | for c in utf16_chars { | ||
112 | if col >= c.start { | ||
113 | col += c.len() - TextUnit::from_usize(1); | ||
114 | } else { | ||
115 | // From here on, all utf16 characters come *after* the character we are mapping, | ||
116 | // so we don't need to take them into account | ||
117 | break; | ||
118 | } | ||
119 | } | ||
120 | } | ||
121 | |||
122 | col | ||
123 | } | ||
124 | } | ||
125 | |||
126 | #[cfg(test)] | ||
127 | /// Simple reference implementation to use in proptests | ||
128 | pub fn to_line_col(text: &str, offset: TextUnit) -> LineCol { | ||
129 | let mut res = LineCol { line: 0, col_utf16: 0 }; | ||
130 | for (i, c) in text.char_indices() { | ||
131 | if i + c.len_utf8() > offset.to_usize() { | ||
132 | // if it's an invalid offset, inside a multibyte char | ||
133 | // return as if it was at the start of the char | ||
134 | break; | ||
135 | } | ||
136 | if c == '\n' { | ||
137 | res.line += 1; | ||
138 | res.col_utf16 = 0; | ||
139 | } else { | ||
140 | res.col_utf16 += 1; | ||
141 | } | ||
142 | } | ||
143 | res | ||
144 | } | ||
145 | |||
146 | #[cfg(test)] | ||
147 | mod test_line_index { | ||
148 | use super::*; | ||
149 | use proptest::{prelude::*, proptest}; | ||
150 | use ra_text_edit::test_utils::{arb_offset, arb_text}; | ||
151 | |||
152 | #[test] | ||
153 | fn test_line_index() { | ||
154 | let text = "hello\nworld"; | ||
155 | let index = LineIndex::new(text); | ||
156 | assert_eq!(index.line_col(0.into()), LineCol { line: 0, col_utf16: 0 }); | ||
157 | assert_eq!(index.line_col(1.into()), LineCol { line: 0, col_utf16: 1 }); | ||
158 | assert_eq!(index.line_col(5.into()), LineCol { line: 0, col_utf16: 5 }); | ||
159 | assert_eq!(index.line_col(6.into()), LineCol { line: 1, col_utf16: 0 }); | ||
160 | assert_eq!(index.line_col(7.into()), LineCol { line: 1, col_utf16: 1 }); | ||
161 | assert_eq!(index.line_col(8.into()), LineCol { line: 1, col_utf16: 2 }); | ||
162 | assert_eq!(index.line_col(10.into()), LineCol { line: 1, col_utf16: 4 }); | ||
163 | assert_eq!(index.line_col(11.into()), LineCol { line: 1, col_utf16: 5 }); | ||
164 | assert_eq!(index.line_col(12.into()), LineCol { line: 1, col_utf16: 6 }); | ||
165 | |||
166 | let text = "\nhello\nworld"; | ||
167 | let index = LineIndex::new(text); | ||
168 | assert_eq!(index.line_col(0.into()), LineCol { line: 0, col_utf16: 0 }); | ||
169 | assert_eq!(index.line_col(1.into()), LineCol { line: 1, col_utf16: 0 }); | ||
170 | assert_eq!(index.line_col(2.into()), LineCol { line: 1, col_utf16: 1 }); | ||
171 | assert_eq!(index.line_col(6.into()), LineCol { line: 1, col_utf16: 5 }); | ||
172 | assert_eq!(index.line_col(7.into()), LineCol { line: 2, col_utf16: 0 }); | ||
173 | } | ||
174 | |||
175 | fn arb_text_with_offset() -> BoxedStrategy<(TextUnit, String)> { | ||
176 | arb_text().prop_flat_map(|text| (arb_offset(&text), Just(text))).boxed() | ||
177 | } | ||
178 | |||
179 | fn to_line_col(text: &str, offset: TextUnit) -> LineCol { | ||
180 | let mut res = LineCol { line: 0, col_utf16: 0 }; | ||
181 | for (i, c) in text.char_indices() { | ||
182 | if i + c.len_utf8() > offset.to_usize() { | ||
183 | // if it's an invalid offset, inside a multibyte char | ||
184 | // return as if it was at the start of the char | ||
185 | break; | ||
186 | } | ||
187 | if c == '\n' { | ||
188 | res.line += 1; | ||
189 | res.col_utf16 = 0; | ||
190 | } else { | ||
191 | res.col_utf16 += 1; | ||
192 | } | ||
193 | } | ||
194 | res | ||
195 | } | ||
196 | |||
197 | proptest! { | ||
198 | #[test] | ||
199 | fn test_line_index_proptest((offset, text) in arb_text_with_offset()) { | ||
200 | let expected = to_line_col(&text, offset); | ||
201 | let line_index = LineIndex::new(&text); | ||
202 | let actual = line_index.line_col(offset); | ||
203 | |||
204 | assert_eq!(actual, expected); | ||
205 | } | ||
206 | } | ||
207 | } | ||
208 | |||
209 | #[cfg(test)] | ||
210 | mod test_utf8_utf16_conv { | ||
211 | use super::*; | ||
212 | |||
213 | #[test] | ||
214 | fn test_char_len() { | ||
215 | assert_eq!('メ'.len_utf8(), 3); | ||
216 | assert_eq!('メ'.len_utf16(), 1); | ||
217 | } | ||
218 | |||
219 | #[test] | ||
220 | fn test_empty_index() { | ||
221 | let col_index = LineIndex::new( | ||
222 | " | ||
223 | const C: char = 'x'; | ||
224 | ", | ||
225 | ); | ||
226 | assert_eq!(col_index.utf16_lines.len(), 0); | ||
227 | } | ||
228 | |||
229 | #[test] | ||
230 | fn test_single_char() { | ||
231 | let col_index = LineIndex::new( | ||
232 | " | ||
233 | const C: char = 'メ'; | ||
234 | ", | ||
235 | ); | ||
236 | |||
237 | assert_eq!(col_index.utf16_lines.len(), 1); | ||
238 | assert_eq!(col_index.utf16_lines[&1].len(), 1); | ||
239 | assert_eq!(col_index.utf16_lines[&1][0], Utf16Char { start: 17.into(), end: 20.into() }); | ||
240 | |||
241 | // UTF-8 to UTF-16, no changes | ||
242 | assert_eq!(col_index.utf8_to_utf16_col(1, 15.into()), 15); | ||
243 | |||
244 | // UTF-8 to UTF-16 | ||
245 | assert_eq!(col_index.utf8_to_utf16_col(1, 22.into()), 20); | ||
246 | |||
247 | // UTF-16 to UTF-8, no changes | ||
248 | assert_eq!(col_index.utf16_to_utf8_col(1, 15), TextUnit::from(15)); | ||
249 | |||
250 | // UTF-16 to UTF-8 | ||
251 | assert_eq!(col_index.utf16_to_utf8_col(1, 19), TextUnit::from(21)); | ||
252 | } | ||
253 | |||
254 | #[test] | ||
255 | fn test_string() { | ||
256 | let col_index = LineIndex::new( | ||
257 | " | ||
258 | const C: char = \"メ メ\"; | ||
259 | ", | ||
260 | ); | ||
261 | |||
262 | assert_eq!(col_index.utf16_lines.len(), 1); | ||
263 | assert_eq!(col_index.utf16_lines[&1].len(), 2); | ||
264 | assert_eq!(col_index.utf16_lines[&1][0], Utf16Char { start: 17.into(), end: 20.into() }); | ||
265 | assert_eq!(col_index.utf16_lines[&1][1], Utf16Char { start: 21.into(), end: 24.into() }); | ||
266 | |||
267 | // UTF-8 to UTF-16 | ||
268 | assert_eq!(col_index.utf8_to_utf16_col(1, 15.into()), 15); | ||
269 | |||
270 | assert_eq!(col_index.utf8_to_utf16_col(1, 21.into()), 19); | ||
271 | assert_eq!(col_index.utf8_to_utf16_col(1, 25.into()), 21); | ||
272 | |||
273 | assert!(col_index.utf8_to_utf16_col(2, 15.into()) == 15); | ||
274 | |||
275 | // UTF-16 to UTF-8 | ||
276 | assert_eq!(col_index.utf16_to_utf8_col(1, 15), TextUnit::from_usize(15)); | ||
277 | |||
278 | assert_eq!(col_index.utf16_to_utf8_col(1, 18), TextUnit::from_usize(20)); | ||
279 | assert_eq!(col_index.utf16_to_utf8_col(1, 19), TextUnit::from_usize(23)); | ||
280 | |||
281 | assert_eq!(col_index.utf16_to_utf8_col(2, 15), TextUnit::from_usize(15)); | ||
282 | } | ||
283 | } | ||