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
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
|
//! a fuzzy pattern matcher for filename filtering / sorting.
//! It's not meant for file contents but for small strings (less than 1000 chars)
//! such as file names.
use {
super::NameMatch,
secular,
smallvec::{smallvec, SmallVec},
std::fmt::{self, Write},
};
type CandChars = SmallVec<[char; 32]>;
// weights used in match score computing
const BONUS_MATCH: i32 = 50_000;
const BONUS_EXACT: i32 = 1_000;
const BONUS_START: i32 = 10;
const BONUS_START_WORD: i32 = 5;
const BONUS_CANDIDATE_LENGTH: i32 = -1; // per char
const BONUS_MATCH_LENGTH: i32 = -10; // per char of length of the match
const BONUS_NB_HOLES: i32 = -30; // there's also a max on that number
const BONUS_SINGLED_CHAR: i32 = -15; // when there's a char, neither first not last, isolated
/// A pattern for fuzzy matching
#[derive(Debug, Clone)]
pub struct FuzzyPattern {
chars: Box<[char]>, // secularized characters
max_nb_holes: usize,
}
impl fmt::Display for FuzzyPattern {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
for &c in self.chars.iter() {
f.write_char(c)?
}
Ok(())
}
}
enum MatchSearchResult {
Perfect(NameMatch), // no need to test other positions
Some(NameMatch),
None,
}
fn is_word_separator(c: char) -> bool {
matches!(c, '_' | ' ' | '-')
}
impl FuzzyPattern {
/// build a pattern which will later be usable for fuzzy search.
/// A pattern should be reused
pub fn from(pat: &str) -> Self {
let chars = secular::normalized_lower_lay_string(pat)
.chars()
.collect::<Vec<char>>()
.into_boxed_slice();
let max_nb_holes = match chars.len() {
1 => 0,
2 => 1,
3 => 2,
4 => 2,
5 => 2,
6 => 3,
7 => 3,
8 => 4,
_ => chars.len() * 4 / 7,
};
FuzzyPattern {
chars,
max_nb_holes,
}
}
/// an "empty" pattern is one which accepts everything because
/// it has no discriminant
pub fn is_empty(&self) -> bool {
self.chars.is_empty()
}
fn tight_match_from_index(
&self,
cand_chars: &CandChars,
start_idx: usize, // start index in candidate, in chars
) -> MatchSearchResult {
let mut pos = smallvec![0; self.chars.len()]; // positions of matching chars in candidate
let mut cand_idx = start_idx;
let mut pat_idx = 0; // index both in self.chars and pos
let mut in_hole = false;
loop {
if cand_chars[cand_idx] == self.chars[pat_idx] {
pos[pat_idx] = cand_idx;
if in_hole {
// We're no more in a hole.
// Let's look if we can bring back the chars before the hole
let mut rev_idx = 1;
loop {
if pat_idx < rev_idx {
break;
}
if cand_chars[cand_idx-rev_idx] == self.chars[pat_idx-rev_idx] {
// we move the pos forward
pos[pat_idx-rev_idx] = cand_idx-rev_idx;
} else {
break;
}
rev_idx += 1;
}
in_hole = false;
}
pat_idx += 1;
if pat_idx == self.chars.len() {
break; // match, finished
}
} else {
// there's a hole
if cand_chars.len() - cand_idx <= self.chars.len() - pat_idx {
return MatchSearchResult::None;
}
in_hole = true;
}
cand_idx += 1;
}
let mut nb_holes = 0;
let mut nb_singled_chars = 0;
for idx in 1..pos.len() {
if pos[idx] > 1 + pos[idx-1] {
nb_holes += 1;
if idx > 1 && pos[idx-1] > 1 + pos[idx-2] {
// we improve a simple case: the one of a singleton which was created
// by pushing forward a char
if cand_chars[pos[idx-2]+1] == cand_chars[pos[idx-1]] {
// in some cases we're really removing another singletons but
// let's forget this
pos[idx-1] = pos[idx-2]+1;
nb_holes -= 1;
} else {
nb_singled_chars += 1;
}
}
}
}
if nb_holes > self.max_nb_holes {
return MatchSearchResult::None;
}
let match_len = 1 + cand_idx - pos[0];
let mut score = BONUS_MATCH;
score += BONUS_CANDIDATE_LENGTH * (cand_chars.len() as i32);
score += BONUS_SINGLED_CHAR * nb_singled_chars;
score += BONUS_NB_HOLES * (nb_holes as i32);
score += match_len as i32 * BONUS_MATCH_LENGTH;
if pos[0] == 0 {
score += BONUS_START + BONUS_START_WORD;
if cand_chars.len() == self.chars.len() {
score += BONUS_EXACT;
return MatchSearchResult::Perfect(NameMatch { score, pos });
}
} else {
let previous = cand_chars[pos[0] - 1];
if is_word_separator(previous) {
score += BONUS_START_WORD;
if cand_chars.len() - pos[0] == self.chars.len() {
return MatchSearchResult::Perfect(NameMatch { score, pos });
}
}
}
MatchSearchResult::Some(NameMatch { score, pos })
}
/// return a match if the pattern can be found in the candidate string.
/// The algorithm tries to return the best one. For example if you search
/// "abc" in "ababca-abc", the returned match would be at the end.
pub fn find(&self, candidate: &str) -> Option<NameMatch> {
if candidate.len() < self.chars.len() {
return None;
}
let mut cand_chars: CandChars = SmallVec::with_capacity(candidate.len());
cand_chars.extend(candidate.chars().map(secular::lower_lay_char));
if cand_chars.len() < self.chars.len() {
return None;
}
let mut best_score = 0;
let mut best_match: Option<NameMatch> = None;
let n = cand_chars.len() + 1 - self.chars.len();
for start_idx in 0..n {
if cand_chars[start_idx] == self.chars[0] {
match self.tight_match_from_index(&cand_chars, start_idx) {
MatchSearchResult::Perfect(m) => {
return Some(m);
}
MatchSearchResult::Some(m) => {
if m.score > best_score {
best_score = m.score;
best_match = Some(m);
}
// we could make start_idx jump to pos[0] here
// but it doesn't improve the perfs (it's rare
// anyway to have pos[0] much greater than the
// start of the search)
}
_ => {}
}
}
}
best_match
}
|