-
Notifications
You must be signed in to change notification settings - Fork 7
/
Copy pathdat.go
652 lines (581 loc) · 13.5 KB
/
dat.go
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
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
package godat
import (
"fmt"
"sort"
"strings"
"unicode"
"unicode/utf8"
)
func assert(cond bool, err string) {
if cond == false {
panic(err)
}
}
func (gd *GoDat) build() (err error) {
for _, s := range gd.pats {
if err = gd.insertPattern(s); err != nil {
return
}
}
return
}
// dump godat
func (gd *GoDat) dump() {
if gd.name != "" {
fmt.Println("dat " + gd.name + ":")
}
fmt.Printf("options: nocase=%v\n", gd.nocase)
fmt.Printf("array length = %d, idles = %d\n", len(gd.base), gd.idles)
if len(gd.base) <= 256 {
for i := 0; i < len(gd.base); i++ {
fmt.Printf("GoDat array index %d: %d %d\n", i, gd.base[i], gd.check[i])
}
}
fmt.Println("aux:", len(gd.auxiliary))
if len(gd.auxiliary) <= 64 {
for k, v := range gd.auxiliary {
fmt.Printf(" %s(%v): %d\n", string(k), k, v)
}
fmt.Println("reverse aux:", gd.revAuxiliary)
}
fmt.Println("patterns:", len(gd.pats))
}
// 增加一个模式
func (gd *GoDat) add(pat string) error {
gd.pats = append(gd.pats, pat)
return nil
}
// 扩大数组长度
func (gd *GoDat) extend(l int) (err error) {
orgLen := len(gd.base)
length := orgLen
if l == 0 {
length = length * 2
} else {
for length <= l {
length = length * 2
}
}
if length > gd.maxLen {
err = fmt.Errorf("Array cannot exceed dat's maxLen: %d", gd.maxLen)
return
}
base := make([]int, length)
check := make([]int, length)
attrs := make([]uint, length)
copy(base, gd.base)
copy(check, gd.check)
copy(attrs, gd.attrs)
// 将base和check数组的新元素连接起来
for i := orgLen; i < length; i++ {
base[i] = -(i - 1)
check[i] = -(i + 1)
}
if check[0] == 0 {
// 原来已经没有空位置
check[0] = orgLen
base[orgLen] = -(length - 1)
check[length-1] = -orgLen
} else {
// 将扩展出的位置与原来的空闲位置连接起来
first := check[0]
last := -1 * base[first]
base[orgLen] = -last
check[last] = -orgLen
base[first] = -length + 1
check[length-1] = -first
}
//fmt.Println("Before extend:\nbase:", gd.base, "\ncheck:", gd.check)
gd.base = base
gd.check = check
gd.attrs = attrs
gd.idles += orgLen
//fmt.Println("After extend:\nbase:", gd.base, "\ncheck:", gd.check)
return nil
}
func (gd *GoDat) buildAux() {
// 字符序号从1开始
chs := 1
if gd.auxiliary == nil {
gd.auxiliary = make(map[rune]int)
}
if gd.revAuxiliary == nil {
gd.revAuxiliary = make(map[int]rune)
}
// 给每个字符标定序号
for _, pat := range gd.pats {
for _, ch := range pat {
if gd.nocase {
ch = unicode.ToLower(ch)
}
if _, ok := gd.auxiliary[ch]; !ok {
gd.auxiliary[ch] = chs
gd.revAuxiliary[chs] = ch
chs++
}
}
}
}
// 将base和check数组组成双向链表,借鉴小稳的C实现原理
func (gd *GoDat) initLink() {
// 数组位置0预留
gd.base[0] = 0
gd.check[0] = 1
// 数组从1到length-1的各个位置连接起来
length := len(gd.base)
for i := 1; i < length; i++ {
gd.base[i] = -(i - 1)
gd.check[i] = -(i + 1)
}
gd.base[1] = -(length - 1)
gd.check[length-1] = -1
gd.idles = length - 1
}
// 节点用完后, 调用addLink使该节点变成idle
func (gd *GoDat) addLink(s int) {
//fmt.Println("addLink:", s)
if gd.idles == 0 {
gd.check[0] = s
gd.base[s] = -s
gd.check[s] = -s
} else {
t := gd.check[0]
// 找到s的下一个空节点位置
if s > t {
//for t = -1 * gd.check[t]; t != gd.check[0] && t < s; {
// t = -1 * gd.check[t]
//}
for t = s; t < gd.maxLen; t++ {
if gd.check[t] < 0 {
break
}
}
if t == gd.maxLen {
t = gd.check[0]
}
}
gd.base[s] = gd.base[t]
gd.check[-1*gd.base[t]] = -1 * s
gd.base[t] = -1 * s
gd.check[s] = -1 * t
if s < gd.check[0] {
gd.check[0] = s
}
}
gd.idles++
}
// 使用节点前调用delLink删除节点
func (gd *GoDat) delLink(s int) {
if gd.idles <= 0 {
return
}
if gd.idles == 1 {
gd.check[0] = 0
} else {
// 是首节点的话, 修改check[0]指向下一个空闲的节点
if s == gd.check[0] {
gd.check[0] = -1 * gd.check[s]
}
//fmt.Println("delLink:", s, gd.base[s], gd.check[s])
//fmt.Println("before delLink: base:", gd.base)
//fmt.Println("before delLink: check:", gd.check)
gd.base[-1*gd.check[s]] = gd.base[s]
gd.check[-1*gd.base[s]] = gd.check[s]
//fmt.Println("after delLink: base:", gd.base)
//fmt.Println("after delLink: check:", gd.check)
}
gd.idles--
}
func (gd *GoDat) sort() {
if len(gd.pats) == 0 {
return
}
// 过滤空字符串, 去重
m := make(map[string]bool)
pats := make([]string, len(gd.pats))
cnt := 0
for i := 0; i < len(gd.pats); i++ {
pat := gd.pats[i]
if pat == "" {
continue
}
if gd.nocase { // 转换大小写
pat = strings.ToLower(pat)
}
if ok := m[pat]; !ok {
m[pat] = true
pats[cnt] = pat
cnt++
}
}
gd.pats = pats[0:cnt]
// 对pattern排序, 节省空间
sort.Sort(sort.StringSlice(gd.pats))
}
func (gd *GoDat) __find_pos(s, c int) int {
if gd.idles > 0 {
start := gd.check[0]
pos := 0
min := len(gd.auxiliary)
_ = min
for pos = start; pos <= min; {
pos = -1 * gd.check[pos]
if pos == start {
break
}
}
if pos > min {
//fmt.Printf("__find_pos: s=%d c=%d pos=%d\n", s, c, pos)
return pos
}
}
//fmt.Printf("__find_pos: s=%d c=%d, extend array\n", s, c)
if err := gd.extend(0); err != nil {
return -1
}
return len(gd.base) / 2
}
// 为s的下一个状态c找到一个位置
// 返回:
// pos: c的位置, -1 为失败
// exist: pos位置是否有字符c
// conflict: 是否冲突
func (gd *GoDat) findPos(s, c int) (pos int, exist, conflict bool) {
if s == 0 {
pos = gd.base[0] + c
} else {
if gd.base[s] == 0 {
// base[s] == 0 该位置是一个中间状态,其base值由找到的位置和c来决定
// 为c找到位置
pos = gd.__find_pos(s, c)
if pos > 0 {
gd.base[s] = pos - c
}
return
} else if gd.base[s] > 0 {
// 只有base[0] = 0
pos = gd.base[s] + c
} else if gd.base[s] == DAT_END_POS {
// base[s] == DAT_END_POS 该位置是一个结束位置
// 为c找到位置
if pos = gd.__find_pos(s, c); pos > 0 {
gd.base[s] = c - pos
}
return
} else {
// base[s] < 0
// base[s]是一个结束字符,且该字符后有其他字符
pos = -1*gd.base[s] + c
}
}
if pos >= len(gd.base) {
if err := gd.extend(pos); err != nil {
pos = -1
fmt.Println("extend failed:", err)
return
}
}
if gd.check[pos] == s {
exist = true
return
}
if gd.check[pos] >= 0 {
// gd.check [pos]不可能=0
conflict = true
}
return
}
// find new base
// return -1 if cannot found a free pos
func (gd *GoDat) findNewBase(s int, ca []int) (int, int) {
if gd.idles == 0 {
if err := gd.extend(0); err != nil {
return -1, 0
}
}
start := gd.check[0]
pos := start
arrLen := len(gd.base)
chs := len(gd.auxiliary)
//chs := 1
//if len(ca) > 0 {
// chs = ca[len(ca)-1]
//}
for ; ; pos = -1 * gd.check[pos] {
if -1*gd.check[pos] == start {
if err := gd.extend(0); err != nil {
return -1, 0
}
}
//fmt.Printf("find new base loop: pos=%d, c=%d\n", pos, ca[0])
// new base 的最小位置
if pos <= chs {
continue
}
found := true
newbase := pos - ca[0]
// 0位置跳过
i := 1
for ; i < len(ca); i++ {
c := ca[i]
if (newbase + c) >= arrLen {
if err := gd.extend(newbase + c); err != nil {
panic(fmt.Sprintf("findNewBase failed: s=%d, ca=%v, pos=%d, array_len=%d, newbase=%d",
s, ca, pos, len(gd.base), newbase))
return -1, 0
}
arrLen = len(gd.base)
// fmt.Printf(">>>>>>>>>>>>>>>>>>extended: newBase=%d, c=%d arrLen=%d\n", newbase, c, arrLen)
}
if gd.check[newbase+c] >= 0 {
found = false
//fmt.Println("breaked at", newbase, c)
break
}
}
if found {
return pos, newbase
}
}
panic(fmt.Sprintf("findNewBase failed: s=%d, ca=%v, pos=%d, array_len=%d", s, ca, pos, len(gd.base)))
return -1, 0
}
// re-locate
// s --> b
// newbase = pos - ca[0]
func (gd *GoDat) reLocate(s, pos, newbase int, ca []int) {
oldbase := gd.base[s]
//fmt.Printf("!!!!!!!!reLocate for s=%d, oldBase=%d, newBase=%d, pos=%d, ca=%v\n", s, oldbase, newbase, pos, ca)
for i, c := range ca {
// 新加入的状态
if i == len(ca)-1 {
// fmt.Printf("i=%d, newbase=%d, c=%d, base[%d]=0\n", i, newbase, c, newbase+c)
gd.delLink(newbase + c)
gd.check[newbase+c] = s
gd.base[newbase+c] = 0
} else {
// s的下一跳字符状态由oldPos迁移到newPos
newPos := newbase + c
oldPos := oldbase + c
if oldbase < 0 {
oldPos = -oldbase + c
}
gd.delLink(newPos)
gd.base[newPos] = gd.base[oldPos]
gd.check[newPos] = s
// 下一跳的check值更新
//fmt.Printf(" to be update pos %d, base=%d next children.\n", oldPos, gd.base[oldPos])
if gd.base[oldPos] != DAT_END_POS {
subca := gd.backtrace(oldPos, 0)
//fmt.Printf(" sub char-array for old pos %d: %v\n", oldPos, subca)
for _, child := range subca {
//fmt.Printf(" set check [%d] to %d oldPos=%d child=%d\n",
// gd.base[oldPos]+child, newPos, oldPos, child)
if gd.base[oldPos] > 0 {
gd.check[gd.base[oldPos]+child] = newPos
} else {
gd.check[-1*gd.base[oldPos]+child] = newPos
}
}
}
// 旧状态节点变为空闲
gd.addLink(oldPos)
//fmt.Printf("i=%d c=%d add idle pos %d, del idle pos %d\n\n", i, c, oldPos, newPos)
}
}
if oldbase < 0 {
gd.base[s] = -newbase
} else {
gd.base[s] = newbase
}
}
// 解除冲突 resolve conflicts
//
func (gd *GoDat) resolve(s, c int) (newPos int, err error) {
// 根据字符得到编码
ca := gd.backtrace(s, gd.revAuxiliary[c])
//fmt.Printf("backtrace %d: %v\n", s, ca)
// find new base with ca
pos, newbase := gd.findNewBase(s, ca)
if pos < 0 {
return -1, fmt.Errorf("find new base for %d, %d failed.", s, c)
}
// fmt.Printf("find new base: %d for s=%d, c=%d\n", pos, s, c)
// 重新定位
gd.reLocate(s, pos, newbase, ca)
newPos = newbase + c
return
}
// 增加一个模式
func (gd *GoDat) insertPattern(pat string) (err error) {
var (
r rune
c, i, s, t = 0, 0, 0, 0
arrLen = len(gd.base)
//patLen = len(pat)
exist bool
conflict bool
)
for i, r = range pat {
c = gd.auxiliary[r]
t, exist, conflict = gd.findPos(s, c)
//fmt.Printf("inserted: ch=%v, code=%d, s=%d[base=%d], t=%d, exist=%v, conflict=%v\n", string(r), c, s, gd.base[s], t, exist, conflict)
if t < 0 {
return fmt.Errorf("build pattern %s(%d) failed: origal array length=%d, now array length=%d",
pat, i, arrLen, len(gd.base))
}
if exist == false {
// exist==false的情况下, conflict有可能为true, 也有可能为false
if conflict {
// resolve已经把c加入到数组中, 且base设为0
t, err = gd.resolve(s, c)
if err != nil {
return
}
} else {
gd.delLink(t)
// base值由下一个确定
gd.base[t] = 0
gd.check[t] = s
}
}
s = t
}
// 最后一个字符
if gd.base[t] > 0 {
gd.base[t] = -1 * gd.base[t]
} else if gd.base[t] == 0 {
gd.base[t] = DAT_END_POS
}
return
}
// return:
// res = -10: pat有字符不在dat的字符表中,pat不在dat中
// res = -20: pat不在dat中
// res = -30: pat字符跳转表错误,pat不在dat中
func (gd *GoDat) removePattern(pat string) (res int, err error) {
var (
i int
r rune
rl int
s, t int
code int
patLen int
idx int
base int
nxtCnt int
)
if pat == "" {
return
}
patLen = len(pat)
if gd.nocase {
pat = strings.ToLower(pat)
}
for i < patLen {
r, rl = utf8.DecodeRuneInString(pat[i:])
code = gd.auxiliary[r]
if code == 0 {
res = -10
return
}
base = gd.base[s]
if base >= 0 {
t = base + code
} else {
if base == DAT_END_POS {
return
}
t = -base + code
}
if t > len(gd.base) {
res = -20
return
}
if gd.check[t] != s {
res = -30
err = fmt.Errorf("pat %s NOT in dat", pat)
return
}
i += rl
s = t
}
i -= rl
if base = gd.base[t]; base > 0 {
// 模式pat是其他模式的子串,但模式pat本身不存在
res = -40
err = fmt.Errorf("pat %s NOT in dat", pat)
return
}
assert(base != 0, "base should never be 0 except base[0]")
s = gd.check[t]
gd.attrs[t] = 0
if base == DAT_END_POS {
gd.addLink(t)
} else {
// 模式pat是其他模式的子串
assert(base < 0, "base should < 0 here")
gd.base[t] = -base
//fmt.Printf("pat %s is part of other pats, just set base positive.\n", pat)
goto delPat
}
// 计算第一个字符在gd.pats的偏移, 使commonPrefix更快
_, rl = utf8.DecodeRuneInString(pat[0:])
idx = gd.binSearch(pat[0:rl])
// 从后向前删除
r, rl = utf8.DecodeLastRuneInString(pat[0:i])
i -= rl
for i >= 0 {
t = s
nxtCnt = gd.commonCount(pat[0:i+rl], idx) - 1
//assert(nxtCnt == gd.nextStateCount(t), fmt.Sprintf("commonCount=%d should equal with nextStateCount=%d, t=%d,i=%d, rl=%d",
// nxtCnt, gd.nextStateCount(t), t, i, rl))
base = gd.base[t]
if base < 0 {
// 该节点是另一个模式的结束
if nxtCnt == 0 {
gd.base[t] = DAT_END_POS
}
goto delPat
}
if nxtCnt > 0 {
goto delPat
}
s = gd.check[t]
gd.addLink(t)
r, rl = utf8.DecodeLastRuneInString(pat[0:i])
if rl == 0 {
goto delPat
}
i -= rl
}
delPat:
// remove pattern from string array
idx = gd.binSearch(pat)
gd.pats = append(gd.pats[0:idx], gd.pats[idx+1:]...)
return
}
// 计算gd有多少个以subpat为开头的模式
func (gd *GoDat) nextStateCount(s int) int {
var (
code int
cnt int
base int
)
base = gd.base[s]
if base == DAT_END_POS {
return 0
}
if base < 0 {
base = -1 * base
}
for _, code = range gd.auxiliary {
if (base + code) >= len(gd.base) {
break
}
if gd.check[base+code] == s {
cnt++
}
}
return cnt
}