forked from PCRE2Project/pcre2
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathpcre2_study.c
1829 lines (1541 loc) · 52.5 KB
/
pcre2_study.c
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
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/*************************************************
* Perl-Compatible Regular Expressions *
*************************************************/
/* PCRE is a library of functions to support regular expressions whose syntax
and semantics are as close as possible to those of the Perl 5 language.
Written by Philip Hazel
Original API code Copyright (c) 1997-2012 University of Cambridge
New API code Copyright (c) 2016-2023 University of Cambridge
-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
* Redistributions of source code must retain the above copyright notice,
this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.
* Neither the name of the University of Cambridge nor the names of its
contributors may be used to endorse or promote products derived from
this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
POSSIBILITY OF SUCH DAMAGE.
-----------------------------------------------------------------------------
*/
/* This module contains functions for scanning a compiled pattern and
collecting data (e.g. minimum matching length). */
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "pcre2_internal.h"
/* The maximum remembered capturing brackets minimum. */
#define MAX_CACHE_BACKREF 128
/* Set a bit in the starting code unit bit map. */
#define SET_BIT(c) re->start_bitmap[(c)/8] |= (1u << ((c)&7))
/* Returns from set_start_bits() */
enum { SSB_FAIL, SSB_DONE, SSB_CONTINUE, SSB_UNKNOWN, SSB_TOODEEP };
/*************************************************
* Find the minimum subject length for a group *
*************************************************/
/* Scan a parenthesized group and compute the minimum length of subject that
is needed to match it. This is a lower bound; it does not mean there is a
string of that length that matches. In UTF mode, the result is in characters
rather than code units. The field in a compiled pattern for storing the minimum
length is 16-bits long (on the grounds that anything longer than that is
pathological), so we give up when we reach that amount. This also means that
integer overflow for really crazy patterns cannot happen.
Backreference minimum lengths are cached to speed up multiple references. This
function is called only when the highest back reference in the pattern is less
than or equal to MAX_CACHE_BACKREF, which is one less than the size of the
caching vector. The zeroth element contains the number of the highest set
value.
Arguments:
re compiled pattern block
code pointer to start of group (the bracket)
startcode pointer to start of the whole pattern's code
utf UTF flag
recurses chain of recurse_check to catch mutual recursion
countptr pointer to call count (to catch over complexity)
backref_cache vector for caching back references.
This function is no longer called when the pattern contains (*ACCEPT); however,
the old code for returning -1 is retained, just in case.
Returns: the minimum length
-1 \C in UTF-8 mode
or (*ACCEPT)
or pattern too complicated
-2 internal error (missing capturing bracket)
-3 internal error (opcode not listed)
*/
static int
find_minlength(const pcre2_real_code *re, PCRE2_SPTR code,
PCRE2_SPTR startcode, BOOL utf, recurse_check *recurses, int *countptr,
int *backref_cache)
{
int length = -1;
int branchlength = 0;
int prev_cap_recno = -1;
int prev_cap_d = 0;
int prev_recurse_recno = -1;
int prev_recurse_d = 0;
uint32_t once_fudge = 0;
BOOL had_recurse = FALSE;
BOOL dupcapused = (re->flags & PCRE2_DUPCAPUSED) != 0;
PCRE2_SPTR nextbranch = code + GET(code, 1);
PCRE2_UCHAR *cc = (PCRE2_UCHAR *)code + 1 + LINK_SIZE;
recurse_check this_recurse;
/* If this is a "could be empty" group, its minimum length is 0. */
if (*code >= OP_SBRA && *code <= OP_SCOND) return 0;
/* Skip over capturing bracket number */
if (*code == OP_CBRA || *code == OP_CBRAPOS) cc += IMM2_SIZE;
/* A large and/or complex regex can take too long to process. */
if ((*countptr)++ > 1000) return -1;
/* Scan along the opcodes for this branch. If we get to the end of the branch,
check the length against that of the other branches. If the accumulated length
passes 16-bits, reset to that value and skip the rest of the branch. */
for (;;)
{
int d, min, recno;
PCRE2_UCHAR op, *cs, *ce;
if (branchlength >= UINT16_MAX)
{
branchlength = UINT16_MAX;
cc = (PCRE2_UCHAR *)nextbranch;
}
op = *cc;
switch (op)
{
case OP_COND:
case OP_SCOND:
/* If there is only one branch in a condition, the implied branch has zero
length, so we don't add anything. This covers the DEFINE "condition"
automatically. If there are two branches we can treat it the same as any
other non-capturing subpattern. */
cs = cc + GET(cc, 1);
if (*cs != OP_ALT)
{
cc = cs + 1 + LINK_SIZE;
break;
}
goto PROCESS_NON_CAPTURE;
case OP_BRA:
/* There's a special case of OP_BRA, when it is wrapped round a repeated
OP_RECURSE. We'd like to process the latter at this level so that
remembering the value works for repeated cases. So we do nothing, but
set a fudge value to skip over the OP_KET after the recurse. */
if (cc[1+LINK_SIZE] == OP_RECURSE && cc[2*(1+LINK_SIZE)] == OP_KET)
{
once_fudge = 1 + LINK_SIZE;
cc += 1 + LINK_SIZE;
break;
}
/* Fall through */
case OP_ONCE:
case OP_SCRIPT_RUN:
case OP_SBRA:
case OP_BRAPOS:
case OP_SBRAPOS:
PROCESS_NON_CAPTURE:
d = find_minlength(re, cc, startcode, utf, recurses, countptr,
backref_cache);
if (d < 0) return d;
branchlength += d;
do cc += GET(cc, 1); while (*cc == OP_ALT);
cc += 1 + LINK_SIZE;
break;
/* To save time for repeated capturing subpatterns, we remember the
length of the previous one. Unfortunately we can't do the same for
the unnumbered ones above. Nor can we do this if (?| is present in the
pattern because captures with the same number are not then identical. */
case OP_CBRA:
case OP_SCBRA:
case OP_CBRAPOS:
case OP_SCBRAPOS:
recno = (int)GET2(cc, 1+LINK_SIZE);
if (dupcapused || recno != prev_cap_recno)
{
prev_cap_recno = recno;
prev_cap_d = find_minlength(re, cc, startcode, utf, recurses, countptr,
backref_cache);
if (prev_cap_d < 0) return prev_cap_d;
}
branchlength += prev_cap_d;
do cc += GET(cc, 1); while (*cc == OP_ALT);
cc += 1 + LINK_SIZE;
break;
/* ACCEPT makes things far too complicated; we have to give up. In fact,
from 10.34 onwards, if a pattern contains (*ACCEPT), this function is not
used. However, leave the code in place, just in case. */
case OP_ACCEPT:
case OP_ASSERT_ACCEPT:
return -1;
/* Reached end of a branch; if it's a ket it is the end of a nested
call. If it's ALT it is an alternation in a nested call. If it is END it's
the end of the outer call. All can be handled by the same code. If the
length of any branch is zero, there is no need to scan any subsequent
branches. */
case OP_ALT:
case OP_KET:
case OP_KETRMAX:
case OP_KETRMIN:
case OP_KETRPOS:
case OP_END:
if (length < 0 || (!had_recurse && branchlength < length))
length = branchlength;
if (op != OP_ALT || length == 0) return length;
nextbranch = cc + GET(cc, 1);
cc += 1 + LINK_SIZE;
branchlength = 0;
had_recurse = FALSE;
break;
/* Skip over assertive subpatterns */
case OP_ASSERT:
case OP_ASSERT_NOT:
case OP_ASSERTBACK:
case OP_ASSERTBACK_NOT:
case OP_ASSERT_NA:
case OP_ASSERTBACK_NA:
do cc += GET(cc, 1); while (*cc == OP_ALT);
/* Fall through */
/* Skip over things that don't match chars */
case OP_REVERSE:
case OP_CREF:
case OP_DNCREF:
case OP_RREF:
case OP_DNRREF:
case OP_FALSE:
case OP_TRUE:
case OP_CALLOUT:
case OP_SOD:
case OP_SOM:
case OP_EOD:
case OP_EODN:
case OP_CIRC:
case OP_CIRCM:
case OP_DOLL:
case OP_DOLLM:
case OP_NOT_WORD_BOUNDARY:
case OP_WORD_BOUNDARY:
case OP_NOT_UCP_WORD_BOUNDARY:
case OP_UCP_WORD_BOUNDARY:
cc += PRIV(OP_lengths)[*cc];
break;
case OP_CALLOUT_STR:
cc += GET(cc, 1 + 2*LINK_SIZE);
break;
/* Skip over a subpattern that has a {0} or {0,x} quantifier */
case OP_BRAZERO:
case OP_BRAMINZERO:
case OP_BRAPOSZERO:
case OP_SKIPZERO:
cc += PRIV(OP_lengths)[*cc];
do cc += GET(cc, 1); while (*cc == OP_ALT);
cc += 1 + LINK_SIZE;
break;
/* Handle literal characters and + repetitions */
case OP_CHAR:
case OP_CHARI:
case OP_NOT:
case OP_NOTI:
case OP_PLUS:
case OP_PLUSI:
case OP_MINPLUS:
case OP_MINPLUSI:
case OP_POSPLUS:
case OP_POSPLUSI:
case OP_NOTPLUS:
case OP_NOTPLUSI:
case OP_NOTMINPLUS:
case OP_NOTMINPLUSI:
case OP_NOTPOSPLUS:
case OP_NOTPOSPLUSI:
branchlength++;
cc += 2;
#ifdef SUPPORT_UNICODE
if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
#endif
break;
case OP_TYPEPLUS:
case OP_TYPEMINPLUS:
case OP_TYPEPOSPLUS:
branchlength++;
cc += (cc[1] == OP_PROP || cc[1] == OP_NOTPROP)? 4 : 2;
break;
/* Handle exact repetitions. The count is already in characters, but we
may need to skip over a multibyte character in UTF mode. */
case OP_EXACT:
case OP_EXACTI:
case OP_NOTEXACT:
case OP_NOTEXACTI:
branchlength += GET2(cc,1);
cc += 2 + IMM2_SIZE;
#ifdef SUPPORT_UNICODE
if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
#endif
break;
case OP_TYPEEXACT:
branchlength += GET2(cc,1);
cc += 2 + IMM2_SIZE + ((cc[1 + IMM2_SIZE] == OP_PROP
|| cc[1 + IMM2_SIZE] == OP_NOTPROP)? 2 : 0);
break;
/* Handle single-char non-literal matchers */
case OP_PROP:
case OP_NOTPROP:
cc += 2;
/* Fall through */
case OP_NOT_DIGIT:
case OP_DIGIT:
case OP_NOT_WHITESPACE:
case OP_WHITESPACE:
case OP_NOT_WORDCHAR:
case OP_WORDCHAR:
case OP_ANY:
case OP_ALLANY:
case OP_EXTUNI:
case OP_HSPACE:
case OP_NOT_HSPACE:
case OP_VSPACE:
case OP_NOT_VSPACE:
branchlength++;
cc++;
break;
/* "Any newline" might match two characters, but it also might match just
one. */
case OP_ANYNL:
branchlength += 1;
cc++;
break;
/* The single-byte matcher means we can't proceed in UTF mode. (In
non-UTF mode \C will actually be turned into OP_ALLANY, so won't ever
appear, but leave the code, just in case.) */
case OP_ANYBYTE:
#ifdef SUPPORT_UNICODE
if (utf) return -1;
#endif
branchlength++;
cc++;
break;
/* For repeated character types, we have to test for \p and \P, which have
an extra two bytes of parameters. */
case OP_TYPESTAR:
case OP_TYPEMINSTAR:
case OP_TYPEQUERY:
case OP_TYPEMINQUERY:
case OP_TYPEPOSSTAR:
case OP_TYPEPOSQUERY:
if (cc[1] == OP_PROP || cc[1] == OP_NOTPROP) cc += 2;
cc += PRIV(OP_lengths)[op];
break;
case OP_TYPEUPTO:
case OP_TYPEMINUPTO:
case OP_TYPEPOSUPTO:
if (cc[1 + IMM2_SIZE] == OP_PROP
|| cc[1 + IMM2_SIZE] == OP_NOTPROP) cc += 2;
cc += PRIV(OP_lengths)[op];
break;
/* Check a class for variable quantification */
case OP_CLASS:
case OP_NCLASS:
#ifdef SUPPORT_WIDE_CHARS
case OP_XCLASS:
/* The original code caused an unsigned overflow in 64 bit systems,
so now we use a conditional statement. */
if (op == OP_XCLASS)
cc += GET(cc, 1);
else
cc += PRIV(OP_lengths)[OP_CLASS];
#else
cc += PRIV(OP_lengths)[OP_CLASS];
#endif
switch (*cc)
{
case OP_CRPLUS:
case OP_CRMINPLUS:
case OP_CRPOSPLUS:
branchlength++;
/* Fall through */
case OP_CRSTAR:
case OP_CRMINSTAR:
case OP_CRQUERY:
case OP_CRMINQUERY:
case OP_CRPOSSTAR:
case OP_CRPOSQUERY:
cc++;
break;
case OP_CRRANGE:
case OP_CRMINRANGE:
case OP_CRPOSRANGE:
branchlength += GET2(cc,1);
cc += 1 + 2 * IMM2_SIZE;
break;
default:
branchlength++;
break;
}
break;
/* Backreferences and subroutine calls (OP_RECURSE) are treated in the same
way: we find the minimum length for the subpattern. A recursion
(backreference or subroutine) causes an a flag to be set that causes the
length of this branch to be ignored. The logic is that a recursion can only
make sense if there is another alternative that stops the recursing. That
will provide the minimum length (when no recursion happens).
If PCRE2_MATCH_UNSET_BACKREF is set, a backreference to an unset bracket
matches an empty string (by default it causes a matching failure), so in
that case we must set the minimum length to zero.
For backreferenes, if duplicate numbers are present in the pattern we check
for a reference to a duplicate. If it is, we don't know which version will
be referenced, so we have to set the minimum length to zero. */
/* Duplicate named pattern back reference. */
case OP_DNREF:
case OP_DNREFI:
if (!dupcapused && (re->overall_options & PCRE2_MATCH_UNSET_BACKREF) == 0)
{
int count = GET2(cc, 1+IMM2_SIZE);
PCRE2_UCHAR *slot =
(PCRE2_UCHAR *)((uint8_t *)re + sizeof(pcre2_real_code)) +
GET2(cc, 1) * re->name_entry_size;
d = INT_MAX;
/* Scan all groups with the same name; find the shortest. */
while (count-- > 0)
{
int dd, i;
recno = GET2(slot, 0);
if (recno <= backref_cache[0] && backref_cache[recno] >= 0)
dd = backref_cache[recno];
else
{
ce = cs = (PCRE2_UCHAR *)PRIV(find_bracket)(startcode, utf, recno);
if (cs == NULL) return -2;
do ce += GET(ce, 1); while (*ce == OP_ALT);
dd = 0;
if (!dupcapused ||
(PCRE2_UCHAR *)PRIV(find_bracket)(ce, utf, recno) == NULL)
{
if (cc > cs && cc < ce) /* Simple recursion */
{
had_recurse = TRUE;
}
else
{
recurse_check *r = recurses;
for (r = recurses; r != NULL; r = r->prev)
if (r->group == cs) break;
if (r != NULL) /* Mutual recursion */
{
had_recurse = TRUE;
}
else
{
this_recurse.prev = recurses; /* No recursion */
this_recurse.group = cs;
dd = find_minlength(re, cs, startcode, utf, &this_recurse,
countptr, backref_cache);
if (dd < 0) return dd;
}
}
}
backref_cache[recno] = dd;
for (i = backref_cache[0] + 1; i < recno; i++) backref_cache[i] = -1;
backref_cache[0] = recno;
}
if (dd < d) d = dd;
if (d <= 0) break; /* No point looking at any more */
slot += re->name_entry_size;
}
}
else d = 0;
cc += 1 + 2*IMM2_SIZE;
goto REPEAT_BACK_REFERENCE;
/* Single back reference by number. References by name are converted to by
number when there is no duplication. */
case OP_REF:
case OP_REFI:
recno = GET2(cc, 1);
if (recno <= backref_cache[0] && backref_cache[recno] >= 0)
d = backref_cache[recno];
else
{
int i;
d = 0;
if ((re->overall_options & PCRE2_MATCH_UNSET_BACKREF) == 0)
{
ce = cs = (PCRE2_UCHAR *)PRIV(find_bracket)(startcode, utf, recno);
if (cs == NULL) return -2;
do ce += GET(ce, 1); while (*ce == OP_ALT);
if (!dupcapused ||
(PCRE2_UCHAR *)PRIV(find_bracket)(ce, utf, recno) == NULL)
{
if (cc > cs && cc < ce) /* Simple recursion */
{
had_recurse = TRUE;
}
else
{
recurse_check *r = recurses;
for (r = recurses; r != NULL; r = r->prev) if (r->group == cs) break;
if (r != NULL) /* Mutual recursion */
{
had_recurse = TRUE;
}
else /* No recursion */
{
this_recurse.prev = recurses;
this_recurse.group = cs;
d = find_minlength(re, cs, startcode, utf, &this_recurse, countptr,
backref_cache);
if (d < 0) return d;
}
}
}
}
backref_cache[recno] = d;
for (i = backref_cache[0] + 1; i < recno; i++) backref_cache[i] = -1;
backref_cache[0] = recno;
}
cc += 1 + IMM2_SIZE;
/* Handle repeated back references */
REPEAT_BACK_REFERENCE:
switch (*cc)
{
case OP_CRSTAR:
case OP_CRMINSTAR:
case OP_CRQUERY:
case OP_CRMINQUERY:
case OP_CRPOSSTAR:
case OP_CRPOSQUERY:
min = 0;
cc++;
break;
case OP_CRPLUS:
case OP_CRMINPLUS:
case OP_CRPOSPLUS:
min = 1;
cc++;
break;
case OP_CRRANGE:
case OP_CRMINRANGE:
case OP_CRPOSRANGE:
min = GET2(cc, 1);
cc += 1 + 2 * IMM2_SIZE;
break;
default:
min = 1;
break;
}
/* Take care not to overflow: (1) min and d are ints, so check that their
product is not greater than INT_MAX. (2) branchlength is limited to
UINT16_MAX (checked at the top of the loop). */
if ((d > 0 && (INT_MAX/d) < min) || UINT16_MAX - branchlength < min*d)
branchlength = UINT16_MAX;
else branchlength += min * d;
break;
/* Recursion always refers to the first occurrence of a subpattern with a
given number. Therefore, we can always make use of caching, even when the
pattern contains multiple subpatterns with the same number. */
case OP_RECURSE:
cs = ce = (PCRE2_UCHAR *)startcode + GET(cc, 1);
recno = GET2(cs, 1+LINK_SIZE);
if (recno == prev_recurse_recno)
{
branchlength += prev_recurse_d;
}
else
{
do ce += GET(ce, 1); while (*ce == OP_ALT);
if (cc > cs && cc < ce) /* Simple recursion */
had_recurse = TRUE;
else
{
recurse_check *r = recurses;
for (r = recurses; r != NULL; r = r->prev) if (r->group == cs) break;
if (r != NULL) /* Mutual recursion */
had_recurse = TRUE;
else
{
this_recurse.prev = recurses;
this_recurse.group = cs;
prev_recurse_d = find_minlength(re, cs, startcode, utf, &this_recurse,
countptr, backref_cache);
if (prev_recurse_d < 0) return prev_recurse_d;
prev_recurse_recno = recno;
branchlength += prev_recurse_d;
}
}
}
cc += 1 + LINK_SIZE + once_fudge;
once_fudge = 0;
break;
/* Anything else does not or need not match a character. We can get the
item's length from the table, but for those that can match zero occurrences
of a character, we must take special action for UTF-8 characters. As it
happens, the "NOT" versions of these opcodes are used at present only for
ASCII characters, so they could be omitted from this list. However, in
future that may change, so we include them here so as not to leave a
gotcha for a future maintainer. */
case OP_UPTO:
case OP_UPTOI:
case OP_NOTUPTO:
case OP_NOTUPTOI:
case OP_MINUPTO:
case OP_MINUPTOI:
case OP_NOTMINUPTO:
case OP_NOTMINUPTOI:
case OP_POSUPTO:
case OP_POSUPTOI:
case OP_NOTPOSUPTO:
case OP_NOTPOSUPTOI:
case OP_STAR:
case OP_STARI:
case OP_NOTSTAR:
case OP_NOTSTARI:
case OP_MINSTAR:
case OP_MINSTARI:
case OP_NOTMINSTAR:
case OP_NOTMINSTARI:
case OP_POSSTAR:
case OP_POSSTARI:
case OP_NOTPOSSTAR:
case OP_NOTPOSSTARI:
case OP_QUERY:
case OP_QUERYI:
case OP_NOTQUERY:
case OP_NOTQUERYI:
case OP_MINQUERY:
case OP_MINQUERYI:
case OP_NOTMINQUERY:
case OP_NOTMINQUERYI:
case OP_POSQUERY:
case OP_POSQUERYI:
case OP_NOTPOSQUERY:
case OP_NOTPOSQUERYI:
cc += PRIV(OP_lengths)[op];
#ifdef SUPPORT_UNICODE
if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
#endif
break;
/* Skip these, but we need to add in the name length. */
case OP_MARK:
case OP_COMMIT_ARG:
case OP_PRUNE_ARG:
case OP_SKIP_ARG:
case OP_THEN_ARG:
cc += PRIV(OP_lengths)[op] + cc[1];
break;
/* The remaining opcodes are just skipped over. */
case OP_CLOSE:
case OP_COMMIT:
case OP_FAIL:
case OP_PRUNE:
case OP_SET_SOM:
case OP_SKIP:
case OP_THEN:
cc += PRIV(OP_lengths)[op];
break;
/* This should not occur: we list all opcodes explicitly so that when
new ones get added they are properly considered. */
default:
return -3;
}
}
/* Control never gets here */
}
/*************************************************
* Set a bit and maybe its alternate case *
*************************************************/
/* Given a character, set its first code unit's bit in the table, and also the
corresponding bit for the other version of a letter if we are caseless.
Arguments:
re points to the regex block
p points to the first code unit of the character
caseless TRUE if caseless
utf TRUE for UTF mode
ucp TRUE for UCP mode
Returns: pointer after the character
*/
static PCRE2_SPTR
set_table_bit(pcre2_real_code *re, PCRE2_SPTR p, BOOL caseless, BOOL utf,
BOOL ucp)
{
uint32_t c = *p++; /* First code unit */
(void)utf; /* Stop compiler warnings when UTF not supported */
(void)ucp;
/* In 16-bit and 32-bit modes, code units greater than 0xff set the bit for
0xff. */
#if PCRE2_CODE_UNIT_WIDTH != 8
if (c > 0xff) SET_BIT(0xff); else
#endif
SET_BIT(c);
/* In UTF-8 or UTF-16 mode, pick up the remaining code units in order to find
the end of the character, even when caseless. */
#ifdef SUPPORT_UNICODE
if (utf)
{
#if PCRE2_CODE_UNIT_WIDTH == 8
if (c >= 0xc0) GETUTF8INC(c, p);
#elif PCRE2_CODE_UNIT_WIDTH == 16
if ((c & 0xfc00) == 0xd800) GETUTF16INC(c, p);
#endif
}
#endif /* SUPPORT_UNICODE */
/* If caseless, handle the other case of the character. */
if (caseless)
{
#ifdef SUPPORT_UNICODE
if (utf || ucp)
{
c = UCD_OTHERCASE(c);
#if PCRE2_CODE_UNIT_WIDTH == 8
if (utf)
{
PCRE2_UCHAR buff[6];
(void)PRIV(ord2utf)(c, buff);
SET_BIT(buff[0]);
}
else if (c < 256) SET_BIT(c);
#else /* 16-bit or 32-bit mode */
if (c > 0xff) SET_BIT(0xff); else SET_BIT(c);
#endif
}
else
#endif /* SUPPORT_UNICODE */
/* Not UTF or UCP */
if (MAX_255(c)) SET_BIT(re->tables[fcc_offset + c]);
}
return p;
}
/*************************************************
* Set bits for a positive character type *
*************************************************/
/* This function sets starting bits for a character type. In UTF-8 mode, we can
only do a direct setting for bytes less than 128, as otherwise there can be
confusion with bytes in the middle of UTF-8 characters. In a "traditional"
environment, the tables will only recognize ASCII characters anyway, but in at
least one Windows environment, some higher bytes bits were set in the tables.
So we deal with that case by considering the UTF-8 encoding.
Arguments:
re the regex block
cbit type the type of character wanted
table_limit 32 for non-UTF-8; 16 for UTF-8
Returns: nothing
*/
static void
set_type_bits(pcre2_real_code *re, int cbit_type, unsigned int table_limit)
{
uint32_t c;
for (c = 0; c < table_limit; c++)
re->start_bitmap[c] |= re->tables[c+cbits_offset+cbit_type];
#if defined SUPPORT_UNICODE && PCRE2_CODE_UNIT_WIDTH == 8
if (table_limit == 32) return;
for (c = 128; c < 256; c++)
{
if ((re->tables[cbits_offset + c/8] & (1u << (c&7))) != 0)
{
PCRE2_UCHAR buff[6];
(void)PRIV(ord2utf)(c, buff);
SET_BIT(buff[0]);
}
}
#endif /* UTF-8 */
}
/*************************************************
* Set bits for a negative character type *
*************************************************/
/* This function sets starting bits for a negative character type such as \D.
In UTF-8 mode, we can only do a direct setting for bytes less than 128, as
otherwise there can be confusion with bytes in the middle of UTF-8 characters.
Unlike in the positive case, where we can set appropriate starting bits for
specific high-valued UTF-8 characters, in this case we have to set the bits for
all high-valued characters. The lowest is 0xc2, but we overkill by starting at
0xc0 (192) for simplicity.
Arguments:
re the regex block
cbit type the type of character wanted
table_limit 32 for non-UTF-8; 16 for UTF-8
Returns: nothing
*/
static void
set_nottype_bits(pcre2_real_code *re, int cbit_type, unsigned int table_limit)
{
uint32_t c;
for (c = 0; c < table_limit; c++)
re->start_bitmap[c] |= (uint8_t)(~(re->tables[c+cbits_offset+cbit_type]));
#if defined SUPPORT_UNICODE && PCRE2_CODE_UNIT_WIDTH == 8
if (table_limit != 32) for (c = 24; c < 32; c++) re->start_bitmap[c] = 0xff;
#endif
}
/*************************************************
* Create bitmap of starting code units *
*************************************************/
/* This function scans a compiled unanchored expression recursively and
attempts to build a bitmap of the set of possible starting code units whose
values are less than 256. In 16-bit and 32-bit mode, values above 255 all cause
the 255 bit to be set. When calling set[_not]_type_bits() in UTF-8 (sic) mode
we pass a value of 16 rather than 32 as the final argument. (See comments in
those functions for the reason.)
The SSB_CONTINUE return is useful for parenthesized groups in patterns such as
(a*)b where the group provides some optional starting code units but scanning
must continue at the outer level to find at least one mandatory code unit. At
the outermost level, this function fails unless the result is SSB_DONE.
We restrict recursion (for nested groups) to 1000 to avoid stack overflow
issues.
Arguments:
re points to the compiled regex block
code points to an expression
utf TRUE if in UTF mode
ucp TRUE if in UCP mode
depthptr pointer to recurse depth
Returns: SSB_FAIL => Failed to find any starting code units
SSB_DONE => Found mandatory starting code units
SSB_CONTINUE => Found optional starting code units
SSB_UNKNOWN => Hit an unrecognized opcode
SSB_TOODEEP => Recursion is too deep
*/
static int
set_start_bits(pcre2_real_code *re, PCRE2_SPTR code, BOOL utf, BOOL ucp,
int *depthptr)
{
uint32_t c;
int yield = SSB_DONE;
#if defined SUPPORT_UNICODE && PCRE2_CODE_UNIT_WIDTH == 8
int table_limit = utf? 16:32;
#else
int table_limit = 32;
#endif
*depthptr += 1;
if (*depthptr > 1000) return SSB_TOODEEP;
do
{
BOOL try_next = TRUE;
PCRE2_SPTR tcode = code + 1 + LINK_SIZE;
if (*code == OP_CBRA || *code == OP_SCBRA ||
*code == OP_CBRAPOS || *code == OP_SCBRAPOS) tcode += IMM2_SIZE;
while (try_next) /* Loop for items in this branch */
{
int rc;
uint8_t *classmap = NULL;
#ifdef SUPPORT_WIDE_CHARS
PCRE2_UCHAR xclassflags;
#endif
switch(*tcode)
{
/* If we reach something we don't understand, it means a new opcode has
been created that hasn't been added to this function. Hopefully this
problem will be discovered during testing. */
default:
return SSB_UNKNOWN;
/* Fail for a valid opcode that implies no starting bits. */
case OP_ACCEPT:
case OP_ASSERT_ACCEPT:
case OP_ALLANY:
case OP_ANY: