summaryrefslogtreecommitdiffstats
path: root/Lib/sre_compile.py
blob: c6398bfb83a5763df1ce9b383274c229e4faa567 (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
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
#
# Secret Labs' Regular Expression Engine
#
# convert template to internal format
#
# Copyright (c) 1997-2001 by Secret Labs AB.  All rights reserved.
#
# See the sre.py file for information on usage and redistribution.
#

"""Internal support module for sre"""

import _sre
import sre_parse
from sre_constants import *

assert _sre.MAGIC == MAGIC, "SRE module mismatch"

_LITERAL_CODES = {LITERAL, NOT_LITERAL}
_REPEATING_CODES = {REPEAT, MIN_REPEAT, MAX_REPEAT}
_SUCCESS_CODES = {SUCCESS, FAILURE}
_ASSERT_CODES = {ASSERT, ASSERT_NOT}
_UNIT_CODES = _LITERAL_CODES | {ANY, IN}

# Sets of lowercase characters which have the same uppercase.
_equivalences = (
    # LATIN SMALL LETTER I, LATIN SMALL LETTER DOTLESS I
    (0x69, 0x131), # iı
    # LATIN SMALL LETTER S, LATIN SMALL LETTER LONG S
    (0x73, 0x17f), # sſ
    # MICRO SIGN, GREEK SMALL LETTER MU
    (0xb5, 0x3bc), # µμ
    # COMBINING GREEK YPOGEGRAMMENI, GREEK SMALL LETTER IOTA, GREEK PROSGEGRAMMENI
    (0x345, 0x3b9, 0x1fbe), # \u0345ιι
    # GREEK SMALL LETTER IOTA WITH DIALYTIKA AND TONOS, GREEK SMALL LETTER IOTA WITH DIALYTIKA AND OXIA
    (0x390, 0x1fd3), # ΐΐ
    # GREEK SMALL LETTER UPSILON WITH DIALYTIKA AND TONOS, GREEK SMALL LETTER UPSILON WITH DIALYTIKA AND OXIA
    (0x3b0, 0x1fe3), # ΰΰ
    # GREEK SMALL LETTER BETA, GREEK BETA SYMBOL
    (0x3b2, 0x3d0), # βϐ
    # GREEK SMALL LETTER EPSILON, GREEK LUNATE EPSILON SYMBOL
    (0x3b5, 0x3f5), # εϵ
    # GREEK SMALL LETTER THETA, GREEK THETA SYMBOL
    (0x3b8, 0x3d1), # θϑ
    # GREEK SMALL LETTER KAPPA, GREEK KAPPA SYMBOL
    (0x3ba, 0x3f0), # κϰ
    # GREEK SMALL LETTER PI, GREEK PI SYMBOL
    (0x3c0, 0x3d6), # πϖ
    # GREEK SMALL LETTER RHO, GREEK RHO SYMBOL
    (0x3c1, 0x3f1), # ρϱ
    # GREEK SMALL LETTER FINAL SIGMA, GREEK SMALL LETTER SIGMA
    (0x3c2, 0x3c3), # ςσ
    # GREEK SMALL LETTER PHI, GREEK PHI SYMBOL
    (0x3c6, 0x3d5), # φϕ
    # LATIN SMALL LETTER S WITH DOT ABOVE, LATIN SMALL LETTER LONG S WITH DOT ABOVE
    (0x1e61, 0x1e9b), # ṡẛ
    # LATIN SMALL LIGATURE LONG S T, LATIN SMALL LIGATURE ST
    (0xfb05, 0xfb06), # ſtst
)

# Maps the lowercase code to lowercase codes which have the same uppercase.
_ignorecase_fixes = {i: tuple(j for j in t if i != j)
                     for t in _equivalences for i in t}

def _combine_flags(flags, add_flags, del_flags,
                   TYPE_FLAGS=sre_parse.TYPE_FLAGS):
    if add_flags & TYPE_FLAGS:
        flags &= ~TYPE_FLAGS
    return (flags | add_flags) & ~del_flags

def _compile(code, pattern, flags):
    # internal: compile a (sub)pattern
    emit = code.append
    _len = len
    LITERAL_CODES = _LITERAL_CODES
    REPEATING_CODES = _REPEATING_CODES
    SUCCESS_CODES = _SUCCESS_CODES
    ASSERT_CODES = _ASSERT_CODES
    iscased = None
    tolower = None
    fixes = None
    if flags & SRE_FLAG_IGNORECASE and not flags & SRE_FLAG_LOCALE:
        if flags & SRE_FLAG_UNICODE:
            iscased = _sre.unicode_iscased
            tolower = _sre.unicode_tolower
            fixes = _ignorecase_fixes
        else:
            iscased = _sre.ascii_iscased
            tolower = _sre.ascii_tolower
    for op, av in pattern:
        if op in LITERAL_CODES:
            if not flags & SRE_FLAG_IGNORECASE:
                emit(op)
                emit(av)
            elif flags & SRE_FLAG_LOCALE:
                emit(OP_LOCALE_IGNORE[op])
                emit(av)
            elif not iscased(av):
                emit(op)
                emit(av)
            else:
                lo = tolower(av)
                if not fixes:  # ascii
                    emit(OP_IGNORE[op])
                    emit(lo)
                elif lo not in fixes:
                    emit(OP_UNICODE_IGNORE[op])
                    emit(lo)
                else:
                    emit(IN_UNI_IGNORE)
                    skip = _len(code); emit(0)
                    if op is NOT_LITERAL:
                        emit(NEGATE)
                    for k in (lo,) + fixes[lo]:
                        emit(LITERAL)
                        emit(k)
                    emit(FAILURE)
                    code[skip] = _len(code) - skip
        elif op is IN:
            charset, hascased = _optimize_charset(av, iscased, tolower, fixes)
            if flags & SRE_FLAG_IGNORECASE and flags & SRE_FLAG_LOCALE:
                emit(IN_LOC_IGNORE)
            elif not hascased:
                emit(IN)
            elif not fixes:  # ascii
                emit(IN_IGNORE)
            else:
                emit(IN_UNI_IGNORE)
            skip = _len(code); emit(0)
            _compile_charset(charset, flags, code)
            code[skip] = _len(code) - skip
        elif op is ANY:
            if flags & SRE_FLAG_DOTALL:
                emit(ANY_ALL)
            else:
                emit(ANY)
        elif op in REPEATING_CODES:
            if flags & SRE_FLAG_TEMPLATE:
                raise error("internal: unsupported template operator %r" % (op,))
            if _simple(av[2]):
                if op is MAX_REPEAT:
                    emit(REPEAT_ONE)
                else:
                    emit(MIN_REPEAT_ONE)
                skip = _len(code); emit(0)
                emit(av[0])
                emit(av[1])
                _compile(code, av[2], flags)
                emit(SUCCESS)
                code[skip] = _len(code) - skip
            else:
                emit(REPEAT)
                skip = _len(code); emit(0)
                emit(av[0])
                emit(av[1])
                _compile(code, av[2], flags)
                code[skip] = _len(code) - skip
                if op is MAX_REPEAT:
                    emit(MAX_UNTIL)
                else:
                    emit(MIN_UNTIL)
        elif op is SUBPATTERN:
            group, add_flags, del_flags, p = av
            if group:
                emit(MARK)
                emit((group-1)*2)
            # _compile_info(code, p, _combine_flags(flags, add_flags, del_flags))
            _compile(code, p, _combine_flags(flags, add_flags, del_flags))
            if group:
                emit(MARK)
                emit((group-1)*2+1)
        elif op in SUCCESS_CODES:
            emit(op)
        elif op in ASSERT_CODES:
            emit(op)
            skip = _len(code); emit(0)
            if av[0] >= 0:
                emit(0) # look ahead
            else:
                lo, hi = av[1].getwidth()
                if lo != hi:
                    raise error("look-behind requires fixed-width pattern")
                emit(lo) # look behind
            _compile(code, av[1], flags)
            emit(SUCCESS)
            code[skip] = _len(code) - skip
        elif op is CALL:
            emit(op)
            skip = _len(code); emit(0)
            _compile(code, av, flags)
            emit(SUCCESS)
            code[skip] = _len(code) - skip
        elif op is AT:
            emit(op)
            if flags & SRE_FLAG_MULTILINE:
                av = AT_MULTILINE.get(av, av)
            if flags & SRE_FLAG_LOCALE:
                av = AT_LOCALE.get(av, av)
            elif flags & SRE_FLAG_UNICODE:
                av = AT_UNICODE.get(av, av)
            emit(av)
        elif op is BRANCH:
            emit(op)
            tail = []
            tailappend = tail.append
            for av in av[1]:
                skip = _len(code); emit(0)
                # _compile_info(code, av, flags)
                _compile(code, av, flags)
                emit(JUMP)
                tailappend(_len(code)); emit(0)
                code[skip] = _len(code) - skip
            emit(FAILURE) # end of branch
            for tail in tail:
                code[tail] = _len(code) - tail
        elif op is CATEGORY:
            emit(op)
            if flags & SRE_FLAG_LOCALE:
                av = CH_LOCALE[av]
            elif flags & SRE_FLAG_UNICODE:
                av = CH_UNICODE[av]
            emit(av)
        elif op is GROUPREF:
            if not flags & SRE_FLAG_IGNORECASE:
                emit(op)
            elif flags & SRE_FLAG_LOCALE:
                emit(GROUPREF_LOC_IGNORE)
            elif not fixes:  # ascii
                emit(GROUPREF_IGNORE)
            else:
                emit(GROUPREF_UNI_IGNORE)
            emit(av-1)
        elif op is GROUPREF_EXISTS:
            emit(op)
            emit(av[0]-1)
            skipyes = _len(code); emit(0)
            _compile(code, av[1], flags)
            if av[2]:
                emit(JUMP)
                skipno = _len(code); emit(0)
                code[skipyes] = _len(code) - skipyes + 1
                _compile(code, av[2], flags)
                code[skipno] = _len(code) - skipno
            else:
                code[skipyes] = _len(code) - skipyes + 1
        else:
            raise error("internal: unsupported operand type %r" % (op,))

def _compile_charset(charset, flags, code):
    # compile charset subprogram
    emit = code.append
    for op, av in charset:
        emit(op)
        if op is NEGATE:
            pass
        elif op is LITERAL:
            emit(av)
        elif op is RANGE or op is RANGE_UNI_IGNORE:
            emit(av[0])
            emit(av[1])
        elif op is CHARSET:
            code.extend(av)
        elif op is BIGCHARSET:
            code.extend(av)
        elif op is CATEGORY:
            if flags & SRE_FLAG_LOCALE:
                emit(CH_LOCALE[av])
            elif flags & SRE_FLAG_UNICODE:
                emit(CH_UNICODE[av])
            else:
                emit(av)
        else:
            raise error("internal: unsupported set operator %r" % (op,))
    emit(FAILURE)

def _optimize_charset(charset, iscased=None, fixup=None, fixes=None):
    # internal: optimize character set
    out = []
    tail = []
    charmap = bytearray(256)
    hascased = False
    for op, av in charset:
        while True:
            try:
                if op is LITERAL:
                    if fixup:
                        lo = fixup(av)
                        charmap[lo] = 1
                        if fixes and lo in fixes:
                            for k in fixes[lo]:
                                charmap[k] = 1
                        if not hascased and iscased(av):
                            hascased = True
                    else:
                        charmap[av] = 1
                elif op is RANGE:
                    r = range(av[0], av[1]+1)
                    if fixup:
                        if fixes:
                            for i in map(fixup, r):
                                charmap[i] = 1
                                if i in fixes:
                                    for k in fixes[i]:
                                        charmap[k] = 1
                        else:
                            for i in map(fixup, r):
                                charmap[i] = 1
                        if not hascased:
                            hascased = any(map(iscased, r))
                    else:
                        for i in r:
                            charmap[i] = 1
                elif op is NEGATE:
                    out.append((op, av))
                else:
                    tail.append((op, av))
            except IndexError:
                if len(charmap) == 256:
                    # character set contains non-UCS1 character codes
                    charmap += b'\0' * 0xff00
                    continue
                # Character set contains non-BMP character codes.
                if fixup:
                    hascased = True
                    # There are only two ranges of cased non-BMP characters:
                    # 10400-1044F (Deseret) and 118A0-118DF (Warang Citi),
                    # and for both ranges RANGE_UNI_IGNORE works.
                    if op is RANGE:
                        op = RANGE_UNI_IGNORE
                tail.append((op, av))
            break

    # compress character map
    runs = []
    q = 0
    while True:
        p = charmap.find(1, q)
        if p < 0:
            break
        if len(runs) >= 2:
            runs = None
            break
        q = charmap.find(0, p)
        if q < 0:
            runs.append((p, len(charmap)))
            break
        runs.append((p, q))
    if runs is not None:
        # use literal/range
        for p, q in runs:
            if q - p == 1:
                out.append((LITERAL, p))
            else:
                out.append((RANGE, (p, q - 1)))
        out += tail
        # if the case was changed or new representation is more compact
        if hascased or len(out) < len(charset):
            return out, hascased
        # else original character set is good enough
        return charset, hascased

    # use bitmap
    if len(charmap) == 256:
        data = _mk_bitmap(charmap)
        out.append((CHARSET, data))
        out += tail
        return out, hascased

    # To represent a big charset, first a bitmap of all characters in the
    # set is constructed. Then, this bitmap is sliced into chunks of 256
    # characters, duplicate chunks are eliminated, and each chunk is
    # given a number. In the compiled expression, the charset is
    # represented by a 32-bit word sequence, consisting of one word for
    # the number of different chunks, a sequence of 256 bytes (64 words)
    # of chunk numbers indexed by their original chunk position, and a
    # sequence of 256-bit chunks (8 words each).

    # Compression is normally good: in a typical charset, large ranges of
    # Unicode will be either completely excluded (e.g. if only cyrillic
    # letters are to be matched), or completely included (e.g. if large
    # subranges of Kanji match). These ranges will be represented by
    # chunks of all one-bits or all zero-bits.

    # Matching can be also done efficiently: the more significant byte of
    # the Unicode character is an index into the chunk number, and the
    # less significant byte is a bit index in the chunk (just like the
    # CHARSET matching).

    charmap = bytes(charmap) # should be hashable
    comps = {}
    mapping = bytearray(256)
    block = 0
    data = bytearray()
    for i in range(0, 65536, 256):
        chunk = charmap[i: i + 256]
        if chunk in comps:
            mapping[i // 256] = comps[chunk]
        else:
            mapping[i // 256] = comps[chunk] = block
            block += 1
            data += chunk
    data = _mk_bitmap(data)
    data[0:0] = [block] + _bytes_to_codes(mapping)
    out.append((BIGCHARSET, data))
    out += tail
    return out, hascased

_CODEBITS = _sre.CODESIZE * 8
MAXCODE = (1 << _CODEBITS) - 1
_BITS_TRANS = b'0' + b'1' * 255
def _mk_bitmap(bits, _CODEBITS=_CODEBITS, _int=int):
    s = bits.translate(_BITS_TRANS)[::-1]
    return [_int(s[i - _CODEBITS: i], 2)
            for i in range(len(s), 0, -_CODEBITS)]

def _bytes_to_codes(b):
    # Convert block indices to word array
    a = memoryview(b).cast('I')
    assert a.itemsize == _sre.CODESIZE
    assert len(a) * a.itemsize == len(b)
    return a.tolist()

def _simple(p):
    # check if this subpattern is a "simple" operator
    if len(p) != 1:
        return False
    op, av = p[0]
    if op is SUBPATTERN:
        return av[0] is None and _simple(av[-1])
    return op in _UNIT_CODES

def _generate_overlap_table(prefix):
    """
    Generate an overlap table for the following prefix.
    An overlap table is a table of the same size as the prefix which
    informs about the potential self-overlap for each index in the prefix:
    - if overlap[i] == 0, prefix[i:] can't overlap prefix[0:...]
    - if overlap[i] == k with 0 < k <= i, prefix[i-k+1:i+1] overlaps with
      prefix[0:k]
    """
    table = [0] * len(prefix)
    for i in range(1, len(prefix)):
        idx = table[i - 1]
        while prefix[i] != prefix[idx]:
            if idx == 0:
                table[i] = 0
                break
            idx = table[idx - 1]
        else:
            table[i] = idx + 1
    return table

def _get_iscased(flags):
    if not flags & SRE_FLAG_IGNORECASE:
        return None
    elif flags & SRE_FLAG_UNICODE:
        return _sre.unicode_iscased
    else:
        return _sre.ascii_iscased

def _get_literal_prefix(pattern, flags):
    # look for literal prefix
    prefix = []
    prefixappend = prefix.append
    prefix_skip = None
    iscased = _get_iscased(flags)
    for op, av in pattern.data:
        if op is LITERAL:
            if iscased and iscased(av):
                break
            prefixappend(av)
        elif op is SUBPATTERN:
            group, add_flags, del_flags, p = av
            flags1 = _combine_flags(flags, add_flags, del_flags)
            if flags1 & SRE_FLAG_IGNORECASE and flags1 & SRE_FLAG_LOCALE:
                break
            prefix1, prefix_skip1, got_all = _get_literal_prefix(p, flags1)
            if prefix_skip is None:
                if group is not None:
                    prefix_skip = len(prefix)
                elif prefix_skip1 is not None:
                    prefix_skip = len(prefix) + prefix_skip1
            prefix.extend(prefix1)
            if not got_all:
                break
        else:
            break
    else:
        return prefix, prefix_skip, True
    return prefix, prefix_skip, False

def _get_charset_prefix(pattern, flags):
    while True:
        if not pattern.data:
            return None
        op, av = pattern.data[0]
        if op is not SUBPATTERN:
            break
        group, add_flags, del_flags, pattern = av
        flags = _combine_flags(flags, add_flags, del_flags)
        if flags & SRE_FLAG_IGNORECASE and flags & SRE_FLAG_LOCALE:
            return None

    iscased = _get_iscased(flags)
    if op is LITERAL:
        if iscased and iscased(av):
            return None
        return [(op, av)]
    elif op is BRANCH:
        charset = []
        charsetappend = charset.append
        for p in av[1]:
            if not p:
                return None
            op, av = p[0]
            if op is LITERAL and not (iscased and iscased(av)):
                charsetappend((op, av))
            else:
                return None
        return charset
    elif op is IN:
        charset = av
        if iscased:
            for op, av in charset:
                if op is LITERAL:
                    if iscased(av):
                        return None
                elif op is RANGE:
                    if av[1] > 0xffff:
                        return None
                    if any(map(iscased, range(av[0], av[1]+1))):
                        return None
        return charset
    return None

def _compile_info(code, pattern, flags):
    # internal: compile an info block.  in the current version,
    # this contains min/max pattern width, and an optional literal
    # prefix or a character map
    lo, hi = pattern.getwidth()
    if hi > MAXCODE:
        hi = MAXCODE
    if lo == 0:
        code.extend([INFO, 4, 0, lo, hi])
        return
    # look for a literal prefix
    prefix = []
    prefix_skip = 0
    charset = [] # not used
    if not (flags & SRE_FLAG_IGNORECASE and flags & SRE_FLAG_LOCALE):
        # look for literal prefix
        prefix, prefix_skip, got_all = _get_literal_prefix(pattern, flags)
        # if no prefix, look for charset prefix
        if not prefix:
            charset = _get_charset_prefix(pattern, flags)
##     if prefix:
##         print("*** PREFIX", prefix, prefix_skip)
##     if charset:
##         print("*** CHARSET", charset)
    # add an info block
    emit = code.append
    emit(INFO)
    skip = len(code); emit(0)
    # literal flag
    mask = 0
    if prefix:
        mask = SRE_INFO_PREFIX
        if prefix_skip is None and got_all:
            mask = mask | SRE_INFO_LITERAL
    elif charset:
        mask = mask | SRE_INFO_CHARSET
    emit(mask)
    # pattern length
    if lo < MAXCODE:
        emit(lo)
    else:
        emit(MAXCODE)
        prefix = prefix[:MAXCODE]
    emit(min(hi, MAXCODE))
    # add literal prefix
    if prefix:
        emit(len(prefix)) # length
        if prefix_skip is None:
            prefix_skip =  len(prefix)
        emit(prefix_skip) # skip
        code.extend(prefix)
        # generate overlap table
        code.extend(_generate_overlap_table(prefix))
    elif charset:
        charset, hascased = _optimize_charset(charset)
        assert not hascased
        _compile_charset(charset, flags, code)
    code[skip] = len(code) - skip

def isstring(obj):
    return isinstance(obj, (str, bytes))

def _code(p, flags):

    flags = p.state.flags | flags
    code = []

    # compile info block
    _compile_info(code, p, flags)

    # compile the pattern
    _compile(code, p.data, flags)

    code.append(SUCCESS)

    return code

def _hex_code(code):
    return '[%s]' % ', '.join('%#0*x' % (_sre.CODESIZE*2+2, x) for x in code)

def dis(code):
    import sys

    labels = set()
    level = 0
    offset_width = len(str(len(code) - 1))

    def dis_(start, end):
        def print_(*args, to=None):
            if to is not None:
                labels.add(to)
                args += ('(to %d)' % (to,),)
            print('%*d%s ' % (offset_width, start, ':' if start in labels else '.'),
                  end='  '*(level-1))
            print(*args)

        def print_2(*args):
            print(end=' '*(offset_width + 2*level))
            print(*args)

        nonlocal level
        level += 1
        i = start
        while i < end:
            start = i
            op = code[i]
            i += 1
            op = OPCODES[op]
            if op in (SUCCESS, FAILURE, ANY, ANY_ALL,
                      MAX_UNTIL, MIN_UNTIL, NEGATE):
                print_(op)
            elif op in (LITERAL, NOT_LITERAL,
                        LITERAL_IGNORE, NOT_LITERAL_IGNORE,
                        LITERAL_UNI_IGNORE, NOT_LITERAL_UNI_IGNORE,
                        LITERAL_LOC_IGNORE, NOT_LITERAL_LOC_IGNORE):
                arg = code[i]
                i += 1
                print_(op, '%#02x (%r)' % (arg, chr(arg)))
            elif op is AT:
                arg = code[i]
                i += 1
                arg = str(ATCODES[arg])
                assert arg[:3] == 'AT_'
                print_(op, arg[3:])
            elif op is CATEGORY:
                arg = code[i]
                i += 1
                arg = str(CHCODES[arg])
                assert arg[:9] == 'CATEGORY_'
                print_(op, arg[9:])
            elif op in (IN, IN_IGNORE, IN_UNI_IGNORE, IN_LOC_IGNORE):
                skip = code[i]
                print_(op, skip, to=i+skip)
                dis_(i+1, i+skip)
                i += skip
            elif op in (RANGE, RANGE_UNI_IGNORE):
                lo, hi = code[i: i+2]
                i += 2
                print_(op, '%#02x %#02x (%r-%r)' % (lo, hi, chr(lo), chr(hi)))
            elif op is CHARSET:
                print_(op, _hex_code(code[i: i + 256//_CODEBITS]))
                i += 256//_CODEBITS
            elif op is BIGCHARSET:
                arg = code[i]
                i += 1
                mapping = list(b''.join(x.to_bytes(_sre.CODESIZE, sys.byteorder)
                                        for x in code[i: i + 256//_sre.CODESIZE]))
                print_(op, arg, mapping)
                i += 256//_sre.CODESIZE
                level += 1
                for j in range(arg):
                    print_2(_hex_code(code[i: i + 256//_CODEBITS]))
                    i += 256//_CODEBITS
                level -= 1
            elif op in (MARK, GROUPREF, GROUPREF_IGNORE, GROUPREF_UNI_IGNORE,
                        GROUPREF_LOC_IGNORE):
                arg = code[i]
                i += 1
                print_(op, arg)
            elif op is JUMP:
                skip = code[i]
                print_(op, skip, to=i+skip)
                i += 1
            elif op is BRANCH:
                skip = code[i]
                print_(op, skip, to=i+skip)
                while skip:
                    dis_(i+1, i+skip)
                    i += skip
                    start = i
                    skip = code[i]
                    if skip:
                        print_('branch', skip, to=i+skip)
                    else:
                        print_(FAILURE)
                i += 1
            elif op in (REPEAT, REPEAT_ONE, MIN_REPEAT_ONE):
                skip, min, max = code[i: i+3]
                if max == MAXREPEAT:
                    max = 'MAXREPEAT'
                print_(op, skip, min, max, to=i+skip)
                dis_(i+3, i+skip)
                i += skip
            elif op is GROUPREF_EXISTS:
                arg, skip = code[i: i+2]
                print_(op, arg, skip, to=i+skip)
                i += 2
            elif op in (ASSERT, ASSERT_NOT):
                skip, arg = code[i: i+2]
                print_(op, skip, arg, to=i+skip)
                dis_(i+2, i+skip)
                i += skip
            elif op is INFO:
                skip, flags, min, max = code[i: i+4]
                if max == MAXREPEAT:
                    max = 'MAXREPEAT'
                print_(op, skip, bin(flags), min, max, to=i+skip)
                start = i+4
                if flags & SRE_INFO_PREFIX:
                    prefix_len, prefix_skip = code[i+4: i+6]
                    print_2('  prefix_skip', prefix_skip)
                    start = i + 6
                    prefix = code[start: start+prefix_len]
                    print_2('  prefix',
                            '[%s]' % ', '.join('%#02x' % x for x in prefix),
                            '(%r)' % ''.join(map(chr, prefix)))
                    start += prefix_len
                    print_2('  overlap', code[start: start+prefix_len])
                    start += prefix_len
                if flags & SRE_INFO_CHARSET:
                    level += 1
                    print_2('in')
                    dis_(start, i+skip)
                    level -= 1
                i += skip
            else:
                raise ValueError(op)

        level -= 1

    dis_(0, len(code))


def compile(p, flags=0):
    # internal: convert pattern list to internal format

    if isstring(p):
        pattern = p
        p = sre_parse.parse(p, flags)
    else:
        pattern = None

    code = _code(p, flags)

    if flags & SRE_FLAG_DEBUG:
        print()
        dis(code)

    # map in either direction
    groupindex = p.state.groupdict
    indexgroup = [None] * p.state.groups
    for k, i in groupindex.items():
        indexgroup[i] = k

    return _sre.compile(
        pattern, flags | p.state.flags, code,
        p.state.groups-1,
        groupindex, tuple(indexgroup)
        )