File: dirstate.py

package info (click to toggle)
mercurial 4.0-1+deb9u1
  • links: PTS, VCS
  • area: main
  • in suites: stretch
  • size: 32,100 kB
  • ctags: 13,376
  • sloc: python: 94,424; ansic: 8,392; tcl: 3,715; lisp: 1,448; sh: 999; makefile: 375; xml: 35
file content (1259 lines) | stat: -rw-r--r-- 46,083 bytes parent folder | download
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
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
# dirstate.py - working directory tracking for mercurial
#
# Copyright 2005-2007 Matt Mackall <mpm@selenic.com>
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.

from __future__ import absolute_import

import collections
import errno
import os
import stat

from .i18n import _
from .node import nullid
from . import (
    encoding,
    error,
    match as matchmod,
    osutil,
    parsers,
    pathutil,
    scmutil,
    util,
)

propertycache = util.propertycache
filecache = scmutil.filecache
_rangemask = 0x7fffffff

dirstatetuple = parsers.dirstatetuple

class repocache(filecache):
    """filecache for files in .hg/"""
    def join(self, obj, fname):
        return obj._opener.join(fname)

class rootcache(filecache):
    """filecache for files in the repository root"""
    def join(self, obj, fname):
        return obj._join(fname)

def _getfsnow(vfs):
    '''Get "now" timestamp on filesystem'''
    tmpfd, tmpname = vfs.mkstemp()
    try:
        return os.fstat(tmpfd).st_mtime
    finally:
        os.close(tmpfd)
        vfs.unlink(tmpname)

def nonnormalentries(dmap):
    '''Compute the nonnormal dirstate entries from the dmap'''
    try:
        return parsers.nonnormalentries(dmap)
    except AttributeError:
        return set(fname for fname, e in dmap.iteritems()
                   if e[0] != 'n' or e[3] == -1)

def _trypending(root, vfs, filename):
    '''Open  file to be read according to HG_PENDING environment variable

    This opens '.pending' of specified 'filename' only when HG_PENDING
    is equal to 'root'.

    This returns '(fp, is_pending_opened)' tuple.
    '''
    if root == os.environ.get('HG_PENDING'):
        try:
            return (vfs('%s.pending' % filename), True)
        except IOError as inst:
            if inst.errno != errno.ENOENT:
                raise
    return (vfs(filename), False)

class dirstate(object):

    def __init__(self, opener, ui, root, validate):
        '''Create a new dirstate object.

        opener is an open()-like callable that can be used to open the
        dirstate file; root is the root of the directory tracked by
        the dirstate.
        '''
        self._opener = opener
        self._validate = validate
        self._root = root
        # ntpath.join(root, '') of Python 2.7.9 does not add sep if root is
        # UNC path pointing to root share (issue4557)
        self._rootdir = pathutil.normasprefix(root)
        # internal config: ui.forcecwd
        forcecwd = ui.config('ui', 'forcecwd')
        if forcecwd:
            self._cwd = forcecwd
        self._dirty = False
        self._dirtypl = False
        self._lastnormaltime = 0
        self._ui = ui
        self._filecache = {}
        self._parentwriters = 0
        self._filename = 'dirstate'
        self._pendingfilename = '%s.pending' % self._filename
        self._plchangecallbacks = {}
        self._origpl = None

        # for consistent view between _pl() and _read() invocations
        self._pendingmode = None

    def beginparentchange(self):
        '''Marks the beginning of a set of changes that involve changing
        the dirstate parents. If there is an exception during this time,
        the dirstate will not be written when the wlock is released. This
        prevents writing an incoherent dirstate where the parent doesn't
        match the contents.
        '''
        self._parentwriters += 1

    def endparentchange(self):
        '''Marks the end of a set of changes that involve changing the
        dirstate parents. Once all parent changes have been marked done,
        the wlock will be free to write the dirstate on release.
        '''
        if self._parentwriters > 0:
            self._parentwriters -= 1

    def pendingparentchange(self):
        '''Returns true if the dirstate is in the middle of a set of changes
        that modify the dirstate parent.
        '''
        return self._parentwriters > 0

    @propertycache
    def _map(self):
        '''Return the dirstate contents as a map from filename to
        (state, mode, size, time).'''
        self._read()
        return self._map

    @propertycache
    def _copymap(self):
        self._read()
        return self._copymap

    @propertycache
    def _nonnormalset(self):
        return nonnormalentries(self._map)

    @propertycache
    def _filefoldmap(self):
        try:
            makefilefoldmap = parsers.make_file_foldmap
        except AttributeError:
            pass
        else:
            return makefilefoldmap(self._map, util.normcasespec,
                                   util.normcasefallback)

        f = {}
        normcase = util.normcase
        for name, s in self._map.iteritems():
            if s[0] != 'r':
                f[normcase(name)] = name
        f['.'] = '.' # prevents useless util.fspath() invocation
        return f

    @propertycache
    def _dirfoldmap(self):
        f = {}
        normcase = util.normcase
        for name in self._dirs:
            f[normcase(name)] = name
        return f

    @repocache('branch')
    def _branch(self):
        try:
            return self._opener.read("branch").strip() or "default"
        except IOError as inst:
            if inst.errno != errno.ENOENT:
                raise
            return "default"

    @propertycache
    def _pl(self):
        try:
            fp = self._opendirstatefile()
            st = fp.read(40)
            fp.close()
            l = len(st)
            if l == 40:
                return st[:20], st[20:40]
            elif l > 0 and l < 40:
                raise error.Abort(_('working directory state appears damaged!'))
        except IOError as err:
            if err.errno != errno.ENOENT:
                raise
        return [nullid, nullid]

    @propertycache
    def _dirs(self):
        return util.dirs(self._map, 'r')

    def dirs(self):
        return self._dirs

    @rootcache('.hgignore')
    def _ignore(self):
        files = self._ignorefiles()
        if not files:
            return util.never

        pats = ['include:%s' % f for f in files]
        return matchmod.match(self._root, '', [], pats, warn=self._ui.warn)

    @propertycache
    def _slash(self):
        return self._ui.configbool('ui', 'slash') and os.sep != '/'

    @propertycache
    def _checklink(self):
        return util.checklink(self._root)

    @propertycache
    def _checkexec(self):
        return util.checkexec(self._root)

    @propertycache
    def _checkcase(self):
        return not util.fscasesensitive(self._join('.hg'))

    def _join(self, f):
        # much faster than os.path.join()
        # it's safe because f is always a relative path
        return self._rootdir + f

    def flagfunc(self, buildfallback):
        if self._checklink and self._checkexec:
            def f(x):
                try:
                    st = os.lstat(self._join(x))
                    if util.statislink(st):
                        return 'l'
                    if util.statisexec(st):
                        return 'x'
                except OSError:
                    pass
                return ''
            return f

        fallback = buildfallback()
        if self._checklink:
            def f(x):
                if os.path.islink(self._join(x)):
                    return 'l'
                if 'x' in fallback(x):
                    return 'x'
                return ''
            return f
        if self._checkexec:
            def f(x):
                if 'l' in fallback(x):
                    return 'l'
                if util.isexec(self._join(x)):
                    return 'x'
                return ''
            return f
        else:
            return fallback

    @propertycache
    def _cwd(self):
        return os.getcwd()

    def getcwd(self):
        '''Return the path from which a canonical path is calculated.

        This path should be used to resolve file patterns or to convert
        canonical paths back to file paths for display. It shouldn't be
        used to get real file paths. Use vfs functions instead.
        '''
        cwd = self._cwd
        if cwd == self._root:
            return ''
        # self._root ends with a path separator if self._root is '/' or 'C:\'
        rootsep = self._root
        if not util.endswithsep(rootsep):
            rootsep += os.sep
        if cwd.startswith(rootsep):
            return cwd[len(rootsep):]
        else:
            # we're outside the repo. return an absolute path.
            return cwd

    def pathto(self, f, cwd=None):
        if cwd is None:
            cwd = self.getcwd()
        path = util.pathto(self._root, cwd, f)
        if self._slash:
            return util.pconvert(path)
        return path

    def __getitem__(self, key):
        '''Return the current state of key (a filename) in the dirstate.

        States are:
          n  normal
          m  needs merging
          r  marked for removal
          a  marked for addition
          ?  not tracked
        '''
        return self._map.get(key, ("?",))[0]

    def __contains__(self, key):
        return key in self._map

    def __iter__(self):
        for x in sorted(self._map):
            yield x

    def iteritems(self):
        return self._map.iteritems()

    def parents(self):
        return [self._validate(p) for p in self._pl]

    def p1(self):
        return self._validate(self._pl[0])

    def p2(self):
        return self._validate(self._pl[1])

    def branch(self):
        return encoding.tolocal(self._branch)

    def setparents(self, p1, p2=nullid):
        """Set dirstate parents to p1 and p2.

        When moving from two parents to one, 'm' merged entries a
        adjusted to normal and previous copy records discarded and
        returned by the call.

        See localrepo.setparents()
        """
        if self._parentwriters == 0:
            raise ValueError("cannot set dirstate parent without "
                             "calling dirstate.beginparentchange")

        self._dirty = self._dirtypl = True
        oldp2 = self._pl[1]
        if self._origpl is None:
            self._origpl = self._pl
        self._pl = p1, p2
        copies = {}
        if oldp2 != nullid and p2 == nullid:
            for f, s in self._map.iteritems():
                # Discard 'm' markers when moving away from a merge state
                if s[0] == 'm':
                    if f in self._copymap:
                        copies[f] = self._copymap[f]
                    self.normallookup(f)
                # Also fix up otherparent markers
                elif s[0] == 'n' and s[2] == -2:
                    if f in self._copymap:
                        copies[f] = self._copymap[f]
                    self.add(f)
        return copies

    def setbranch(self, branch):
        self._branch = encoding.fromlocal(branch)
        f = self._opener('branch', 'w', atomictemp=True, checkambig=True)
        try:
            f.write(self._branch + '\n')
            f.close()

            # make sure filecache has the correct stat info for _branch after
            # replacing the underlying file
            ce = self._filecache['_branch']
            if ce:
                ce.refresh()
        except: # re-raises
            f.discard()
            raise

    def _opendirstatefile(self):
        fp, mode = _trypending(self._root, self._opener, self._filename)
        if self._pendingmode is not None and self._pendingmode != mode:
            fp.close()
            raise error.Abort(_('working directory state may be '
                                'changed parallelly'))
        self._pendingmode = mode
        return fp

    def _read(self):
        self._map = {}
        self._copymap = {}
        try:
            fp = self._opendirstatefile()
            try:
                st = fp.read()
            finally:
                fp.close()
        except IOError as err:
            if err.errno != errno.ENOENT:
                raise
            return
        if not st:
            return

        if util.safehasattr(parsers, 'dict_new_presized'):
            # Make an estimate of the number of files in the dirstate based on
            # its size. From a linear regression on a set of real-world repos,
            # all over 10,000 files, the size of a dirstate entry is 85
            # bytes. The cost of resizing is significantly higher than the cost
            # of filling in a larger presized dict, so subtract 20% from the
            # size.
            #
            # This heuristic is imperfect in many ways, so in a future dirstate
            # format update it makes sense to just record the number of entries
            # on write.
            self._map = parsers.dict_new_presized(len(st) / 71)

        # Python's garbage collector triggers a GC each time a certain number
        # of container objects (the number being defined by
        # gc.get_threshold()) are allocated. parse_dirstate creates a tuple
        # for each file in the dirstate. The C version then immediately marks
        # them as not to be tracked by the collector. However, this has no
        # effect on when GCs are triggered, only on what objects the GC looks
        # into. This means that O(number of files) GCs are unavoidable.
        # Depending on when in the process's lifetime the dirstate is parsed,
        # this can get very expensive. As a workaround, disable GC while
        # parsing the dirstate.
        #
        # (we cannot decorate the function directly since it is in a C module)
        parse_dirstate = util.nogc(parsers.parse_dirstate)
        p = parse_dirstate(self._map, self._copymap, st)
        if not self._dirtypl:
            self._pl = p

    def invalidate(self):
        for a in ("_map", "_copymap", "_filefoldmap", "_dirfoldmap", "_branch",
                  "_pl", "_dirs", "_ignore", "_nonnormalset"):
            if a in self.__dict__:
                delattr(self, a)
        self._lastnormaltime = 0
        self._dirty = False
        self._parentwriters = 0
        self._origpl = None

    def copy(self, source, dest):
        """Mark dest as a copy of source. Unmark dest if source is None."""
        if source == dest:
            return
        self._dirty = True
        if source is not None:
            self._copymap[dest] = source
        elif dest in self._copymap:
            del self._copymap[dest]

    def copied(self, file):
        return self._copymap.get(file, None)

    def copies(self):
        return self._copymap

    def _droppath(self, f):
        if self[f] not in "?r" and "_dirs" in self.__dict__:
            self._dirs.delpath(f)

        if "_filefoldmap" in self.__dict__:
            normed = util.normcase(f)
            if normed in self._filefoldmap:
                del self._filefoldmap[normed]

    def _addpath(self, f, state, mode, size, mtime):
        oldstate = self[f]
        if state == 'a' or oldstate == 'r':
            scmutil.checkfilename(f)
            if f in self._dirs:
                raise error.Abort(_('directory %r already in dirstate') % f)
            # shadows
            for d in util.finddirs(f):
                if d in self._dirs:
                    break
                if d in self._map and self[d] != 'r':
                    raise error.Abort(
                        _('file %r in dirstate clashes with %r') % (d, f))
        if oldstate in "?r" and "_dirs" in self.__dict__:
            self._dirs.addpath(f)
        self._dirty = True
        self._map[f] = dirstatetuple(state, mode, size, mtime)
        if state != 'n' or mtime == -1:
            self._nonnormalset.add(f)

    def normal(self, f):
        '''Mark a file normal and clean.'''
        s = os.lstat(self._join(f))
        mtime = s.st_mtime
        self._addpath(f, 'n', s.st_mode,
                      s.st_size & _rangemask, mtime & _rangemask)
        if f in self._copymap:
            del self._copymap[f]
        if f in self._nonnormalset:
            self._nonnormalset.remove(f)
        if mtime > self._lastnormaltime:
            # Remember the most recent modification timeslot for status(),
            # to make sure we won't miss future size-preserving file content
            # modifications that happen within the same timeslot.
            self._lastnormaltime = mtime

    def normallookup(self, f):
        '''Mark a file normal, but possibly dirty.'''
        if self._pl[1] != nullid and f in self._map:
            # if there is a merge going on and the file was either
            # in state 'm' (-1) or coming from other parent (-2) before
            # being removed, restore that state.
            entry = self._map[f]
            if entry[0] == 'r' and entry[2] in (-1, -2):
                source = self._copymap.get(f)
                if entry[2] == -1:
                    self.merge(f)
                elif entry[2] == -2:
                    self.otherparent(f)
                if source:
                    self.copy(source, f)
                return
            if entry[0] == 'm' or entry[0] == 'n' and entry[2] == -2:
                return
        self._addpath(f, 'n', 0, -1, -1)
        if f in self._copymap:
            del self._copymap[f]
        if f in self._nonnormalset:
            self._nonnormalset.remove(f)

    def otherparent(self, f):
        '''Mark as coming from the other parent, always dirty.'''
        if self._pl[1] == nullid:
            raise error.Abort(_("setting %r to other parent "
                               "only allowed in merges") % f)
        if f in self and self[f] == 'n':
            # merge-like
            self._addpath(f, 'm', 0, -2, -1)
        else:
            # add-like
            self._addpath(f, 'n', 0, -2, -1)

        if f in self._copymap:
            del self._copymap[f]

    def add(self, f):
        '''Mark a file added.'''
        self._addpath(f, 'a', 0, -1, -1)
        if f in self._copymap:
            del self._copymap[f]

    def remove(self, f):
        '''Mark a file removed.'''
        self._dirty = True
        self._droppath(f)
        size = 0
        if self._pl[1] != nullid and f in self._map:
            # backup the previous state
            entry = self._map[f]
            if entry[0] == 'm': # merge
                size = -1
            elif entry[0] == 'n' and entry[2] == -2: # other parent
                size = -2
        self._map[f] = dirstatetuple('r', 0, size, 0)
        self._nonnormalset.add(f)
        if size == 0 and f in self._copymap:
            del self._copymap[f]

    def merge(self, f):
        '''Mark a file merged.'''
        if self._pl[1] == nullid:
            return self.normallookup(f)
        return self.otherparent(f)

    def drop(self, f):
        '''Drop a file from the dirstate'''
        if f in self._map:
            self._dirty = True
            self._droppath(f)
            del self._map[f]
            if f in self._nonnormalset:
                self._nonnormalset.remove(f)
            if f in self._copymap:
                del self._copymap[f]

    def _discoverpath(self, path, normed, ignoremissing, exists, storemap):
        if exists is None:
            exists = os.path.lexists(os.path.join(self._root, path))
        if not exists:
            # Maybe a path component exists
            if not ignoremissing and '/' in path:
                d, f = path.rsplit('/', 1)
                d = self._normalize(d, False, ignoremissing, None)
                folded = d + "/" + f
            else:
                # No path components, preserve original case
                folded = path
        else:
            # recursively normalize leading directory components
            # against dirstate
            if '/' in normed:
                d, f = normed.rsplit('/', 1)
                d = self._normalize(d, False, ignoremissing, True)
                r = self._root + "/" + d
                folded = d + "/" + util.fspath(f, r)
            else:
                folded = util.fspath(normed, self._root)
            storemap[normed] = folded

        return folded

    def _normalizefile(self, path, isknown, ignoremissing=False, exists=None):
        normed = util.normcase(path)
        folded = self._filefoldmap.get(normed, None)
        if folded is None:
            if isknown:
                folded = path
            else:
                folded = self._discoverpath(path, normed, ignoremissing, exists,
                                            self._filefoldmap)
        return folded

    def _normalize(self, path, isknown, ignoremissing=False, exists=None):
        normed = util.normcase(path)
        folded = self._filefoldmap.get(normed, None)
        if folded is None:
            folded = self._dirfoldmap.get(normed, None)
        if folded is None:
            if isknown:
                folded = path
            else:
                # store discovered result in dirfoldmap so that future
                # normalizefile calls don't start matching directories
                folded = self._discoverpath(path, normed, ignoremissing, exists,
                                            self._dirfoldmap)
        return folded

    def normalize(self, path, isknown=False, ignoremissing=False):
        '''
        normalize the case of a pathname when on a casefolding filesystem

        isknown specifies whether the filename came from walking the
        disk, to avoid extra filesystem access.

        If ignoremissing is True, missing path are returned
        unchanged. Otherwise, we try harder to normalize possibly
        existing path components.

        The normalized case is determined based on the following precedence:

        - version of name already stored in the dirstate
        - version of name stored on disk
        - version provided via command arguments
        '''

        if self._checkcase:
            return self._normalize(path, isknown, ignoremissing)
        return path

    def clear(self):
        self._map = {}
        self._nonnormalset = set()
        if "_dirs" in self.__dict__:
            delattr(self, "_dirs")
        self._copymap = {}
        self._pl = [nullid, nullid]
        self._lastnormaltime = 0
        self._dirty = True

    def rebuild(self, parent, allfiles, changedfiles=None):
        if changedfiles is None:
            # Rebuild entire dirstate
            changedfiles = allfiles
            lastnormaltime = self._lastnormaltime
            self.clear()
            self._lastnormaltime = lastnormaltime

        if self._origpl is None:
            self._origpl = self._pl
        self._pl = (parent, nullid)
        for f in changedfiles:
            if f in allfiles:
                self.normallookup(f)
            else:
                self.drop(f)

        self._dirty = True

    def write(self, tr):
        if not self._dirty:
            return

        filename = self._filename
        if tr:
            # 'dirstate.write()' is not only for writing in-memory
            # changes out, but also for dropping ambiguous timestamp.
            # delayed writing re-raise "ambiguous timestamp issue".
            # See also the wiki page below for detail:
            # https://www.mercurial-scm.org/wiki/DirstateTransactionPlan

            # emulate dropping timestamp in 'parsers.pack_dirstate'
            now = _getfsnow(self._opener)
            dmap = self._map
            for f, e in dmap.iteritems():
                if e[0] == 'n' and e[3] == now:
                    dmap[f] = dirstatetuple(e[0], e[1], e[2], -1)
                    self._nonnormalset.add(f)

            # emulate that all 'dirstate.normal' results are written out
            self._lastnormaltime = 0

            # delay writing in-memory changes out
            tr.addfilegenerator('dirstate', (self._filename,),
                                self._writedirstate, location='plain')
            return

        st = self._opener(filename, "w", atomictemp=True, checkambig=True)
        self._writedirstate(st)

    def addparentchangecallback(self, category, callback):
        """add a callback to be called when the wd parents are changed

        Callback will be called with the following arguments:
            dirstate, (oldp1, oldp2), (newp1, newp2)

        Category is a unique identifier to allow overwriting an old callback
        with a newer callback.
        """
        self._plchangecallbacks[category] = callback

    def _writedirstate(self, st):
        # notify callbacks about parents change
        if self._origpl is not None and self._origpl != self._pl:
            for c, callback in sorted(self._plchangecallbacks.iteritems()):
                callback(self, self._origpl, self._pl)
            self._origpl = None
        # use the modification time of the newly created temporary file as the
        # filesystem's notion of 'now'
        now = util.fstat(st).st_mtime & _rangemask

        # enough 'delaywrite' prevents 'pack_dirstate' from dropping
        # timestamp of each entries in dirstate, because of 'now > mtime'
        delaywrite = self._ui.configint('debug', 'dirstate.delaywrite', 0)
        if delaywrite > 0:
            # do we have any files to delay for?
            for f, e in self._map.iteritems():
                if e[0] == 'n' and e[3] == now:
                    import time # to avoid useless import
                    # rather than sleep n seconds, sleep until the next
                    # multiple of n seconds
                    clock = time.time()
                    start = int(clock) - (int(clock) % delaywrite)
                    end = start + delaywrite
                    time.sleep(end - clock)
                    now = end # trust our estimate that the end is near now
                    break

        st.write(parsers.pack_dirstate(self._map, self._copymap, self._pl, now))
        self._nonnormalset = nonnormalentries(self._map)
        st.close()
        self._lastnormaltime = 0
        self._dirty = self._dirtypl = False

    def _dirignore(self, f):
        if f == '.':
            return False
        if self._ignore(f):
            return True
        for p in util.finddirs(f):
            if self._ignore(p):
                return True
        return False

    def _ignorefiles(self):
        files = []
        if os.path.exists(self._join('.hgignore')):
            files.append(self._join('.hgignore'))
        for name, path in self._ui.configitems("ui"):
            if name == 'ignore' or name.startswith('ignore.'):
                # we need to use os.path.join here rather than self._join
                # because path is arbitrary and user-specified
                files.append(os.path.join(self._rootdir, util.expandpath(path)))
        return files

    def _ignorefileandline(self, f):
        files = collections.deque(self._ignorefiles())
        visited = set()
        while files:
            i = files.popleft()
            patterns = matchmod.readpatternfile(i, self._ui.warn,
                                                sourceinfo=True)
            for pattern, lineno, line in patterns:
                kind, p = matchmod._patsplit(pattern, 'glob')
                if kind == "subinclude":
                    if p not in visited:
                        files.append(p)
                    continue
                m = matchmod.match(self._root, '', [], [pattern],
                                   warn=self._ui.warn)
                if m(f):
                    return (i, lineno, line)
            visited.add(i)
        return (None, -1, "")

    def _walkexplicit(self, match, subrepos):
        '''Get stat data about the files explicitly specified by match.

        Return a triple (results, dirsfound, dirsnotfound).
        - results is a mapping from filename to stat result. It also contains
          listings mapping subrepos and .hg to None.
        - dirsfound is a list of files found to be directories.
        - dirsnotfound is a list of files that the dirstate thinks are
          directories and that were not found.'''

        def badtype(mode):
            kind = _('unknown')
            if stat.S_ISCHR(mode):
                kind = _('character device')
            elif stat.S_ISBLK(mode):
                kind = _('block device')
            elif stat.S_ISFIFO(mode):
                kind = _('fifo')
            elif stat.S_ISSOCK(mode):
                kind = _('socket')
            elif stat.S_ISDIR(mode):
                kind = _('directory')
            return _('unsupported file type (type is %s)') % kind

        matchedir = match.explicitdir
        badfn = match.bad
        dmap = self._map
        lstat = os.lstat
        getkind = stat.S_IFMT
        dirkind = stat.S_IFDIR
        regkind = stat.S_IFREG
        lnkkind = stat.S_IFLNK
        join = self._join
        dirsfound = []
        foundadd = dirsfound.append
        dirsnotfound = []
        notfoundadd = dirsnotfound.append

        if not match.isexact() and self._checkcase:
            normalize = self._normalize
        else:
            normalize = None

        files = sorted(match.files())
        subrepos.sort()
        i, j = 0, 0
        while i < len(files) and j < len(subrepos):
            subpath = subrepos[j] + "/"
            if files[i] < subpath:
                i += 1
                continue
            while i < len(files) and files[i].startswith(subpath):
                del files[i]
            j += 1

        if not files or '.' in files:
            files = ['.']
        results = dict.fromkeys(subrepos)
        results['.hg'] = None

        alldirs = None
        for ff in files:
            # constructing the foldmap is expensive, so don't do it for the
            # common case where files is ['.']
            if normalize and ff != '.':
                nf = normalize(ff, False, True)
            else:
                nf = ff
            if nf in results:
                continue

            try:
                st = lstat(join(nf))
                kind = getkind(st.st_mode)
                if kind == dirkind:
                    if nf in dmap:
                        # file replaced by dir on disk but still in dirstate
                        results[nf] = None
                    if matchedir:
                        matchedir(nf)
                    foundadd((nf, ff))
                elif kind == regkind or kind == lnkkind:
                    results[nf] = st
                else:
                    badfn(ff, badtype(kind))
                    if nf in dmap:
                        results[nf] = None
            except OSError as inst: # nf not found on disk - it is dirstate only
                if nf in dmap: # does it exactly match a missing file?
                    results[nf] = None
                else: # does it match a missing directory?
                    if alldirs is None:
                        alldirs = util.dirs(dmap)
                    if nf in alldirs:
                        if matchedir:
                            matchedir(nf)
                        notfoundadd(nf)
                    else:
                        badfn(ff, inst.strerror)

        # Case insensitive filesystems cannot rely on lstat() failing to detect
        # a case-only rename.  Prune the stat object for any file that does not
        # match the case in the filesystem, if there are multiple files that
        # normalize to the same path.
        if match.isexact() and self._checkcase:
            normed = {}

            for f, st in results.iteritems():
                if st is None:
                    continue

                nc = util.normcase(f)
                paths = normed.get(nc)

                if paths is None:
                    paths = set()
                    normed[nc] = paths

                paths.add(f)

            for norm, paths in normed.iteritems():
                if len(paths) > 1:
                    for path in paths:
                        folded = self._discoverpath(path, norm, True, None,
                                                    self._dirfoldmap)
                        if path != folded:
                            results[path] = None

        return results, dirsfound, dirsnotfound

    def walk(self, match, subrepos, unknown, ignored, full=True):
        '''
        Walk recursively through the directory tree, finding all files
        matched by match.

        If full is False, maybe skip some known-clean files.

        Return a dict mapping filename to stat-like object (either
        mercurial.osutil.stat instance or return value of os.stat()).

        '''
        # full is a flag that extensions that hook into walk can use -- this
        # implementation doesn't use it at all. This satisfies the contract
        # because we only guarantee a "maybe".

        if ignored:
            ignore = util.never
            dirignore = util.never
        elif unknown:
            ignore = self._ignore
            dirignore = self._dirignore
        else:
            # if not unknown and not ignored, drop dir recursion and step 2
            ignore = util.always
            dirignore = util.always

        matchfn = match.matchfn
        matchalways = match.always()
        matchtdir = match.traversedir
        dmap = self._map
        listdir = osutil.listdir
        lstat = os.lstat
        dirkind = stat.S_IFDIR
        regkind = stat.S_IFREG
        lnkkind = stat.S_IFLNK
        join = self._join

        exact = skipstep3 = False
        if match.isexact(): # match.exact
            exact = True
            dirignore = util.always # skip step 2
        elif match.prefix(): # match.match, no patterns
            skipstep3 = True

        if not exact and self._checkcase:
            normalize = self._normalize
            normalizefile = self._normalizefile
            skipstep3 = False
        else:
            normalize = self._normalize
            normalizefile = None

        # step 1: find all explicit files
        results, work, dirsnotfound = self._walkexplicit(match, subrepos)

        skipstep3 = skipstep3 and not (work or dirsnotfound)
        work = [d for d in work if not dirignore(d[0])]

        # step 2: visit subdirectories
        def traverse(work, alreadynormed):
            wadd = work.append
            while work:
                nd = work.pop()
                skip = None
                if nd == '.':
                    nd = ''
                else:
                    skip = '.hg'
                try:
                    entries = listdir(join(nd), stat=True, skip=skip)
                except OSError as inst:
                    if inst.errno in (errno.EACCES, errno.ENOENT):
                        match.bad(self.pathto(nd), inst.strerror)
                        continue
                    raise
                for f, kind, st in entries:
                    if normalizefile:
                        # even though f might be a directory, we're only
                        # interested in comparing it to files currently in the
                        # dmap -- therefore normalizefile is enough
                        nf = normalizefile(nd and (nd + "/" + f) or f, True,
                                           True)
                    else:
                        nf = nd and (nd + "/" + f) or f
                    if nf not in results:
                        if kind == dirkind:
                            if not ignore(nf):
                                if matchtdir:
                                    matchtdir(nf)
                                wadd(nf)
                            if nf in dmap and (matchalways or matchfn(nf)):
                                results[nf] = None
                        elif kind == regkind or kind == lnkkind:
                            if nf in dmap:
                                if matchalways or matchfn(nf):
                                    results[nf] = st
                            elif ((matchalways or matchfn(nf))
                                  and not ignore(nf)):
                                # unknown file -- normalize if necessary
                                if not alreadynormed:
                                    nf = normalize(nf, False, True)
                                results[nf] = st
                        elif nf in dmap and (matchalways or matchfn(nf)):
                            results[nf] = None

        for nd, d in work:
            # alreadynormed means that processwork doesn't have to do any
            # expensive directory normalization
            alreadynormed = not normalize or nd == d
            traverse([d], alreadynormed)

        for s in subrepos:
            del results[s]
        del results['.hg']

        # step 3: visit remaining files from dmap
        if not skipstep3 and not exact:
            # If a dmap file is not in results yet, it was either
            # a) not matching matchfn b) ignored, c) missing, or d) under a
            # symlink directory.
            if not results and matchalways:
                visit = dmap.keys()
            else:
                visit = [f for f in dmap if f not in results and matchfn(f)]
            visit.sort()

            if unknown:
                # unknown == True means we walked all dirs under the roots
                # that wasn't ignored, and everything that matched was stat'ed
                # and is already in results.
                # The rest must thus be ignored or under a symlink.
                audit_path = pathutil.pathauditor(self._root, cached=True)

                for nf in iter(visit):
                    # If a stat for the same file was already added with a
                    # different case, don't add one for this, since that would
                    # make it appear as if the file exists under both names
                    # on disk.
                    if (normalizefile and
                        normalizefile(nf, True, True) in results):
                        results[nf] = None
                    # Report ignored items in the dmap as long as they are not
                    # under a symlink directory.
                    elif audit_path.check(nf):
                        try:
                            results[nf] = lstat(join(nf))
                            # file was just ignored, no links, and exists
                        except OSError:
                            # file doesn't exist
                            results[nf] = None
                    else:
                        # It's either missing or under a symlink directory
                        # which we in this case report as missing
                        results[nf] = None
            else:
                # We may not have walked the full directory tree above,
                # so stat and check everything we missed.
                nf = iter(visit).next
                for st in util.statfiles([join(i) for i in visit]):
                    results[nf()] = st
        return results

    def status(self, match, subrepos, ignored, clean, unknown):
        '''Determine the status of the working copy relative to the
        dirstate and return a pair of (unsure, status), where status is of type
        scmutil.status and:

          unsure:
            files that might have been modified since the dirstate was
            written, but need to be read to be sure (size is the same
            but mtime differs)
          status.modified:
            files that have definitely been modified since the dirstate
            was written (different size or mode)
          status.clean:
            files that have definitely not been modified since the
            dirstate was written
        '''
        listignored, listclean, listunknown = ignored, clean, unknown
        lookup, modified, added, unknown, ignored = [], [], [], [], []
        removed, deleted, clean = [], [], []

        dmap = self._map
        ladd = lookup.append            # aka "unsure"
        madd = modified.append
        aadd = added.append
        uadd = unknown.append
        iadd = ignored.append
        radd = removed.append
        dadd = deleted.append
        cadd = clean.append
        mexact = match.exact
        dirignore = self._dirignore
        checkexec = self._checkexec
        copymap = self._copymap
        lastnormaltime = self._lastnormaltime

        # We need to do full walks when either
        # - we're listing all clean files, or
        # - match.traversedir does something, because match.traversedir should
        #   be called for every dir in the working dir
        full = listclean or match.traversedir is not None
        for fn, st in self.walk(match, subrepos, listunknown, listignored,
                                full=full).iteritems():
            if fn not in dmap:
                if (listignored or mexact(fn)) and dirignore(fn):
                    if listignored:
                        iadd(fn)
                else:
                    uadd(fn)
                continue

            # This is equivalent to 'state, mode, size, time = dmap[fn]' but not
            # written like that for performance reasons. dmap[fn] is not a
            # Python tuple in compiled builds. The CPython UNPACK_SEQUENCE
            # opcode has fast paths when the value to be unpacked is a tuple or
            # a list, but falls back to creating a full-fledged iterator in
            # general. That is much slower than simply accessing and storing the
            # tuple members one by one.
            t = dmap[fn]
            state = t[0]
            mode = t[1]
            size = t[2]
            time = t[3]

            if not st and state in "nma":
                dadd(fn)
            elif state == 'n':
                if (size >= 0 and
                    ((size != st.st_size and size != st.st_size & _rangemask)
                     or ((mode ^ st.st_mode) & 0o100 and checkexec))
                    or size == -2 # other parent
                    or fn in copymap):
                    madd(fn)
                elif time != st.st_mtime and time != st.st_mtime & _rangemask:
                    ladd(fn)
                elif st.st_mtime == lastnormaltime:
                    # fn may have just been marked as normal and it may have
                    # changed in the same second without changing its size.
                    # This can happen if we quickly do multiple commits.
                    # Force lookup, so we don't miss such a racy file change.
                    ladd(fn)
                elif listclean:
                    cadd(fn)
            elif state == 'm':
                madd(fn)
            elif state == 'a':
                aadd(fn)
            elif state == 'r':
                radd(fn)

        return (lookup, scmutil.status(modified, added, removed, deleted,
                                       unknown, ignored, clean))

    def matches(self, match):
        '''
        return files in the dirstate (in whatever state) filtered by match
        '''
        dmap = self._map
        if match.always():
            return dmap.keys()
        files = match.files()
        if match.isexact():
            # fast path -- filter the other way around, since typically files is
            # much smaller than dmap
            return [f for f in files if f in dmap]
        if match.prefix() and all(fn in dmap for fn in files):
            # fast path -- all the values are known to be files, so just return
            # that
            return list(files)
        return [f for f in dmap if match(f)]

    def _actualfilename(self, tr):
        if tr:
            return self._pendingfilename
        else:
            return self._filename

    def savebackup(self, tr, suffix='', prefix=''):
        '''Save current dirstate into backup file with suffix'''
        assert len(suffix) > 0 or len(prefix) > 0
        filename = self._actualfilename(tr)

        # use '_writedirstate' instead of 'write' to write changes certainly,
        # because the latter omits writing out if transaction is running.
        # output file will be used to create backup of dirstate at this point.
        self._writedirstate(self._opener(filename, "w", atomictemp=True,
                                         checkambig=True))

        if tr:
            # ensure that subsequent tr.writepending returns True for
            # changes written out above, even if dirstate is never
            # changed after this
            tr.addfilegenerator('dirstate', (self._filename,),
                                self._writedirstate, location='plain')

            # ensure that pending file written above is unlinked at
            # failure, even if tr.writepending isn't invoked until the
            # end of this transaction
            tr.registertmp(filename, location='plain')

        self._opener.write(prefix + self._filename + suffix,
                           self._opener.tryread(filename))

    def restorebackup(self, tr, suffix='', prefix=''):
        '''Restore dirstate by backup file with suffix'''
        assert len(suffix) > 0 or len(prefix) > 0
        # this "invalidate()" prevents "wlock.release()" from writing
        # changes of dirstate out after restoring from backup file
        self.invalidate()
        filename = self._actualfilename(tr)
        # using self._filename to avoid having "pending" in the backup filename
        self._opener.rename(prefix + self._filename + suffix, filename,
                            checkambig=True)

    def clearbackup(self, tr, suffix='', prefix=''):
        '''Clear backup file with suffix'''
        assert len(suffix) > 0 or len(prefix) > 0
        # using self._filename to avoid having "pending" in the backup filename
        self._opener.unlink(prefix + self._filename + suffix)