~bzr-pqm/bzr/bzr.dev

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
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
# Copyright (C) 2005, 2006 Canonical Ltd
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA

"""WorkingTree object and friends.

A WorkingTree represents the editable working copy of a branch.
Operations which represent the WorkingTree are also done here, 
such as renaming or adding files.  The WorkingTree has an inventory 
which is updated by these operations.  A commit produces a 
new revision based on the workingtree and its inventory.

At the moment every WorkingTree has its own branch.  Remote
WorkingTrees aren't supported.

To get a WorkingTree, call bzrdir.open_workingtree() or
WorkingTree.open(dir).
"""

MERGE_MODIFIED_HEADER_1 = "BZR merge-modified list format 1"
CONFLICT_HEADER_1 = "BZR conflict list format 1"

# TODO: Give the workingtree sole responsibility for the working inventory;
# remove the variable and references to it from the branch.  This may require
# updating the commit code so as to update the inventory within the working
# copy, and making sure there's only one WorkingTree for any directory on disk.
# At the moment they may alias the inventory and have old copies of it in
# memory.  (Now done? -- mbp 20060309)

from binascii import hexlify
import collections
from copy import deepcopy
from cStringIO import StringIO
import errno
import fnmatch
import os
import re
import stat
from time import time
import warnings

import bzrlib
from bzrlib import bzrdir, errors, ignores, osutils, urlutils
from bzrlib.atomicfile import AtomicFile
import bzrlib.branch
from bzrlib.conflicts import Conflict, ConflictList, CONFLICT_SUFFIXES
from bzrlib.decorators import needs_read_lock, needs_write_lock
from bzrlib.errors import (BzrCheckError,
                           BzrError,
                           ConflictFormatError,
                           WeaveRevisionNotPresent,
                           NotBranchError,
                           NoSuchFile,
                           NotVersionedError,
                           MergeModifiedFormatError,
                           UnsupportedOperation,
                           )
from bzrlib.inventory import InventoryEntry, Inventory
from bzrlib.lockable_files import LockableFiles, TransportLock
from bzrlib.lockdir import LockDir
from bzrlib.merge import merge_inner, transform_tree
from bzrlib.osutils import (
                            abspath,
                            compact_date,
                            file_kind,
                            isdir,
                            getcwd,
                            pathjoin,
                            pumpfile,
                            safe_unicode,
                            splitpath,
                            rand_chars,
                            normpath,
                            realpath,
                            relpath,
                            rename,
                            supports_executable,
                            )
from bzrlib.progress import DummyProgress, ProgressPhase
from bzrlib.revision import NULL_REVISION
from bzrlib.rio import RioReader, rio_file, Stanza
from bzrlib.symbol_versioning import (deprecated_passed,
        deprecated_method,
        deprecated_function,
        DEPRECATED_PARAMETER,
        zero_eight,
        )
from bzrlib.trace import mutter, note
from bzrlib.transform import build_tree
from bzrlib.transport import get_transport
from bzrlib.transport.local import LocalTransport
from bzrlib.textui import show_status
import bzrlib.tree
import bzrlib.ui
import bzrlib.xml5


# the regex removes any weird characters; we don't escape them 
# but rather just pull them out
_gen_file_id_re = re.compile(r'[^\w.]')
_gen_id_suffix = None
_gen_id_serial = 0


def _next_id_suffix():
    """Create a new file id suffix that is reasonably unique.
    
    On the first call we combine the current time with 64 bits of randomness
    to give a highly probably globally unique number. Then each call in the same
    process adds 1 to a serial number we append to that unique value.
    """
    # XXX TODO: change bzrlib.add.smart_add to call workingtree.add() rather 
    # than having to move the id randomness out of the inner loop like this.
    # XXX TODO: for the global randomness this uses we should add the thread-id
    # before the serial #.
    global _gen_id_suffix, _gen_id_serial
    if _gen_id_suffix is None:
        _gen_id_suffix = "-%s-%s-" % (compact_date(time()), rand_chars(16))
    _gen_id_serial += 1
    return _gen_id_suffix + str(_gen_id_serial)


def gen_file_id(name):
    """Return new file id for the basename 'name'.

    The uniqueness is supplied from _next_id_suffix.
    """
    # The real randomness is in the _next_id_suffix, the
    # rest of the identifier is just to be nice.
    # So we:
    # 1) Remove non-ascii word characters to keep the ids portable
    # 2) squash to lowercase, so the file id doesn't have to
    #    be escaped (case insensitive filesystems would bork for ids
    #    that only differred in case without escaping).
    # 3) truncate the filename to 20 chars. Long filenames also bork on some
    #    filesystems
    # 4) Removing starting '.' characters to prevent the file ids from
    #    being considered hidden.
    ascii_word_only = _gen_file_id_re.sub('', name.lower())
    short_no_dots = ascii_word_only.lstrip('.')[:20]
    return short_no_dots + _next_id_suffix()


def gen_root_id():
    """Return a new tree-root file id."""
    return gen_file_id('TREE_ROOT')


class TreeEntry(object):
    """An entry that implements the minimum interface used by commands.

    This needs further inspection, it may be better to have 
    InventoryEntries without ids - though that seems wrong. For now,
    this is a parallel hierarchy to InventoryEntry, and needs to become
    one of several things: decorates to that hierarchy, children of, or
    parents of it.
    Another note is that these objects are currently only used when there is
    no InventoryEntry available - i.e. for unversioned objects.
    Perhaps they should be UnversionedEntry et al. ? - RBC 20051003
    """
 
    def __eq__(self, other):
        # yes, this us ugly, TODO: best practice __eq__ style.
        return (isinstance(other, TreeEntry)
                and other.__class__ == self.__class__)
 
    def kind_character(self):
        return "???"


class TreeDirectory(TreeEntry):
    """See TreeEntry. This is a directory in a working tree."""

    def __eq__(self, other):
        return (isinstance(other, TreeDirectory)
                and other.__class__ == self.__class__)

    def kind_character(self):
        return "/"


class TreeFile(TreeEntry):
    """See TreeEntry. This is a regular file in a working tree."""

    def __eq__(self, other):
        return (isinstance(other, TreeFile)
                and other.__class__ == self.__class__)

    def kind_character(self):
        return ''


class TreeLink(TreeEntry):
    """See TreeEntry. This is a symlink in a working tree."""

    def __eq__(self, other):
        return (isinstance(other, TreeLink)
                and other.__class__ == self.__class__)

    def kind_character(self):
        return ''


class WorkingTree(bzrlib.tree.Tree):
    """Working copy tree.

    The inventory is held in the `Branch` working-inventory, and the
    files are in a directory on disk.

    It is possible for a `WorkingTree` to have a filename which is
    not listed in the Inventory and vice versa.
    """

    def __init__(self, basedir='.',
                 branch=DEPRECATED_PARAMETER,
                 _inventory=None,
                 _control_files=None,
                 _internal=False,
                 _format=None,
                 _bzrdir=None):
        """Construct a WorkingTree for basedir.

        If the branch is not supplied, it is opened automatically.
        If the branch is supplied, it must be the branch for this basedir.
        (branch.base is not cross checked, because for remote branches that
        would be meaningless).
        """
        self._format = _format
        self.bzrdir = _bzrdir
        if not _internal:
            # not created via open etc.
            warnings.warn("WorkingTree() is deprecated as of bzr version 0.8. "
                 "Please use bzrdir.open_workingtree or WorkingTree.open().",
                 DeprecationWarning,
                 stacklevel=2)
            wt = WorkingTree.open(basedir)
            self._branch = wt.branch
            self.basedir = wt.basedir
            self._control_files = wt._control_files
            self._hashcache = wt._hashcache
            self._set_inventory(wt._inventory)
            self._format = wt._format
            self.bzrdir = wt.bzrdir
        from bzrlib.hashcache import HashCache
        from bzrlib.trace import note, mutter
        assert isinstance(basedir, basestring), \
            "base directory %r is not a string" % basedir
        basedir = safe_unicode(basedir)
        mutter("opening working tree %r", basedir)
        if deprecated_passed(branch):
            if not _internal:
                warnings.warn("WorkingTree(..., branch=XXX) is deprecated as of bzr 0.8."
                     " Please use bzrdir.open_workingtree() or"
                     " WorkingTree.open().",
                     DeprecationWarning,
                     stacklevel=2
                     )
            self._branch = branch
        else:
            self._branch = self.bzrdir.open_branch()
        self.basedir = realpath(basedir)
        # if branch is at our basedir and is a format 6 or less
        if isinstance(self._format, WorkingTreeFormat2):
            # share control object
            self._control_files = self.branch.control_files
        else:
            # assume all other formats have their own control files.
            assert isinstance(_control_files, LockableFiles), \
                    "_control_files must be a LockableFiles, not %r" \
                    % _control_files
            self._control_files = _control_files
        # update the whole cache up front and write to disk if anything changed;
        # in the future we might want to do this more selectively
        # two possible ways offer themselves : in self._unlock, write the cache
        # if needed, or, when the cache sees a change, append it to the hash
        # cache file, and have the parser take the most recent entry for a
        # given path only.
        cache_filename = self.bzrdir.get_workingtree_transport(None).local_abspath('stat-cache')
        hc = self._hashcache = HashCache(basedir, cache_filename, self._control_files._file_mode)
        hc.read()
        # is this scan needed ? it makes things kinda slow.
        #hc.scan()

        if hc.needs_write:
            mutter("write hc")
            hc.write()

        if _inventory is None:
            self._set_inventory(self.read_working_inventory())
        else:
            self._set_inventory(_inventory)

    branch = property(
        fget=lambda self: self._branch,
        doc="""The branch this WorkingTree is connected to.

            This cannot be set - it is reflective of the actual disk structure
            the working tree has been constructed from.
            """)

    def break_lock(self):
        """Break a lock if one is present from another instance.

        Uses the ui factory to ask for confirmation if the lock may be from
        an active process.

        This will probe the repository for its lock as well.
        """
        self._control_files.break_lock()
        self.branch.break_lock()

    def _set_inventory(self, inv):
        assert inv.root is not None
        self._inventory = inv
        self.path2id = self._inventory.path2id

    def is_control_filename(self, filename):
        """True if filename is the name of a control file in this tree.
        
        :param filename: A filename within the tree. This is a relative path
        from the root of this tree.

        This is true IF and ONLY IF the filename is part of the meta data
        that bzr controls in this tree. I.E. a random .bzr directory placed
        on disk will not be a control file for this tree.
        """
        return self.bzrdir.is_control_filename(filename)

    @staticmethod
    def open(path=None, _unsupported=False):
        """Open an existing working tree at path.

        """
        if path is None:
            path = os.path.getcwdu()
        control = bzrdir.BzrDir.open(path, _unsupported)
        return control.open_workingtree(_unsupported)
        
    @staticmethod
    def open_containing(path=None):
        """Open an existing working tree which has its root about path.
        
        This probes for a working tree at path and searches upwards from there.

        Basically we keep looking up until we find the control directory or
        run into /.  If there isn't one, raises NotBranchError.
        TODO: give this a new exception.
        If there is one, it is returned, along with the unused portion of path.

        :return: The WorkingTree that contains 'path', and the rest of path
        """
        if path is None:
            path = osutils.getcwd()
        control, relpath = bzrdir.BzrDir.open_containing(path)

        return control.open_workingtree(), relpath

    @staticmethod
    def open_downlevel(path=None):
        """Open an unsupported working tree.

        Only intended for advanced situations like upgrading part of a bzrdir.
        """
        return WorkingTree.open(path, _unsupported=True)

    def __iter__(self):
        """Iterate through file_ids for this tree.

        file_ids are in a WorkingTree if they are in the working inventory
        and the working file exists.
        """
        inv = self._inventory
        for path, ie in inv.iter_entries():
            if osutils.lexists(self.abspath(path)):
                yield ie.file_id

    def __repr__(self):
        return "<%s of %s>" % (self.__class__.__name__,
                               getattr(self, 'basedir', None))

    def abspath(self, filename):
        return pathjoin(self.basedir, filename)
    
    def basis_tree(self):
        """Return RevisionTree for the current last revision."""
        revision_id = self.last_revision()
        if revision_id is not None:
            try:
                xml = self.read_basis_inventory()
                inv = bzrlib.xml5.serializer_v5.read_inventory_from_string(xml)
                inv.root.revision = revision_id
            except NoSuchFile:
                inv = None
            if inv is not None and inv.revision_id == revision_id:
                return bzrlib.tree.RevisionTree(self.branch.repository, inv,
                                                revision_id)
        # FIXME? RBC 20060403 should we cache the inventory here ?
        return self.branch.repository.revision_tree(revision_id)

    @staticmethod
    @deprecated_method(zero_eight)
    def create(branch, directory):
        """Create a workingtree for branch at directory.

        If existing_directory already exists it must have a .bzr directory.
        If it does not exist, it will be created.

        This returns a new WorkingTree object for the new checkout.

        TODO FIXME RBC 20060124 when we have checkout formats in place this
        should accept an optional revisionid to checkout [and reject this if
        checking out into the same dir as a pre-checkout-aware branch format.]

        XXX: When BzrDir is present, these should be created through that 
        interface instead.
        """
        warnings.warn('delete WorkingTree.create', stacklevel=3)
        transport = get_transport(directory)
        if branch.bzrdir.root_transport.base == transport.base:
            # same dir 
            return branch.bzrdir.create_workingtree()
        # different directory, 
        # create a branch reference
        # and now a working tree.
        raise NotImplementedError
 
    @staticmethod
    @deprecated_method(zero_eight)
    def create_standalone(directory):
        """Create a checkout and a branch and a repo at directory.

        Directory must exist and be empty.

        please use BzrDir.create_standalone_workingtree
        """
        return bzrdir.BzrDir.create_standalone_workingtree(directory)

    def relpath(self, path):
        """Return the local path portion from a given path.
        
        The path may be absolute or relative. If its a relative path it is 
        interpreted relative to the python current working directory.
        """
        return relpath(self.basedir, path)

    def has_filename(self, filename):
        return osutils.lexists(self.abspath(filename))

    def get_file(self, file_id):
        return self.get_file_byname(self.id2path(file_id))

    def get_file_text(self, file_id):
        return self.get_file(file_id).read()

    def get_file_byname(self, filename):
        return file(self.abspath(filename), 'rb')

    def get_parent_ids(self):
        """See Tree.get_parent_ids.
        
        This implementation reads the pending merges list and last_revision
        value and uses that to decide what the parents list should be.
        """
        last_rev = self.last_revision()
        if last_rev is None:
            parents = []
        else:
            parents = [last_rev]
        other_parents = self.pending_merges()
        return parents + other_parents

    def get_root_id(self):
        """Return the id of this trees root"""
        inv = self.read_working_inventory()
        return inv.root.file_id
        
    def _get_store_filename(self, file_id):
        ## XXX: badly named; this is not in the store at all
        return self.abspath(self.id2path(file_id))

    @needs_read_lock
    def clone(self, to_bzrdir, revision_id=None, basis=None):
        """Duplicate this working tree into to_bzr, including all state.
        
        Specifically modified files are kept as modified, but
        ignored and unknown files are discarded.

        If you want to make a new line of development, see bzrdir.sprout()

        revision
            If not None, the cloned tree will have its last revision set to 
            revision, and and difference between the source trees last revision
            and this one merged in.

        basis
            If not None, a closer copy of a tree which may have some files in
            common, and which file content should be preferentially copied from.
        """
        # assumes the target bzr dir format is compatible.
        result = self._format.initialize(to_bzrdir)
        self.copy_content_into(result, revision_id)
        return result

    @needs_read_lock
    def copy_content_into(self, tree, revision_id=None):
        """Copy the current content and user files of this tree into tree."""
        if revision_id is None:
            transform_tree(tree, self)
        else:
            # TODO now merge from tree.last_revision to revision
            transform_tree(tree, self)
            tree.set_last_revision(revision_id)

    @needs_write_lock
    def commit(self, message=None, revprops=None, *args, **kwargs):
        # avoid circular imports
        from bzrlib.commit import Commit
        if revprops is None:
            revprops = {}
        if not 'branch-nick' in revprops:
            revprops['branch-nick'] = self.branch.nick
        # args for wt.commit start at message from the Commit.commit method,
        # but with branch a kwarg now, passing in args as is results in the
        #message being used for the branch
        args = (DEPRECATED_PARAMETER, message, ) + args
        committed_id = Commit().commit( working_tree=self, revprops=revprops,
            *args, **kwargs)
        self._set_inventory(self.read_working_inventory())
        return committed_id

    def id2abspath(self, file_id):
        return self.abspath(self.id2path(file_id))

    def has_id(self, file_id):
        # files that have been deleted are excluded
        inv = self._inventory
        if not inv.has_id(file_id):
            return False
        path = inv.id2path(file_id)
        return osutils.lexists(self.abspath(path))

    def has_or_had_id(self, file_id):
        if file_id == self.inventory.root.file_id:
            return True
        return self.inventory.has_id(file_id)

    __contains__ = has_id

    def get_file_size(self, file_id):
        return os.path.getsize(self.id2abspath(file_id))

    @needs_read_lock
    def get_file_sha1(self, file_id, path=None):
        if not path:
            path = self._inventory.id2path(file_id)
        return self._hashcache.get_sha1(path)

    def get_file_mtime(self, file_id, path=None):
        if not path:
            path = self._inventory.id2path(file_id)
        return os.lstat(self.abspath(path)).st_mtime

    if not supports_executable():
        def is_executable(self, file_id, path=None):
            return self._inventory[file_id].executable
    else:
        def is_executable(self, file_id, path=None):
            if not path:
                path = self._inventory.id2path(file_id)
            mode = os.lstat(self.abspath(path)).st_mode
            return bool(stat.S_ISREG(mode) and stat.S_IEXEC & mode)

    @needs_write_lock
    def add(self, files, ids=None):
        """Make files versioned.

        Note that the command line normally calls smart_add instead,
        which can automatically recurse.

        This adds the files to the inventory, so that they will be
        recorded by the next commit.

        files
            List of paths to add, relative to the base of the tree.

        ids
            If set, use these instead of automatically generated ids.
            Must be the same length as the list of files, but may
            contain None for ids that are to be autogenerated.

        TODO: Perhaps have an option to add the ids even if the files do
              not (yet) exist.

        TODO: Perhaps callback with the ids and paths as they're added.
        """
        # TODO: Re-adding a file that is removed in the working copy
        # should probably put it back with the previous ID.
        if isinstance(files, basestring):
            assert(ids is None or isinstance(ids, basestring))
            files = [files]
            if ids is not None:
                ids = [ids]

        if ids is None:
            ids = [None] * len(files)
        else:
            assert(len(ids) == len(files))

        inv = self.read_working_inventory()
        for f,file_id in zip(files, ids):
            if self.is_control_filename(f):
                raise errors.ForbiddenControlFileError(filename=f)

            fp = splitpath(f)

            if len(fp) == 0:
                raise BzrError("cannot add top-level %r" % f)

            fullpath = normpath(self.abspath(f))
            try:
                kind = file_kind(fullpath)
            except OSError, e:
                if e.errno == errno.ENOENT:
                    raise NoSuchFile(fullpath)
            if not InventoryEntry.versionable_kind(kind):
                raise errors.BadFileKindError(filename=f, kind=kind)
            if file_id is None:
                inv.add_path(f, kind=kind)
            else:
                inv.add_path(f, kind=kind, file_id=file_id)

        self._write_inventory(inv)

    @needs_write_lock
    def add_pending_merge(self, *revision_ids):
        # TODO: Perhaps should check at this point that the
        # history of the revision is actually present?
        p = self.pending_merges()
        updated = False
        for rev_id in revision_ids:
            if rev_id in p:
                continue
            p.append(rev_id)
            updated = True
        if updated:
            self.set_pending_merges(p)

    @needs_read_lock
    def pending_merges(self):
        """Return a list of pending merges.

        These are revisions that have been merged into the working
        directory but not yet committed.
        """
        try:
            merges_file = self._control_files.get_utf8('pending-merges')
        except NoSuchFile:
            return []
        p = []
        for l in merges_file.readlines():
            p.append(l.rstrip('\n'))
        return p

    @needs_write_lock
    def set_pending_merges(self, rev_list):
        self._control_files.put_utf8('pending-merges', '\n'.join(rev_list))

    @needs_write_lock
    def set_merge_modified(self, modified_hashes):
        def iter_stanzas():
            for file_id, hash in modified_hashes.iteritems():
                yield Stanza(file_id=file_id, hash=hash)
        self._put_rio('merge-hashes', iter_stanzas(), MERGE_MODIFIED_HEADER_1)

    @needs_write_lock
    def _put_rio(self, filename, stanzas, header):
        my_file = rio_file(stanzas, header)
        self._control_files.put(filename, my_file)

    @needs_read_lock
    def merge_modified(self):
        try:
            hashfile = self._control_files.get('merge-hashes')
        except NoSuchFile:
            return {}
        merge_hashes = {}
        try:
            if hashfile.next() != MERGE_MODIFIED_HEADER_1 + '\n':
                raise MergeModifiedFormatError()
        except StopIteration:
            raise MergeModifiedFormatError()
        for s in RioReader(hashfile):
            file_id = s.get("file_id")
            if file_id not in self.inventory:
                continue
            hash = s.get("hash")
            if hash == self.get_file_sha1(file_id):
                merge_hashes[file_id] = hash
        return merge_hashes

    def get_symlink_target(self, file_id):
        return os.readlink(self.id2abspath(file_id))

    def file_class(self, filename):
        if self.path2id(filename):
            return 'V'
        elif self.is_ignored(filename):
            return 'I'
        else:
            return '?'

    def list_files(self):
        """Recursively list all files as (path, class, kind, id, entry).

        Lists, but does not descend into unversioned directories.

        This does not include files that have been deleted in this
        tree.

        Skips the control directory.
        """
        inv = self._inventory
        # Convert these into local objects to save lookup times
        pathjoin = osutils.pathjoin
        file_kind = osutils.file_kind

        # transport.base ends in a slash, we want the piece
        # between the last two slashes
        transport_base_dir = self.bzrdir.transport.base.rsplit('/', 2)[1]

        fk_entries = {'directory':TreeDirectory, 'file':TreeFile, 'symlink':TreeLink}

        # directory file_id, relative path, absolute path, reverse sorted children
        children = os.listdir(self.basedir)
        children.sort()
        # jam 20060527 The kernel sized tree seems equivalent whether we 
        # use a deque and popleft to keep them sorted, or if we use a plain
        # list and just reverse() them.
        children = collections.deque(children)
        stack = [(inv.root.file_id, u'', self.basedir, children)]
        while stack:
            from_dir_id, from_dir_relpath, from_dir_abspath, children = stack[-1]

            while children:
                f = children.popleft()
                ## TODO: If we find a subdirectory with its own .bzr
                ## directory, then that is a separate tree and we
                ## should exclude it.

                # the bzrdir for this tree
                if transport_base_dir == f:
                    continue

                # we know that from_dir_relpath and from_dir_abspath never end in a slash
                # and 'f' doesn't begin with one, we can do a string op, rather
                # than the checks of pathjoin(), all relative paths will have an extra slash
                # at the beginning
                fp = from_dir_relpath + '/' + f

                # absolute path
                fap = from_dir_abspath + '/' + f
                
                f_ie = inv.get_child(from_dir_id, f)
                if f_ie:
                    c = 'V'
                elif self.is_ignored(fp[1:]):
                    c = 'I'
                else:
                    # we may not have found this file, because of a unicode issue
                    f_norm, can_access = osutils.normalized_filename(f)
                    if f == f_norm or not can_access:
                        # No change, so treat this file normally
                        c = '?'
                    else:
                        # this file can be accessed by a normalized path
                        # check again if it is versioned
                        # these lines are repeated here for performance
                        f = f_norm
                        fp = from_dir_relpath + '/' + f
                        fap = from_dir_abspath + '/' + f
                        f_ie = inv.get_child(from_dir_id, f)
                        if f_ie:
                            c = 'V'
                        elif self.is_ignored(fp[1:]):
                            c = 'I'
                        else:
                            c = '?'

                fk = file_kind(fap)

                if f_ie:
                    if f_ie.kind != fk:
                        raise BzrCheckError("file %r entered as kind %r id %r, "
                                            "now of kind %r"
                                            % (fap, f_ie.kind, f_ie.file_id, fk))

                # make a last minute entry
                if f_ie:
                    yield fp[1:], c, fk, f_ie.file_id, f_ie
                else:
                    try:
                        yield fp[1:], c, fk, None, fk_entries[fk]()
                    except KeyError:
                        yield fp[1:], c, fk, None, TreeEntry()
                    continue
                
                if fk != 'directory':
                    continue

                # But do this child first
                new_children = os.listdir(fap)
                new_children.sort()
                new_children = collections.deque(new_children)
                stack.append((f_ie.file_id, fp, fap, new_children))
                # Break out of inner loop, so that we start outer loop with child
                break
            else:
                # if we finished all children, pop it off the stack
                stack.pop()


    @needs_write_lock
    def move(self, from_paths, to_name):
        """Rename files.

        to_name must exist in the inventory.

        If to_name exists and is a directory, the files are moved into
        it, keeping their old names.  

        Note that to_name is only the last component of the new name;
        this doesn't change the directory.

        This returns a list of (from_path, to_path) pairs for each
        entry that is moved.
        """
        result = []
        ## TODO: Option to move IDs only
        assert not isinstance(from_paths, basestring)
        inv = self.inventory
        to_abs = self.abspath(to_name)
        if not isdir(to_abs):
            raise BzrError("destination %r is not a directory" % to_abs)
        if not self.has_filename(to_name):
            raise BzrError("destination %r not in working directory" % to_abs)
        to_dir_id = inv.path2id(to_name)
        if to_dir_id == None and to_name != '':
            raise BzrError("destination %r is not a versioned directory" % to_name)
        to_dir_ie = inv[to_dir_id]
        if to_dir_ie.kind != 'directory':
            raise BzrError("destination %r is not a directory" % to_abs)

        to_idpath = inv.get_idpath(to_dir_id)

        for f in from_paths:
            if not self.has_filename(f):
                raise BzrError("%r does not exist in working tree" % f)
            f_id = inv.path2id(f)
            if f_id == None:
                raise BzrError("%r is not versioned" % f)
            name_tail = splitpath(f)[-1]
            dest_path = pathjoin(to_name, name_tail)
            if self.has_filename(dest_path):
                raise BzrError("destination %r already exists" % dest_path)
            if f_id in to_idpath:
                raise BzrError("can't move %r to a subdirectory of itself" % f)

        # OK, so there's a race here, it's possible that someone will
        # create a file in this interval and then the rename might be
        # left half-done.  But we should have caught most problems.
        orig_inv = deepcopy(self.inventory)
        try:
            for f in from_paths:
                name_tail = splitpath(f)[-1]
                dest_path = pathjoin(to_name, name_tail)
                result.append((f, dest_path))
                inv.rename(inv.path2id(f), to_dir_id, name_tail)
                try:
                    rename(self.abspath(f), self.abspath(dest_path))
                except OSError, e:
                    raise BzrError("failed to rename %r to %r: %s" %
                                   (f, dest_path, e[1]),
                            ["rename rolled back"])
        except:
            # restore the inventory on error
            self._set_inventory(orig_inv)
            raise
        self._write_inventory(inv)
        return result

    @needs_write_lock
    def rename_one(self, from_rel, to_rel):
        """Rename one file.

        This can change the directory or the filename or both.
        """
        inv = self.inventory
        if not self.has_filename(from_rel):
            raise BzrError("can't rename: old working file %r does not exist" % from_rel)
        if self.has_filename(to_rel):
            raise BzrError("can't rename: new working file %r already exists" % to_rel)

        file_id = inv.path2id(from_rel)
        if file_id == None:
            raise BzrError("can't rename: old name %r is not versioned" % from_rel)

        entry = inv[file_id]
        from_parent = entry.parent_id
        from_name = entry.name
        
        if inv.path2id(to_rel):
            raise BzrError("can't rename: new name %r is already versioned" % to_rel)

        to_dir, to_tail = os.path.split(to_rel)
        to_dir_id = inv.path2id(to_dir)
        if to_dir_id == None and to_dir != '':
            raise BzrError("can't determine destination directory id for %r" % to_dir)

        mutter("rename_one:")
        mutter("  file_id    {%s}" % file_id)
        mutter("  from_rel   %r" % from_rel)
        mutter("  to_rel     %r" % to_rel)
        mutter("  to_dir     %r" % to_dir)
        mutter("  to_dir_id  {%s}" % to_dir_id)

        inv.rename(file_id, to_dir_id, to_tail)

        from_abs = self.abspath(from_rel)
        to_abs = self.abspath(to_rel)
        try:
            rename(from_abs, to_abs)
        except OSError, e:
            inv.rename(file_id, from_parent, from_name)
            raise BzrError("failed to rename %r to %r: %s"
                    % (from_abs, to_abs, e[1]),
                    ["rename rolled back"])
        self._write_inventory(inv)

    @needs_read_lock
    def unknowns(self):
        """Return all unknown files.

        These are files in the working directory that are not versioned or
        control files or ignored.
        """
        for subp in self.extras():
            if not self.is_ignored(subp):
                yield subp

    @deprecated_method(zero_eight)
    def iter_conflicts(self):
        """List all files in the tree that have text or content conflicts.
        DEPRECATED.  Use conflicts instead."""
        return self._iter_conflicts()

    def _iter_conflicts(self):
        conflicted = set()
        for info in self.list_files():
            path = info[0]
            stem = get_conflicted_stem(path)
            if stem is None:
                continue
            if stem not in conflicted:
                conflicted.add(stem)
                yield stem

    @needs_write_lock
    def pull(self, source, overwrite=False, stop_revision=None):
        top_pb = bzrlib.ui.ui_factory.nested_progress_bar()
        source.lock_read()
        try:
            pp = ProgressPhase("Pull phase", 2, top_pb)
            pp.next_phase()
            old_revision_history = self.branch.revision_history()
            basis_tree = self.basis_tree()
            count = self.branch.pull(source, overwrite, stop_revision)
            new_revision_history = self.branch.revision_history()
            if new_revision_history != old_revision_history:
                pp.next_phase()
                if len(old_revision_history):
                    other_revision = old_revision_history[-1]
                else:
                    other_revision = None
                repository = self.branch.repository
                pb = bzrlib.ui.ui_factory.nested_progress_bar()
                try:
                    merge_inner(self.branch,
                                self.branch.basis_tree(),
                                basis_tree, 
                                this_tree=self, 
                                pb=pb)
                finally:
                    pb.finished()
                self.set_last_revision(self.branch.last_revision())
            return count
        finally:
            source.unlock()
            top_pb.finished()

    def extras(self):
        """Yield all unknown files in this WorkingTree.

        If there are any unknown directories then only the directory is
        returned, not all its children.  But if there are unknown files
        under a versioned subdirectory, they are returned.

        Currently returned depth-first, sorted by name within directories.
        """
        ## TODO: Work from given directory downwards
        for path, dir_entry in self.inventory.directories():
            # mutter("search for unknowns in %r", path)
            dirabs = self.abspath(path)
            if not isdir(dirabs):
                # e.g. directory deleted
                continue

            fl = []
            for subf in os.listdir(dirabs):
                if subf == '.bzr':
                    continue
                if subf not in dir_entry.children:
                    subf_norm, can_access = osutils.normalized_filename(subf)
                    if subf_norm != subf and can_access:
                        if subf_norm not in dir_entry.children:
                            fl.append(subf_norm)
                    else:
                        fl.append(subf)
            
            fl.sort()
            for subf in fl:
                subp = pathjoin(path, subf)
                yield subp

    def _translate_ignore_rule(self, rule):
        """Translate a single ignore rule to a regex.

        There are two types of ignore rules.  Those that do not contain a / are
        matched against the tail of the filename (that is, they do not care
        what directory the file is in.)  Rules which do contain a slash must
        match the entire path.  As a special case, './' at the start of the
        string counts as a slash in the string but is removed before matching
        (e.g. ./foo.c, ./src/foo.c)

        :return: The translated regex.
        """
        if rule[:2] in ('./', '.\\'):
            # rootdir rule
            result = fnmatch.translate(rule[2:])
        elif '/' in rule or '\\' in rule:
            # path prefix 
            result = fnmatch.translate(rule)
        else:
            # default rule style.
            result = "(?:.*/)?(?!.*/)" + fnmatch.translate(rule)
        assert result[-1] == '$', "fnmatch.translate did not add the expected $"
        return "(" + result + ")"

    def _combine_ignore_rules(self, rules):
        """Combine a list of ignore rules into a single regex object.

        Each individual rule is combined with | to form a big regex, which then
        has $ added to it to form something like ()|()|()$. The group index for
        each subregex's outermost group is placed in a dictionary mapping back 
        to the rule. This allows quick identification of the matching rule that
        triggered a match.
        :return: a list of the compiled regex and the matching-group index 
        dictionaries. We return a list because python complains if you try to 
        combine more than 100 regexes.
        """
        result = []
        groups = {}
        next_group = 0
        translated_rules = []
        for rule in rules:
            translated_rule = self._translate_ignore_rule(rule)
            compiled_rule = re.compile(translated_rule)
            groups[next_group] = rule
            next_group += compiled_rule.groups
            translated_rules.append(translated_rule)
            if next_group == 99:
                result.append((re.compile("|".join(translated_rules)), groups))
                groups = {}
                next_group = 0
                translated_rules = []
        if len(translated_rules):
            result.append((re.compile("|".join(translated_rules)), groups))
        return result

    def ignored_files(self):
        """Yield list of PATH, IGNORE_PATTERN"""
        for subp in self.extras():
            pat = self.is_ignored(subp)
            if pat != None:
                yield subp, pat

    def get_ignore_list(self):
        """Return list of ignore patterns.

        Cached in the Tree object after the first call.
        """
        ignoreset = getattr(self, '_ignoreset', None)
        if ignoreset is not None:
            return ignoreset

        ignore_globs = set(bzrlib.DEFAULT_IGNORE)
        ignore_globs.update(ignores.get_runtime_ignores())

        ignore_globs.update(ignores.get_user_ignores())

        if self.has_filename(bzrlib.IGNORE_FILENAME):
            f = self.get_file_byname(bzrlib.IGNORE_FILENAME)
            try:
                ignore_globs.update(ignores.parse_ignore_file(f))
            finally:
                f.close()

        self._ignoreset = ignore_globs
        self._ignore_regex = self._combine_ignore_rules(ignore_globs)
        return ignore_globs

    def _get_ignore_rules_as_regex(self):
        """Return a regex of the ignore rules and a mapping dict.

        :return: (ignore rules compiled regex, dictionary mapping rule group 
        indices to original rule.)
        """
        if getattr(self, '_ignoreset', None) is None:
            self.get_ignore_list()
        return self._ignore_regex

    def is_ignored(self, filename):
        r"""Check whether the filename matches an ignore pattern.

        Patterns containing '/' or '\' need to match the whole path;
        others match against only the last component.

        If the file is ignored, returns the pattern which caused it to
        be ignored, otherwise None.  So this can simply be used as a
        boolean if desired."""

        # TODO: Use '**' to match directories, and other extended
        # globbing stuff from cvs/rsync.

        # XXX: fnmatch is actually not quite what we want: it's only
        # approximately the same as real Unix fnmatch, and doesn't
        # treat dotfiles correctly and allows * to match /.
        # Eventually it should be replaced with something more
        # accurate.
    
        rules = self._get_ignore_rules_as_regex()
        for regex, mapping in rules:
            match = regex.match(filename)
            if match is not None:
                # one or more of the groups in mapping will have a non-None group 
                # match.
                groups = match.groups()
                rules = [mapping[group] for group in 
                    mapping if groups[group] is not None]
                return rules[0]
        return None

    def kind(self, file_id):
        return file_kind(self.id2abspath(file_id))

    @needs_read_lock
    def last_revision(self):
        """Return the last revision id of this working tree.

        In early branch formats this was == the branch last_revision,
        but that cannot be relied upon - for working tree operations,
        always use tree.last_revision().
        """
        return self.branch.last_revision()

    def is_locked(self):
        return self._control_files.is_locked()

    def lock_read(self):
        """See Branch.lock_read, and WorkingTree.unlock."""
        self.branch.lock_read()
        try:
            return self._control_files.lock_read()
        except:
            self.branch.unlock()
            raise

    def lock_write(self):
        """See Branch.lock_write, and WorkingTree.unlock."""
        self.branch.lock_write()
        try:
            return self._control_files.lock_write()
        except:
            self.branch.unlock()
            raise

    def get_physical_lock_status(self):
        return self._control_files.get_physical_lock_status()

    def _basis_inventory_name(self):
        return 'basis-inventory'

    @needs_write_lock
    def set_last_revision(self, new_revision):
        """Change the last revision in the working tree."""
        if self._change_last_revision(new_revision):
            self._cache_basis_inventory(new_revision)

    def _change_last_revision(self, new_revision):
        """Template method part of set_last_revision to perform the change.
        
        This is used to allow WorkingTree3 instances to not affect branch
        when their last revision is set.
        """
        if new_revision is None:
            self.branch.set_revision_history([])
            return False
        try:
            self.branch.generate_revision_history(new_revision)
        except errors.NoSuchRevision:
            # not present in the repo - dont try to set it deeper than the tip
            self.branch.set_revision_history([new_revision])
        return True

    def _cache_basis_inventory(self, new_revision):
        """Cache new_revision as the basis inventory."""
        # TODO: this should allow the ready-to-use inventory to be passed in,
        # as commit already has that ready-to-use [while the format is the
        # same, that is].
        try:
            # this double handles the inventory - unpack and repack - 
            # but is easier to understand. We can/should put a conditional
            # in here based on whether the inventory is in the latest format
            # - perhaps we should repack all inventories on a repository
            # upgrade ?
            # the fast path is to copy the raw xml from the repository. If the
            # xml contains 'revision_id="', then we assume the right 
            # revision_id is set. We must check for this full string, because a
            # root node id can legitimately look like 'revision_id' but cannot
            # contain a '"'.
            xml = self.branch.repository.get_inventory_xml(new_revision)
            if not 'revision_id="' in xml.split('\n', 1)[0]:
                inv = self.branch.repository.deserialise_inventory(
                    new_revision, xml)
                inv.revision_id = new_revision
                xml = bzrlib.xml5.serializer_v5.write_inventory_to_string(inv)
            assert isinstance(xml, str), 'serialised xml must be bytestring.'
            path = self._basis_inventory_name()
            sio = StringIO(xml)
            self._control_files.put(path, sio)
        except (errors.NoSuchRevision, errors.RevisionNotPresent):
            pass

    def read_basis_inventory(self):
        """Read the cached basis inventory."""
        path = self._basis_inventory_name()
        return self._control_files.get(path).read()
        
    @needs_read_lock
    def read_working_inventory(self):
        """Read the working inventory."""
        # ElementTree does its own conversion from UTF-8, so open in
        # binary.
        result = bzrlib.xml5.serializer_v5.read_inventory(
            self._control_files.get('inventory'))
        self._set_inventory(result)
        return result

    @needs_write_lock
    def remove(self, files, verbose=False, to_file=None):
        """Remove nominated files from the working inventory..

        This does not remove their text.  This does not run on XXX on what? RBC

        TODO: Refuse to remove modified files unless --force is given?

        TODO: Do something useful with directories.

        TODO: Should this remove the text or not?  Tough call; not
        removing may be useful and the user can just use use rm, and
        is the opposite of add.  Removing it is consistent with most
        other tools.  Maybe an option.
        """
        ## TODO: Normalize names
        ## TODO: Remove nested loops; better scalability
        if isinstance(files, basestring):
            files = [files]

        inv = self.inventory

        # do this before any modifications
        for f in files:
            fid = inv.path2id(f)
            if not fid:
                # TODO: Perhaps make this just a warning, and continue?
                # This tends to happen when 
                raise NotVersionedError(path=f)
            if verbose:
                # having remove it, it must be either ignored or unknown
                if self.is_ignored(f):
                    new_status = 'I'
                else:
                    new_status = '?'
                show_status(new_status, inv[fid].kind, f, to_file=to_file)
            del inv[fid]

        self._write_inventory(inv)

    @needs_write_lock
    def revert(self, filenames, old_tree=None, backups=True, 
               pb=DummyProgress()):
        from transform import revert
        from conflicts import resolve
        if old_tree is None:
            old_tree = self.basis_tree()
        conflicts = revert(self, old_tree, filenames, backups, pb)
        if not len(filenames):
            self.set_pending_merges([])
            resolve(self)
        else:
            resolve(self, filenames, ignore_misses=True)
        return conflicts

    # XXX: This method should be deprecated in favour of taking in a proper
    # new Inventory object.
    @needs_write_lock
    def set_inventory(self, new_inventory_list):
        from bzrlib.inventory import (Inventory,
                                      InventoryDirectory,
                                      InventoryEntry,
                                      InventoryFile,
                                      InventoryLink)
        inv = Inventory(self.get_root_id())
        for path, file_id, parent, kind in new_inventory_list:
            name = os.path.basename(path)
            if name == "":
                continue
            # fixme, there should be a factory function inv,add_?? 
            if kind == 'directory':
                inv.add(InventoryDirectory(file_id, name, parent))
            elif kind == 'file':
                inv.add(InventoryFile(file_id, name, parent))
            elif kind == 'symlink':
                inv.add(InventoryLink(file_id, name, parent))
            else:
                raise BzrError("unknown kind %r" % kind)
        self._write_inventory(inv)

    @needs_write_lock
    def set_root_id(self, file_id):
        """Set the root id for this tree."""
        inv = self.read_working_inventory()
        orig_root_id = inv.root.file_id
        del inv._byid[inv.root.file_id]
        inv.root.file_id = file_id
        inv._byid[inv.root.file_id] = inv.root
        for fid in inv:
            entry = inv[fid]
            if entry.parent_id == orig_root_id:
                entry.parent_id = inv.root.file_id
        self._write_inventory(inv)

    def unlock(self):
        """See Branch.unlock.
        
        WorkingTree locking just uses the Branch locking facilities.
        This is current because all working trees have an embedded branch
        within them. IF in the future, we were to make branch data shareable
        between multiple working trees, i.e. via shared storage, then we 
        would probably want to lock both the local tree, and the branch.
        """
        raise NotImplementedError(self.unlock)

    @needs_write_lock
    def update(self):
        """Update a working tree along its branch.

        This will update the branch if its bound too, which means we have multiple trees involved:
        The new basis tree of the master.
        The old basis tree of the branch.
        The old basis tree of the working tree.
        The current working tree state.
        pathologically all three may be different, and non ancestors of each other.
        Conceptually we want to:
        Preserve the wt.basis->wt.state changes
        Transform the wt.basis to the new master basis.
        Apply a merge of the old branch basis to get any 'local' changes from it into the tree.
        Restore the wt.basis->wt.state changes.

        There isn't a single operation at the moment to do that, so we:
        Merge current state -> basis tree of the master w.r.t. the old tree basis.
        Do a 'normal' merge of the old branch basis if it is relevant.
        """
        old_tip = self.branch.update()
        if old_tip is not None:
            self.add_pending_merge(old_tip)
        self.branch.lock_read()
        try:
            result = 0
            if self.last_revision() != self.branch.last_revision():
                # merge tree state up to new branch tip.
                basis = self.basis_tree()
                to_tree = self.branch.basis_tree()
                result += merge_inner(self.branch,
                                      to_tree,
                                      basis,
                                      this_tree=self)
                self.set_last_revision(self.branch.last_revision())
            if old_tip and old_tip != self.last_revision():
                # our last revision was not the prior branch last revision
                # and we have converted that last revision to a pending merge.
                # base is somewhere between the branch tip now
                # and the now pending merge
                from bzrlib.revision import common_ancestor
                try:
                    base_rev_id = common_ancestor(self.branch.last_revision(),
                                                  old_tip,
                                                  self.branch.repository)
                except errors.NoCommonAncestor:
                    base_rev_id = None
                base_tree = self.branch.repository.revision_tree(base_rev_id)
                other_tree = self.branch.repository.revision_tree(old_tip)
                result += merge_inner(self.branch,
                                      other_tree,
                                      base_tree,
                                      this_tree=self)
            return result
        finally:
            self.branch.unlock()

    @needs_write_lock
    def _write_inventory(self, inv):
        """Write inventory as the current inventory."""
        sio = StringIO()
        bzrlib.xml5.serializer_v5.write_inventory(inv, sio)
        sio.seek(0)
        self._control_files.put('inventory', sio)
        self._set_inventory(inv)
        mutter('wrote working inventory')

    def set_conflicts(self, arg):
        raise UnsupportedOperation(self.set_conflicts, self)

    def add_conflicts(self, arg):
        raise UnsupportedOperation(self.add_conflicts, self)

    @needs_read_lock
    def conflicts(self):
        conflicts = ConflictList()
        for conflicted in self._iter_conflicts():
            text = True
            try:
                if file_kind(self.abspath(conflicted)) != "file":
                    text = False
            except errors.NoSuchFile:
                text = False
            if text is True:
                for suffix in ('.THIS', '.OTHER'):
                    try:
                        kind = file_kind(self.abspath(conflicted+suffix))
                        if kind != "file":
                            text = False
                    except errors.NoSuchFile:
                        text = False
                    if text == False:
                        break
            ctype = {True: 'text conflict', False: 'contents conflict'}[text]
            conflicts.append(Conflict.factory(ctype, path=conflicted,
                             file_id=self.path2id(conflicted)))
        return conflicts


class WorkingTree2(WorkingTree):
    """This is the Format 2 working tree.

    This was the first weave based working tree. 
     - uses os locks for locking.
     - uses the branch last-revision.
    """

    def unlock(self):
        # we share control files:
        if self._hashcache.needs_write and self._control_files._lock_count==3:
            self._hashcache.write()
        # reverse order of locking.
        try:
            return self._control_files.unlock()
        finally:
            self.branch.unlock()


class WorkingTree3(WorkingTree):
    """This is the Format 3 working tree.

    This differs from the base WorkingTree by:
     - having its own file lock
     - having its own last-revision property.

    This is new in bzr 0.8
    """

    @needs_read_lock
    def last_revision(self):
        """See WorkingTree.last_revision."""
        try:
            return self._control_files.get_utf8('last-revision').read()
        except NoSuchFile:
            return None

    def _change_last_revision(self, revision_id):
        """See WorkingTree._change_last_revision."""
        if revision_id is None or revision_id == NULL_REVISION:
            try:
                self._control_files._transport.delete('last-revision')
            except errors.NoSuchFile:
                pass
            return False
        else:
            self._control_files.put_utf8('last-revision', revision_id)
            return True

    @needs_write_lock
    def set_conflicts(self, conflicts):
        self._put_rio('conflicts', conflicts.to_stanzas(), 
                      CONFLICT_HEADER_1)

    @needs_write_lock
    def add_conflicts(self, new_conflicts):
        conflict_set = set(self.conflicts())
        conflict_set.update(set(list(new_conflicts)))
        self.set_conflicts(ConflictList(sorted(conflict_set,
                                               key=Conflict.sort_key)))

    @needs_read_lock
    def conflicts(self):
        try:
            confile = self._control_files.get('conflicts')
        except NoSuchFile:
            return ConflictList()
        try:
            if confile.next() != CONFLICT_HEADER_1 + '\n':
                raise ConflictFormatError()
        except StopIteration:
            raise ConflictFormatError()
        return ConflictList.from_stanzas(RioReader(confile))

    def unlock(self):
        if self._hashcache.needs_write and self._control_files._lock_count==1:
            self._hashcache.write()
        # reverse order of locking.
        try:
            return self._control_files.unlock()
        finally:
            self.branch.unlock()


def get_conflicted_stem(path):
    for suffix in CONFLICT_SUFFIXES:
        if path.endswith(suffix):
            return path[:-len(suffix)]

@deprecated_function(zero_eight)
def is_control_file(filename):
    """See WorkingTree.is_control_filename(filename)."""
    ## FIXME: better check
    filename = normpath(filename)
    while filename != '':
        head, tail = os.path.split(filename)
        ## mutter('check %r for control file' % ((head, tail),))
        if tail == '.bzr':
            return True
        if filename == head:
            break
        filename = head
    return False


class WorkingTreeFormat(object):
    """An encapsulation of the initialization and open routines for a format.

    Formats provide three things:
     * An initialization routine,
     * a format string,
     * an open routine.

    Formats are placed in an dict by their format string for reference 
    during workingtree opening. Its not required that these be instances, they
    can be classes themselves with class methods - it simply depends on 
    whether state is needed for a given format or not.

    Once a format is deprecated, just deprecate the initialize and open
    methods on the format class. Do not deprecate the object, as the 
    object will be created every time regardless.
    """

    _default_format = None
    """The default format used for new trees."""

    _formats = {}
    """The known formats."""

    @classmethod
    def find_format(klass, a_bzrdir):
        """Return the format for the working tree object in a_bzrdir."""
        try:
            transport = a_bzrdir.get_workingtree_transport(None)
            format_string = transport.get("format").read()
            return klass._formats[format_string]
        except NoSuchFile:
            raise errors.NoWorkingTree(base=transport.base)
        except KeyError:
            raise errors.UnknownFormatError(format=format_string)

    @classmethod
    def get_default_format(klass):
        """Return the current default format."""
        return klass._default_format

    def get_format_string(self):
        """Return the ASCII format string that identifies this format."""
        raise NotImplementedError(self.get_format_string)

    def get_format_description(self):
        """Return the short description for this format."""
        raise NotImplementedError(self.get_format_description)

    def is_supported(self):
        """Is this format supported?

        Supported formats can be initialized and opened.
        Unsupported formats may not support initialization or committing or 
        some other features depending on the reason for not being supported.
        """
        return True

    @classmethod
    def register_format(klass, format):
        klass._formats[format.get_format_string()] = format

    @classmethod
    def set_default_format(klass, format):
        klass._default_format = format

    @classmethod
    def unregister_format(klass, format):
        assert klass._formats[format.get_format_string()] is format
        del klass._formats[format.get_format_string()]



class WorkingTreeFormat2(WorkingTreeFormat):
    """The second working tree format. 

    This format modified the hash cache from the format 1 hash cache.
    """

    def get_format_description(self):
        """See WorkingTreeFormat.get_format_description()."""
        return "Working tree format 2"

    def stub_initialize_remote(self, control_files):
        """As a special workaround create critical control files for a remote working tree
        
        This ensures that it can later be updated and dealt with locally,
        since BzrDirFormat6 and BzrDirFormat5 cannot represent dirs with 
        no working tree.  (See bug #43064).
        """
        sio = StringIO()
        inv = Inventory()
        bzrlib.xml5.serializer_v5.write_inventory(inv, sio)
        sio.seek(0)
        control_files.put('inventory', sio)

        control_files.put_utf8('pending-merges', '')
        

    def initialize(self, a_bzrdir, revision_id=None):
        """See WorkingTreeFormat.initialize()."""
        if not isinstance(a_bzrdir.transport, LocalTransport):
            raise errors.NotLocalUrl(a_bzrdir.transport.base)
        branch = a_bzrdir.open_branch()
        if revision_id is not None:
            branch.lock_write()
            try:
                revision_history = branch.revision_history()
                try:
                    position = revision_history.index(revision_id)
                except ValueError:
                    raise errors.NoSuchRevision(branch, revision_id)
                branch.set_revision_history(revision_history[:position + 1])
            finally:
                branch.unlock()
        revision = branch.last_revision()
        inv = Inventory() 
        wt = WorkingTree2(a_bzrdir.root_transport.local_abspath('.'),
                         branch,
                         inv,
                         _internal=True,
                         _format=self,
                         _bzrdir=a_bzrdir)
        wt._write_inventory(inv)
        wt.set_root_id(inv.root.file_id)
        wt.set_last_revision(revision)
        wt.set_pending_merges([])
        build_tree(wt.basis_tree(), wt)
        return wt

    def __init__(self):
        super(WorkingTreeFormat2, self).__init__()
        self._matchingbzrdir = bzrdir.BzrDirFormat6()

    def open(self, a_bzrdir, _found=False):
        """Return the WorkingTree object for a_bzrdir

        _found is a private parameter, do not use it. It is used to indicate
               if format probing has already been done.
        """
        if not _found:
            # we are being called directly and must probe.
            raise NotImplementedError
        if not isinstance(a_bzrdir.transport, LocalTransport):
            raise errors.NotLocalUrl(a_bzrdir.transport.base)
        return WorkingTree2(a_bzrdir.root_transport.local_abspath('.'),
                           _internal=True,
                           _format=self,
                           _bzrdir=a_bzrdir)


class WorkingTreeFormat3(WorkingTreeFormat):
    """The second working tree format updated to record a format marker.

    This format:
        - exists within a metadir controlling .bzr
        - includes an explicit version marker for the workingtree control
          files, separate from the BzrDir format
        - modifies the hash cache format
        - is new in bzr 0.8
        - uses a LockDir to guard access for writes.
    """

    def get_format_string(self):
        """See WorkingTreeFormat.get_format_string()."""
        return "Bazaar-NG Working Tree format 3"

    def get_format_description(self):
        """See WorkingTreeFormat.get_format_description()."""
        return "Working tree format 3"

    _lock_file_name = 'lock'
    _lock_class = LockDir

    def _open_control_files(self, a_bzrdir):
        transport = a_bzrdir.get_workingtree_transport(None)
        return LockableFiles(transport, self._lock_file_name, 
                             self._lock_class)

    def initialize(self, a_bzrdir, revision_id=None):
        """See WorkingTreeFormat.initialize().
        
        revision_id allows creating a working tree at a different
        revision than the branch is at.
        """
        if not isinstance(a_bzrdir.transport, LocalTransport):
            raise errors.NotLocalUrl(a_bzrdir.transport.base)
        transport = a_bzrdir.get_workingtree_transport(self)
        control_files = self._open_control_files(a_bzrdir)
        control_files.create_lock()
        control_files.lock_write()
        control_files.put_utf8('format', self.get_format_string())
        branch = a_bzrdir.open_branch()
        if revision_id is None:
            revision_id = branch.last_revision()
        inv = Inventory() 
        wt = WorkingTree3(a_bzrdir.root_transport.local_abspath('.'),
                         branch,
                         inv,
                         _internal=True,
                         _format=self,
                         _bzrdir=a_bzrdir,
                         _control_files=control_files)
        wt.lock_write()
        try:
            wt._write_inventory(inv)
            wt.set_root_id(inv.root.file_id)
            wt.set_last_revision(revision_id)
            wt.set_pending_merges([])
            build_tree(wt.basis_tree(), wt)
        finally:
            wt.unlock()
            control_files.unlock()
        return wt

    def __init__(self):
        super(WorkingTreeFormat3, self).__init__()
        self._matchingbzrdir = bzrdir.BzrDirMetaFormat1()

    def open(self, a_bzrdir, _found=False):
        """Return the WorkingTree object for a_bzrdir

        _found is a private parameter, do not use it. It is used to indicate
               if format probing has already been done.
        """
        if not _found:
            # we are being called directly and must probe.
            raise NotImplementedError
        if not isinstance(a_bzrdir.transport, LocalTransport):
            raise errors.NotLocalUrl(a_bzrdir.transport.base)
        return self._open(a_bzrdir, self._open_control_files(a_bzrdir))

    def _open(self, a_bzrdir, control_files):
        """Open the tree itself.
        
        :param a_bzrdir: the dir for the tree.
        :param control_files: the control files for the tree.
        """
        return WorkingTree3(a_bzrdir.root_transport.local_abspath('.'),
                           _internal=True,
                           _format=self,
                           _bzrdir=a_bzrdir,
                           _control_files=control_files)

    def __str__(self):
        return self.get_format_string()


# formats which have no format string are not discoverable
# and not independently creatable, so are not registered.
__default_format = WorkingTreeFormat3()
WorkingTreeFormat.register_format(__default_format)
WorkingTreeFormat.set_default_format(__default_format)
_legacy_formats = [WorkingTreeFormat2(),
                   ]


class WorkingTreeTestProviderAdapter(object):
    """A tool to generate a suite testing multiple workingtree formats at once.

    This is done by copying the test once for each transport and injecting
    the transport_server, transport_readonly_server, and workingtree_format
    classes into each copy. Each copy is also given a new id() to make it
    easy to identify.
    """

    def __init__(self, transport_server, transport_readonly_server, formats):
        self._transport_server = transport_server
        self._transport_readonly_server = transport_readonly_server
        self._formats = formats
    
    def _clone_test(self, test, bzrdir_format, workingtree_format, variation):
        """Clone test for adaption."""
        new_test = deepcopy(test)
        new_test.transport_server = self._transport_server
        new_test.transport_readonly_server = self._transport_readonly_server
        new_test.bzrdir_format = bzrdir_format
        new_test.workingtree_format = workingtree_format
        def make_new_test_id():
            new_id = "%s(%s)" % (test.id(), variation)
            return lambda: new_id
        new_test.id = make_new_test_id()
        return new_test
    
    def adapt(self, test):
        from bzrlib.tests import TestSuite
        result = TestSuite()
        for workingtree_format, bzrdir_format in self._formats:
            new_test = self._clone_test(
                test,
                bzrdir_format,
                workingtree_format, workingtree_format.__class__.__name__)
            result.addTest(new_test)
        return result