« back to all changes in this revision
Viewing changes to bzrlib/weave.py
- browse files at revision 1185.12.13
- compare with another revision
- download diff
- download tarball
- view history from revision 1185.12.13
- Committer: Aaron Bentley
- Date: 2005-10-06 04:29:10 UTC
- mfrom: (1185.14.11)
- mto: (1185.25.1)
- mto: This revision was merged to the branch mainline in revision 1419.
- Revision ID: aaron.bentley@utoronto.ca-20051006042910-a3263271389bfdc2
Merged conflict handling work
- files added:
- HACKING
- bzrlib/store
- bzrlib/transport
- bzrlib/util
- contrib/emacs
- files removed:
- bzrlib/meta_store.py
- plugins/changeset
- files renamed:
- plugins/ => bzrlib/plugins/
- tools/testweave.py => bzrlib/selftest/test_weave.py
- bzrlib/store.py => bzrlib/store/__init__.py
- effbot/ => bzrlib/util/effbot/
- elementtree/ => bzrlib/util/elementtree/
- urlgrabber/ => bzrlib/util/urlgrabber/
- bzrlib/newinventory.py => contrib/newinventory.py
- files modified:
- .bzrignore
added
removed
bzrlib/weave.py
21
21
22
22
"""Weave - storage of related text file versions"""
23
23
24
# before intset (r923) 2000 versions in 41.5s
25
# with intset (r926) 2000 versions in 93s !!!
26
# better to just use plain sets.
27
28
# making _extract build and return a list, rather than being a generator
29
# takes 37.94s
30
31
# with python -O, r923 does 2000 versions in 36.87s
32
33
# with optimizations to avoid mutating lists - 35.75! I guess copying
34
# all the elements every time costs more than the small manipulations.
35
# a surprisingly small change.
36
37
# r931, which avoids using a generator for extract, does 36.98s
38
39
# with memoized inclusions, takes 41.49s; not very good
40
41
# with slots, takes 37.35s; without takes 39.16, a bit surprising
42
43
# with the delta calculation mixed in with the add method, rather than
44
# separated, takes 36.78s
45
46
# with delta folded in and mutation of the list, 36.13s
47
48
# with all this and simplification of add code, 33s
49
50
51
# TODO: Perhaps have copy method for Weave instances?
52
24
53
25
# XXX: If we do weaves this way, will a merge still behave the same
54
26
# way if it's done in a different order? That's a pretty desirable
59
31
# binaries, perhaps by naively splitting on \n or perhaps using
60
32
# something like a rolling checksum.
61
33
62
# TODO: Track version names as well as indexes.
63
64
34
# TODO: End marker for each version so we can stop reading?
65
35
66
36
# TODO: Check that no insertion occurs inside a deletion that was
73
43
74
44
# TODO: Parallel-extract that passes back each line along with a
75
45
# description of which revisions include it. Nice for checking all
76
# shas in parallel.
46
# shas or calculating stats in parallel.
47
48
# TODO: Using a single _extract routine and then processing the output
49
# is probably inefficient. It's simple enough that we can afford to
50
# have slight specializations for different ways its used: annotate,
51
# basis for add, get, etc.
52
53
# TODO: Perhaps the API should work only in names to hide the integer
54
# indexes from the user?
55
56
# TODO: Is there any potential performance win by having an add()
57
# variant that is passed a pre-cooked version of the single basis
58
# version?
59
60
# TODO: Have a way to go back and insert a revision V1 that is a parent
61
# of an already-stored revision V2. This means some lines previously
62
# counted as new in V2 will be discovered to have actually come from V1.
63
# It is probably necessary to insert V1, then compute a whole new diff
64
# from the mashed ancestors to V2. This must be repeated for every
65
# direct child of V1. The deltas from V2 to its descendents won't change,
66
# although their location within the weave may change. It may be possible
67
# to just adjust the location of those instructions rather than
68
# re-weaving the whole thing. This is expected to be a fairly rare
69
# operation, only used when inserting data that was previously a ghost.
70
71
72
73
74
import sha
75
from difflib import SequenceMatcher
77
76
78
77
79
78
118
117
The instruction can be '{' or '}' for an insertion block, and '['
119
118
and ']' for a deletion block respectively. The version is the
120
119
integer version index. There is no replace operator, only deletes
121
and inserts.
120
and inserts. For '}', the end of an insertion, there is no
121
version parameter because it always closes the most recently
122
opened insertion.
122
123
123
124
Constraints/notes:
124
125
160
161
each version; the parent's parents are implied.
161
162
162
163
_sha1s
163
List of hex SHA-1 of each version, or None if not recorded.
164
List of hex SHA-1 of each version.
165
166
_names
167
List of symbolic names for each version. Each should be unique.
168
169
_name_map
170
For each name, the version number.
171
172
_weave_name
173
Descriptive name of this weave; typically the filename if known.
174
Set by read_weave.
164
175
"""
165
176
166
__slots__ = ['_weave', '_parents', '_sha1s']
177
__slots__ = ['_weave', '_parents', '_sha1s', '_names', '_name_map',
178
'_weave_name']
167
179
168
def __init__(self):
180
def __init__(self, weave_name=None):
169
181
self._weave = []
170
182
self._parents = []
171
183
self._sha1s = []
172
184
self._names = []
185
self._name_map = {}
186
self._weave_name = weave_name
187
188
189
def copy(self):
190
"""Return a deep copy of self.
191
192
The copy can be modified without affecting the original weave."""
193
other = Weave()
194
other._weave = self._weave[:]
195
other._parents = self._parents[:]
196
other._sha1s = self._sha1s[:]
197
other._names = self._names[:]
198
other._name_map = self._name_map.copy()
199
other._weave_name = self._weave_name
200
return other
173
201
174
202
def __eq__(self, other):
175
203
if not isinstance(other, Weave):
176
204
return False
177
205
return self._parents == other._parents \
178
and self._weave == other._weave
206
and self._weave == other._weave \
207
and self._sha1s == other._sha1s
208
179
209
180
181
210
def __ne__(self, other):
182
211
return not self.__eq__(other)
183
212
184
185
def add(self, parents, text):
213
214
def maybe_lookup(self, name_or_index):
215
"""Convert possible symbolic name to index, or pass through indexes."""
216
if isinstance(name_or_index, (int, long)):
217
return name_or_index
218
else:
219
return self.lookup(name_or_index)
220
221
222
def lookup(self, name):
223
"""Convert symbolic version name to index."""
224
try:
225
return self._name_map[name]
226
except KeyError:
227
raise WeaveError("name %r not present in weave %r" %
228
(name, self._weave_name))
229
230
231
def iter_names(self):
232
"""Yield a list of all names in this weave."""
233
return iter(self._names)
234
235
def idx_to_name(self, version):
236
return self._names[version]
237
238
239
def _check_repeated_add(self, name, parents, text, sha1):
240
"""Check that a duplicated add is OK.
241
242
If it is, return the (old) index; otherwise raise an exception.
243
"""
244
idx = self.lookup(name)
245
if sorted(self._parents[idx]) != sorted(parents):
246
raise WeaveError("name \"%s\" already present in weave "
247
"with different parents" % name)
248
if sha1 != self._sha1s[idx]:
249
raise WeaveError("name \"%s\" already present in weave "
250
"with different text" % name)
251
return idx
252
253
254
255
def add(self, name, parents, text, sha1=None):
186
256
"""Add a single text on top of the weave.
187
257
188
258
Returns the index number of the newly added version.
189
259
260
name
261
Symbolic name for this version.
262
(Typically the revision-id of the revision that added it.)
263
190
264
parents
191
265
List or set of direct parent version numbers.
192
266
193
267
text
194
Sequence of lines to be added in the new version."""
195
268
Sequence of lines to be added in the new version.
269
270
sha -- SHA-1 of the file, if known. This is trusted to be
271
correct if supplied.
272
"""
273
from bzrlib.osutils import sha_strings
274
275
assert isinstance(name, basestring)
276
if sha1 is None:
277
sha1 = sha_strings(text)
278
if name in self._name_map:
279
return self._check_repeated_add(name, parents, text, sha1)
280
281
parents = map(self.maybe_lookup, parents)
196
282
self._check_versions(parents)
197
283
## self._check_lines(text)
198
284
new_version = len(self._parents)
199
285
200
import sha
201
s = sha.new()
202
map(s.update, text)
203
sha1 = s.hexdigest()
204
del s
205
286
206
# if we abort after here the weave will be corrupt
207
self._parents.append(frozenset(parents))
287
# if we abort after here the (in-memory) weave will be corrupt because only
288
# some fields are updated
289
self._parents.append(parents[:])
208
290
self._sha1s.append(sha1)
291
self._names.append(name)
292
self._name_map[name] = new_version
209
293
210
294
211
295
if not parents:
216
300
if text:
217
301
self._weave.append(('{', new_version))
218
302
self._weave.extend(text)
219
self._weave.append(('}', new_version))
303
self._weave.append(('}', None))
220
304
221
305
return new_version
222
306
238
322
basis_lineno.append(lineno)
239
323
basis_lines.append(line)
240
324
241
# another small special case: a merge, producing the same text as auto-merge
325
# another small special case: a merge, producing the same text
326
# as auto-merge
242
327
if text == basis_lines:
243
328
return new_version
244
329
252
337
#print 'basis_lines:', basis_lines
253
338
#print 'new_lines: ', lines
254
339
255
from difflib import SequenceMatcher
256
340
s = SequenceMatcher(None, basis_lines, text)
257
341
258
342
# offset gives the number of lines that have been inserted
287
371
# we don't destroy ourselves
288
372
i = i2 + offset
289
373
self._weave[i:i] = ([('{', new_version)]
290
+ text[j1:j2]
291
+ [('}', new_version)])
374
+ text[j1:j2]
375
+ [('}', None)])
292
376
offset += 2 + (j2 - j1)
293
377
294
378
return new_version
295
379
380
def add_identical(self, old_rev_id, new_rev_id, parents):
381
"""Add an identical text to old_rev_id as new_rev_id."""
382
old_lines = self.get(self.lookup(old_rev_id))
383
self.add(new_rev_id, parents, old_lines)
296
384
297
385
def inclusions(self, versions):
298
386
"""Return set of all ancestors of given version(s)."""
299
387
i = set(versions)
300
v = max(versions)
301
try:
302
while v >= 0:
303
if v in i:
304
# include all its parents
305
i.update(self._parents[v])
306
v -= 1
307
return i
308
except IndexError:
309
raise ValueError("version %d not present in weave" % v)
388
for v in xrange(max(versions), 0, -1):
389
if v in i:
390
# include all its parents
391
i.update(self._parents[v])
392
return i
393
## except IndexError:
394
## raise ValueError("version %d not present in weave" % v)
395
396
397
def parents(self, version):
398
return self._parents[version]
310
399
311
400
312
401
def minimal_parents(self, version):
352
441
raise IndexError("invalid version number %r" % i)
353
442
354
443
355
def annotate(self, index):
356
return list(self.annotate_iter(index))
357
358
359
def annotate_iter(self, version):
444
def annotate(self, name_or_index):
445
return list(self.annotate_iter(name_or_index))
446
447
448
def annotate_iter(self, name_or_index):
360
449
"""Yield list of (index-id, line) pairs for the specified version.
361
450
362
451
The index indicates when the line originated in the weave."""
363
for origin, lineno, text in self._extract([version]):
452
incls = [self.maybe_lookup(name_or_index)]
453
for origin, lineno, text in self._extract(incls):
364
454
yield origin, text
365
455
366
456
384
474
if c == '{':
385
475
istack.append(v)
386
476
elif c == '}':
387
oldv = istack.pop()
477
istack.pop()
388
478
elif c == '[':
389
479
assert v not in dset
390
480
dset.add(v)
410
500
411
501
The set typically but not necessarily corresponds to a version.
412
502
"""
503
for i in versions:
504
if not isinstance(i, int):
505
raise ValueError(i)
506
413
507
included = self.inclusions(versions)
414
508
415
509
istack = []
431
525
assert v not in istack
432
526
istack.append(v)
433
527
elif c == '}':
434
oldv = istack.pop()
435
assert oldv == v
528
istack.pop()
436
529
elif c == '[':
437
530
if v in included:
438
531
assert v not in dset
461
554
462
555
463
556
464
def get_iter(self, version):
557
def get_iter(self, name_or_index):
465
558
"""Yield lines for the specified version."""
466
for origin, lineno, line in self._extract([version]):
559
incls = [self.maybe_lookup(name_or_index)]
560
for origin, lineno, line in self._extract(incls):
467
561
yield line
468
562
469
563
470
def get(self, index):
471
return list(self.get_iter(index))
564
def get_text(self, name_or_index):
565
return ''.join(self.get_iter(name_or_index))
566
assert isinstance(version, int)
567
568
569
def get_lines(self, name_or_index):
570
return list(self.get_iter(name_or_index))
571
572
573
get = get_lines
472
574
473
575
474
576
def mash_iter(self, included):
475
577
"""Return composed version of multiple included versions."""
578
included = map(self.maybe_lookup, included)
476
579
for origin, lineno, text in self._extract(included):
477
580
yield text
478
581
508
611
509
612
# try extracting all versions; this is a bit slow and parallel
510
613
# extraction could be used
511
import sha
512
614
nv = self.numversions()
513
615
for version in range(nv):
514
616
if progress_bar:
539
641
If there is a chunk of the file where there's no diagreement,
540
642
only one alternative is given.
541
643
"""
542
543
644
# approach: find the included versions common to all the
544
645
# merged versions
545
646
raise NotImplementedError()
565
666
If line1=line2, this is a pure insert; if newlines=[] this is a
566
667
pure delete. (Similar to difflib.)
567
668
"""
568
669
raise NotImplementedError()
569
670
570
671
571
672
def plan_merge(self, ver_a, ver_b):
665
766
state
666
767
667
768
668
669
670
671
672
673
def weave_info(w):
674
"""Show some text information about the weave."""
675
print '%6s %40s %20s' % ('ver', 'sha1', 'parents')
676
for i in (6, 40, 20):
769
def join(self, other):
770
"""Integrate versions from other into this weave.
771
772
The resulting weave contains all the history of both weaves;
773
any version you could retrieve from either self or other can be
774
retrieved from self after this call.
775
776
It is illegal for the two weaves to contain different values
777
for any version.
778
"""
779
if other.numversions() == 0:
780
return # nothing to update, easy
781
# work through in index order to make sure we get all dependencies
782
for other_idx, name in enumerate(other._names):
783
# TODO: If all the parents of the other version are already
784
# present then we can avoid some work by just taking the delta
785
# and adjusting the offsets.
786
if self._check_version_consistent(other, other_idx, name):
787
continue
788
new_parents = self._imported_parents(other, other_idx)
789
lines = other.get_lines(other_idx)
790
sha1 = other._sha1s[other_idx]
791
self.add(name, new_parents, lines, sha1)
792
793
794
def _imported_parents(self, other, other_idx):
795
"""Return list of parents in self corresponding to indexes in other."""
796
new_parents = []
797
for parent_idx in other._parents[other_idx]:
798
parent_name = other._names[parent_idx]
799
if parent_name not in self._names:
800
# should not be possible
801
raise WeaveError("missing parent {%s} of {%s} in %r"
802
% (parent_name, other._name_map[other_idx], self))
803
new_parents.append(self._name_map[parent_name])
804
return new_parents
805
806
def _check_version_consistent(self, other, other_idx, name):
807
"""Check if a version in consistent in this and other.
808
809
If present & correct return True;
810
if not present in self return False;
811
if inconsistent raise error."""
812
this_idx = self._name_map.get(name, -1)
813
if this_idx != -1:
814
if self._sha1s[this_idx] != other._sha1s[other_idx]:
815
raise WeaveError("inconsistent texts for version {%s} in %r and %r"
816
% (name, self, other))
817
elif set(self._parents[this_idx]) != set(other._parents[other_idx]):
818
raise WeaveError("inconsistent parents for version {%s} in %r and %r"
819
% (name, self, other))
820
else:
821
return True # ok!
822
else:
823
return False
824
825
826
def weave_toc(w):
827
"""Show the weave's table-of-contents"""
828
print '%6s %50s %10s %10s' % ('ver', 'name', 'sha1', 'parents')
829
for i in (6, 50, 10, 10):
677
830
print '-' * i,
678
831
print
679
832
for i in range(w.numversions()):
680
833
sha1 = w._sha1s[i]
681
print '%6d %40s %s' % (i, sha1, ' '.join(map(str, w._parents[i])))
682
683
684
685
def weave_stats(weave_file):
686
from bzrlib.progress import ProgressBar
834
name = w._names[i]
835
parent_str = ' '.join(map(str, w._parents[i]))
836
print '%6d %-50.50s %10.10s %s' % (i, name, sha1, parent_str)
837
838
839
840
def weave_stats(weave_file, pb):
687
841
from bzrlib.weavefile import read_weave
688
842
689
pb = ProgressBar()
690
691
843
wf = file(weave_file, 'rb')
692
844
w = read_weave(wf)
693
845
# FIXME: doesn't work on pipes
697
849
vers = len(w)
698
850
for i in range(vers):
699
851
pb.update('checking sizes', i, vers)
700
for line in w.get_iter(i):
852
for origin, lineno, line in w._extract([i]):
701
853
total += len(line)
702
854
703
855
pb.clear()
706
858
print 'weave file %9d bytes' % weave_size
707
859
print 'total contents %9d bytes' % total
708
860
print 'compression ratio %9.2fx' % (float(total) / float(weave_size))
709
861
if vers:
862
avg = total/vers
863
print 'average size %9d bytes' % avg
864
print 'relative size %9.2fx' % (float(weave_size) / float(avg))
710
865
711
866
712
867
def usage():
721
876
Write out specified version.
722
877
weave check WEAVEFILE
723
878
Check consistency of all versions.
724
weave info WEAVEFILE
879
weave toc WEAVEFILE
725
880
Display table of contents.
726
weave add WEAVEFILE [BASE...] < NEWTEXT
881
weave add WEAVEFILE NAME [BASE...] < NEWTEXT
727
882
Add NEWTEXT, with specified parent versions.
728
883
weave annotate WEAVEFILE VERSION
729
884
Display origin of each line.
731
886
Display composite of all selected versions.
732
887
weave merge WEAVEFILE VERSION1 VERSION2 > OUT
733
888
Auto-merge two versions and display conflicts.
889
weave diff WEAVEFILE VERSION1 VERSION2
890
Show differences between two versions.
734
891
735
892
example:
736
893
737
894
% weave init foo.weave
738
895
% vi foo.txt
739
% weave add foo.weave < foo.txt
896
% weave add foo.weave ver0 < foo.txt
740
897
added version 0
741
898
742
899
(create updated version)
743
900
% vi foo.txt
744
901
% weave get foo.weave 0 | diff -u - foo.txt
745
% weave add foo.weave 0 < foo.txt
902
% weave add foo.weave ver1 0 < foo.txt
746
903
added version 1
747
904
748
905
% weave get foo.weave 0 > foo.txt (create forked version)
749
906
% vi foo.txt
750
% weave add foo.weave 0 < foo.txt
907
% weave add foo.weave ver2 0 < foo.txt
751
908
added version 2
752
909
753
910
% weave merge foo.weave 1 2 > foo.txt (merge them)
754
911
% vi foo.txt (resolve conflicts)
755
% weave add foo.weave 1 2 < foo.txt (commit merged version)
912
% weave add foo.weave merged 1 2 < foo.txt (commit merged version)
756
913
757
914
"""
758
915
761
918
def main(argv):
762
919
import sys
763
920
import os
764
from weavefile import write_weave, read_weave
921
try:
922
import bzrlib
923
except ImportError:
924
# in case we're run directly from the subdirectory
925
sys.path.append('..')
926
import bzrlib
927
from bzrlib.weavefile import write_weave, read_weave
765
928
from bzrlib.progress import ProgressBar
766
929
767
#import psyco
768
#psyco.full()
930
try:
931
import psyco
932
psyco.full()
933
except ImportError:
934
pass
935
936
if len(argv) < 2:
937
usage()
938
return 0
769
939
770
940
cmd = argv[1]
771
941
777
947
elif cmd == 'add':
778
948
w = readit()
779
949
# at the moment, based on everything in the file
780
parents = map(int, argv[3:])
950
name = argv[3]
951
parents = map(int, argv[4:])
781
952
lines = sys.stdin.readlines()
782
ver = w.add(parents, lines)
953
ver = w.add(name, parents, lines)
783
954
write_weave(w, file(argv[2], 'wb'))
784
print 'added version %d' % ver
955
print 'added version %r %d' % (name, ver)
785
956
elif cmd == 'init':
786
957
fn = argv[2]
787
958
if os.path.exists(fn):
796
967
w = readit()
797
968
sys.stdout.writelines(w.mash_iter(map(int, argv[3:])))
798
969
970
elif cmd == 'diff':
971
from difflib import unified_diff
972
w = readit()
973
fn = argv[2]
974
v1, v2 = map(int, argv[3:5])
975
lines1 = w.get(v1)
976
lines2 = w.get(v2)
977
diff_gen = unified_diff(lines1, lines2,
978
'%s version %d' % (fn, v1),
979
'%s version %d' % (fn, v2))
980
sys.stdout.writelines(diff_gen)
981
799
982
elif cmd == 'annotate':
800
983
w = readit()
801
984
# newline is added to all lines regardless; too hard to get
809
992
print '%5d | %s' % (origin, text)
810
993
lasto = origin
811
994
812
elif cmd == 'info':
813
weave_info(readit())
995
elif cmd == 'toc':
996
weave_toc(readit())
814
997
815
998
elif cmd == 'stats':
816
weave_stats(argv[2])
999
weave_stats(argv[2], ProgressBar())
817
1000
818
1001
elif cmd == 'check':
819
1002
w = readit()
865
1048
raise ValueError('unknown command %r' % cmd)
866
1049
867
1050
1051
1052
def profile_main(argv):
1053
import tempfile, hotshot, hotshot.stats
1054
1055
prof_f = tempfile.NamedTemporaryFile()
1056
1057
prof = hotshot.Profile(prof_f.name)
1058
1059
ret = prof.runcall(main, argv)
1060
prof.close()
1061
1062
stats = hotshot.stats.load(prof_f.name)
1063
#stats.strip_dirs()
1064
stats.sort_stats('cumulative')
1065
## XXX: Might like to write to stderr or the trace file instead but
1066
## print_stats seems hardcoded to stdout
1067
stats.print_stats(20)
1068
1069
return ret
1070
1071
868
1072
if __name__ == '__main__':
869
1073
import sys
870
sys.exit(main(sys.argv))
1074
if '--profile' in sys.argv:
1075
args = sys.argv[:]
1076
args.remove('--profile')
1077
sys.exit(profile_main(args))
1078
else:
1079
sys.exit(main(sys.argv))
1080
Loggerhead is a web-based interface for Breezy
Version: 2.0.1