59
61
# where the basis and destination are unchanged.
61
63
# FIXME: Sometimes we will be given a parents list for a revision
62
# that includes some redundant parents (i.e. already a parent of
63
# something in the list.) We should eliminate them. This can
64
# that includes some redundant parents (i.e. already a parent of
65
# something in the list.) We should eliminate them. This can
64
66
# be done fairly efficiently because the sequence numbers constrain
65
67
# the possible relationships.
67
# FIXME: the conflict markers should be *7* characters
70
from cStringIO import StringIO
75
from bzrlib.lazy_import import lazy_import
76
lazy_import(globals(), """
77
from bzrlib import tsort
84
from bzrlib.errors import (WeaveError, WeaveFormatError, WeaveParentMismatch,
85
RevisionAlreadyPresent,
87
UnavailableRepresentation,
88
WeaveRevisionAlreadyPresent,
89
WeaveRevisionNotPresent,
91
from bzrlib.osutils import dirname, sha, sha_strings, split_lines
92
import bzrlib.patiencediff
93
from bzrlib.revision import NULL_REVISION
94
from bzrlib.symbol_versioning import *
71
from difflib import SequenceMatcher
95
73
from bzrlib.trace import mutter
96
from bzrlib.versionedfile import (
103
from bzrlib.weavefile import _read_weave_v5, write_weave_v5
106
class WeaveContentFactory(ContentFactory):
107
"""Content factory for streaming from weaves.
109
:seealso ContentFactory:
112
def __init__(self, version, weave):
113
"""Create a WeaveContentFactory for version from weave."""
114
ContentFactory.__init__(self)
115
self.sha1 = weave.get_sha1s([version])[version]
116
self.key = (version,)
117
parents = weave.get_parent_map([version])[version]
118
self.parents = tuple((parent,) for parent in parents)
119
self.storage_kind = 'fulltext'
122
def get_bytes_as(self, storage_kind):
123
if storage_kind == 'fulltext':
124
return self._weave.get_text(self.key[-1])
125
elif storage_kind == 'chunked':
126
return self._weave.get_lines(self.key[-1])
128
raise UnavailableRepresentation(self.key, storage_kind, 'fulltext')
131
class Weave(VersionedFile):
76
class WeaveError(Exception):
77
"""Exception in processing weave"""
80
class WeaveFormatError(WeaveError):
81
"""Weave invariant violated"""
84
class WeaveParentMismatch(WeaveError):
85
"""Parents are mismatched between two revisions."""
132
89
"""weave - versioned text file storage.
134
91
A Weave manages versions of line-based text files, keeping track
135
92
of the originating version for each line.
275
205
return self._parents == other._parents \
276
206
and self._weave == other._weave \
277
and self._sha1s == other._sha1s
207
and self._sha1s == other._sha1s
279
210
def __ne__(self, other):
280
211
return not self.__eq__(other)
282
def _idx_to_name(self, version):
283
return self._names[version]
285
def _lookup(self, name):
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)):
219
return self.lookup(name_or_index)
222
def lookup(self, name):
286
223
"""Convert symbolic version name to index."""
287
if not self._allow_reserved:
288
self.check_not_reserved_id(name)
290
225
return self._name_map[name]
292
raise RevisionNotPresent(name, self._weave_name)
295
"""See VersionedFile.versions."""
296
return self._names[:]
298
def has_version(self, version_id):
299
"""See VersionedFile.has_version."""
300
return (version_id in self._name_map)
302
__contains__ = has_version
304
def get_record_stream(self, versions, ordering, include_delta_closure):
305
"""Get a stream of records for versions.
307
:param versions: The versions to include. Each version is a tuple
309
:param ordering: Either 'unordered' or 'topological'. A topologically
310
sorted stream has compression parents strictly before their
312
:param include_delta_closure: If True then the closure across any
313
compression parents will be included (in the opaque data).
314
:return: An iterator of ContentFactory objects, each of which is only
315
valid until the iterator is advanced.
317
versions = [version[-1] for version in versions]
318
if ordering == 'topological':
319
parents = self.get_parent_map(versions)
320
new_versions = tsort.topo_sort(parents)
321
new_versions.extend(set(versions).difference(set(parents)))
322
versions = new_versions
323
elif ordering == 'groupcompress':
324
parents = self.get_parent_map(versions)
325
new_versions = sort_groupcompress(parents)
326
new_versions.extend(set(versions).difference(set(parents)))
327
versions = new_versions
328
for version in versions:
330
yield WeaveContentFactory(version, self)
332
yield AbsentContentFactory((version,))
334
def get_parent_map(self, version_ids):
335
"""See VersionedFile.get_parent_map."""
337
for version_id in version_ids:
338
if version_id == NULL_REVISION:
343
map(self._idx_to_name,
344
self._parents[self._lookup(version_id)]))
345
except RevisionNotPresent:
347
result[version_id] = parents
350
def get_parents_with_ghosts(self, version_id):
351
raise NotImplementedError(self.get_parents_with_ghosts)
353
def insert_record_stream(self, stream):
354
"""Insert a record stream into this versioned file.
356
:param stream: A stream of records to insert.
358
:seealso VersionedFile.get_record_stream:
361
for record in stream:
362
# Raise an error when a record is missing.
363
if record.storage_kind == 'absent':
364
raise RevisionNotPresent([record.key[0]], self)
365
# adapt to non-tuple interface
366
parents = [parent[0] for parent in record.parents]
367
if (record.storage_kind == 'fulltext'
368
or record.storage_kind == 'chunked'):
369
self.add_lines(record.key[0], parents,
370
osutils.chunks_to_lines(record.get_bytes_as('chunked')))
372
adapter_key = record.storage_kind, 'fulltext'
374
adapter = adapters[adapter_key]
376
adapter_factory = adapter_registry.get(adapter_key)
377
adapter = adapter_factory(self)
378
adapters[adapter_key] = adapter
379
lines = split_lines(adapter.get_bytes(record))
381
self.add_lines(record.key[0], parents, lines)
382
except RevisionAlreadyPresent:
227
raise WeaveError("name %r not present in weave %r" %
228
(name, self._weave_name))
231
def iter_names(self):
232
"""Yield a list of all names in this weave."""
233
return iter(self._names)
235
def idx_to_name(self, version):
236
return self._names[version]
385
239
def _check_repeated_add(self, name, parents, text, sha1):
386
240
"""Check that a duplicated add is OK.
388
242
If it is, return the (old) index; otherwise raise an exception.
390
idx = self._lookup(name)
391
if sorted(self._parents[idx]) != sorted(parents) \
392
or sha1 != self._sha1s[idx]:
393
raise RevisionAlreadyPresent(name, self._weave_name)
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)
396
def _add_lines(self, version_id, parents, lines, parent_texts,
397
left_matching_blocks, nostore_sha, random_id, check_content):
398
"""See VersionedFile.add_lines."""
399
idx = self._add(version_id, lines, map(self._lookup, parents),
400
nostore_sha=nostore_sha)
401
return sha_strings(lines), sum(map(len, lines)), idx
403
def _add(self, version_id, lines, parents, sha1=None, nostore_sha=None):
255
def add(self, name, parents, text, sha1=None):
404
256
"""Add a single text on top of the weave.
406
258
Returns the index number of the newly added version.
409
261
Symbolic name for this version.
410
262
(Typically the revision-id of the revision that added it.)
411
If None, a name will be allocated based on the hash. (sha1:SHAHASH)
414
265
List or set of direct parent version numbers.
417
268
Sequence of lines to be added in the new version.
419
:param nostore_sha: See VersionedFile.add_lines.
270
sha -- SHA-1 of the file, if known. This is trusted to be
421
self._check_lines_not_unicode(lines)
422
self._check_lines_are_lines(lines)
424
sha1 = sha_strings(lines)
425
if sha1 == nostore_sha:
426
raise errors.ExistingContent
427
if version_id is None:
428
version_id = "sha1:" + sha1
429
if version_id in self._name_map:
430
return self._check_repeated_add(version_id, parents, lines, sha1)
273
from bzrlib.osutils import sha_strings
275
assert isinstance(name, basestring)
277
sha1 = sha_strings(text)
278
if name in self._name_map:
279
return self._check_repeated_add(name, parents, text, sha1)
281
parents = map(self.maybe_lookup, parents)
432
282
self._check_versions(parents)
433
## self._check_lines(lines)
283
## self._check_lines(text)
434
284
new_version = len(self._parents)
436
287
# if we abort after here the (in-memory) weave will be corrupt because only
437
288
# some fields are updated
438
# XXX: FIXME implement a succeed-or-fail of the rest of this routine.
439
# - Robert Collins 20060226
440
289
self._parents.append(parents[:])
441
290
self._sha1s.append(sha1)
442
self._names.append(version_id)
443
self._name_map[version_id] = new_version
291
self._names.append(name)
292
self._name_map[name] = new_version
447
296
# special case; adding with no parents revision; can do
448
297
# this more quickly by just appending unconditionally.
449
298
# even more specially, if we're adding an empty text we
450
299
# need do nothing at all.
452
301
self._weave.append(('{', new_version))
453
self._weave.extend(lines)
302
self._weave.extend(text)
454
303
self._weave.append(('}', None))
455
305
return new_version
457
307
if len(parents) == 1:
559
445
except IndexError:
560
446
raise IndexError("invalid version number %r" % i)
562
def _compatible_parents(self, my_parents, other_parents):
563
"""During join check that other_parents are joinable with my_parents.
565
Joinable is defined as 'is a subset of' - supersets may require
566
regeneration of diffs, but subsets do not.
568
return len(other_parents.difference(my_parents)) == 0
570
def annotate(self, version_id):
571
"""Return a list of (version-id, line) tuples for version_id.
449
def annotate(self, name_or_index):
450
return list(self.annotate_iter(name_or_index))
453
def annotate_iter(self, name_or_index):
454
"""Yield list of (index-id, line) pairs for the specified version.
573
456
The index indicates when the line originated in the weave."""
574
incls = [self._lookup(version_id)]
575
return [(self._idx_to_name(origin), text) for origin, lineno, text in
576
self._extract(incls)]
578
def iter_lines_added_or_present_in_versions(self, version_ids=None,
580
"""See VersionedFile.iter_lines_added_or_present_in_versions()."""
581
if version_ids is None:
582
version_ids = self.versions()
583
version_ids = set(version_ids)
584
for lineno, inserted, deletes, line in self._walk_internal(version_ids):
585
if inserted not in version_ids: continue
587
yield line + '\n', inserted
591
def _walk_internal(self, version_ids=None):
592
"""Helper method for weave actions."""
457
incls = [self.maybe_lookup(name_or_index)]
458
for origin, lineno, text in self._extract(incls):
466
(lineno, insert, deletes, text)
467
for each literal line.
597
473
lineno = 0 # line of weave, 0-based
599
475
for l in self._weave:
600
if l.__class__ == tuple:
476
if isinstance(l, tuple):
604
istack.append(self._names[v])
608
dset.add(self._names[v])
610
dset.remove(self._names[v])
612
raise WeaveFormatError('unexpected instruction %r' % v)
614
yield lineno, istack[-1], frozenset(dset), l
489
raise WeaveFormatError('unexpected instruction %r'
492
assert isinstance(l, basestring)
494
yield lineno, istack[-1], dset, l
499
def _extract(self, versions):
500
"""Yield annotation of lines in included set.
502
Yields a sequence of tuples (origin, lineno, text), where
503
origin is the origin version, lineno the index in the weave,
504
and text the text of the line.
506
The set typically but not necessarily corresponds to a version.
509
if not isinstance(i, int):
512
included = self.inclusions(versions)
517
lineno = 0 # line of weave, 0-based
523
WFE = WeaveFormatError
525
for l in self._weave:
526
if isinstance(l, tuple):
530
assert v not in istack
544
assert isinstance(l, basestring)
546
isactive = (not dset) and istack and (istack[-1] in included)
548
result.append((istack[-1], lineno, l))
618
raise WeaveFormatError("unclosed insertion blocks "
619
"at end of weave: %s" % istack)
552
raise WFE("unclosed insertion blocks at end of weave",
621
raise WeaveFormatError("unclosed deletion blocks at end of weave: %s"
555
raise WFE("unclosed deletion blocks at end of weave",
562
def get_iter(self, name_or_index):
563
"""Yield lines for the specified version."""
564
incls = [self.maybe_lookup(name_or_index)]
565
for origin, lineno, line in self._extract(incls):
569
def get_text(self, name_or_index):
570
return ''.join(self.get_iter(name_or_index))
571
assert isinstance(version, int)
574
def get_lines(self, name_or_index):
575
return list(self.get_iter(name_or_index))
581
def mash_iter(self, included):
582
"""Return composed version of multiple included versions."""
583
included = map(self.maybe_lookup, included)
584
for origin, lineno, text in self._extract(included):
588
def dump(self, to_file):
589
from pprint import pprint
590
print >>to_file, "Weave._weave = ",
591
pprint(self._weave, to_file)
592
print >>to_file, "Weave._parents = ",
593
pprint(self._parents, to_file)
597
def numversions(self):
598
l = len(self._parents)
599
assert l == len(self._sha1s)
604
return self.numversions()
607
def check(self, progress_bar=None):
608
# check no circular inclusions
609
for version in range(self.numversions()):
610
inclusions = list(self._parents[version])
613
if inclusions[-1] >= version:
614
raise WeaveFormatError("invalid included version %d for index %d"
615
% (inclusions[-1], version))
617
# try extracting all versions; this is a bit slow and parallel
618
# extraction could be used
619
nv = self.numversions()
620
for version in range(nv):
622
progress_bar.update('checking text', version, nv)
624
for l in self.get_iter(version):
627
expected = self._sha1s[version]
629
raise WeaveError("mismatched sha1 for version %d; "
630
"got %s, expected %s"
631
% (version, hd, expected))
633
# TODO: check insertions are properly nested, that there are
634
# no lines outside of insertion blocks, that deletions are
635
# properly paired, etc.
639
def merge(self, merge_versions):
640
"""Automerge and mark conflicts between versions.
642
This returns a sequence, each entry describing alternatives
643
for a chunk of the file. Each of the alternatives is given as
646
If there is a chunk of the file where there's no diagreement,
647
only one alternative is given.
649
# approach: find the included versions common to all the
651
raise NotImplementedError()
655
def _delta(self, included, lines):
656
"""Return changes from basis to new revision.
658
The old text for comparison is the union of included revisions.
660
This is used in inserting a new text.
662
Delta is returned as a sequence of
663
(weave1, weave2, newlines).
665
This indicates that weave1:weave2 of the old weave should be
666
replaced by the sequence of lines in newlines. Note that
667
these line numbers are positions in the total weave and don't
668
correspond to the lines in any extracted version, or even the
669
extracted union of included versions.
671
If line1=line2, this is a pure insert; if newlines=[] this is a
672
pure delete. (Similar to difflib.)
674
raise NotImplementedError()
624
677
def plan_merge(self, ver_a, ver_b):
625
678
"""Return pseudo-annotation indicating how the two versions merge.
665
718
# not in either revision
666
719
yield 'irrelevant', line
668
def _extract(self, versions):
669
"""Yield annotation of lines in included set.
671
Yields a sequence of tuples (origin, lineno, text), where
672
origin is the origin version, lineno the index in the weave,
673
and text the text of the line.
675
The set typically but not necessarily corresponds to a version.
678
if not isinstance(i, int):
681
included = self._inclusions(versions)
687
lineno = 0 # line of weave, 0-based
693
WFE = WeaveFormatError
696
# 449 0 4474.6820 2356.5590 bzrlib.weave:556(_extract)
697
# +285282 0 1676.8040 1676.8040 +<isinstance>
698
# 1.6 seconds in 'isinstance'.
699
# changing the first isinstance:
700
# 449 0 2814.2660 1577.1760 bzrlib.weave:556(_extract)
701
# +140414 0 762.8050 762.8050 +<isinstance>
702
# note that the inline time actually dropped (less function calls)
703
# and total processing time was halved.
704
# we're still spending ~1/4 of the method in isinstance though.
705
# so lets hard code the acceptable string classes we expect:
706
# 449 0 1202.9420 786.2930 bzrlib.weave:556(_extract)
707
# +71352 0 377.5560 377.5560 +<method 'append' of 'list'
709
# yay, down to ~1/4 the initial extract time, and our inline time
710
# has shrunk again, with isinstance no longer dominating.
711
# tweaking the stack inclusion test to use a set gives:
712
# 449 0 1122.8030 713.0080 bzrlib.weave:556(_extract)
713
# +71352 0 354.9980 354.9980 +<method 'append' of 'list'
715
# - a 5% win, or possibly just noise. However with large istacks that
716
# 'in' test could dominate, so I'm leaving this change in place -
717
# when its fast enough to consider profiling big datasets we can review.
722
for l in self._weave:
723
if l.__class__ == tuple:
730
iset.remove(istack.pop())
721
yield 'unchanged', '' # terminator
725
def weave_merge(self, plan):
730
for state, line in plan:
731
if state == 'unchanged' or state == 'killed-both':
732
# resync and flush queued conflicts changes if any
733
if not lines_a and not lines_b:
735
elif ch_a and not ch_b:
737
for l in lines_a: yield l
738
elif ch_b and not ch_a:
739
for l in lines_b: yield l
740
elif lines_a == lines_b:
741
for l in lines_a: yield l
738
raise AssertionError()
744
for l in lines_a: yield l
746
for l in lines_b: yield l
753
if state == 'unchanged':
756
elif state == 'killed-a':
759
elif state == 'killed-b':
762
elif state == 'new-a':
765
elif state == 'new-b':
741
isactive = (not dset) and istack and (istack[-1] in included)
743
result.append((istack[-1], lineno, l))
746
raise WeaveFormatError("unclosed insertion blocks "
747
"at end of weave: %s" % istack)
749
raise WeaveFormatError("unclosed deletion blocks at end of weave: %s"
753
def _maybe_lookup(self, name_or_index):
754
"""Convert possible symbolic name to index, or pass through indexes.
769
assert state in ('irrelevant', 'ghost-a', 'ghost-b', 'killed-base',
774
def join(self, other):
775
"""Integrate versions from other into this weave.
777
The resulting weave contains all the history of both weaves;
778
any version you could retrieve from either self or other can be
779
retrieved from self after this call.
781
It is illegal for the two weaves to contain different values
782
or different parents for any version. See also reweave().
758
if isinstance(name_or_index, (int, long)):
761
return self._lookup(name_or_index)
763
def get_lines(self, version_id):
764
"""See VersionedFile.get_lines()."""
765
int_index = self._maybe_lookup(version_id)
766
result = [line for (origin, lineno, line) in self._extract([int_index])]
767
expected_sha1 = self._sha1s[int_index]
768
measured_sha1 = sha_strings(result)
769
if measured_sha1 != expected_sha1:
770
raise errors.WeaveInvalidChecksum(
771
'file %s, revision %s, expected: %s, measured %s'
772
% (self._weave_name, version_id,
773
expected_sha1, measured_sha1))
776
def get_sha1s(self, version_ids):
777
"""See VersionedFile.get_sha1s()."""
779
for v in version_ids:
780
result[v] = self._sha1s[self._lookup(v)]
783
def num_versions(self):
784
"""How many versions are in this weave?"""
785
l = len(self._parents)
788
__len__ = num_versions
790
def check(self, progress_bar=None):
791
# TODO evaluate performance hit of using string sets in this routine.
792
# TODO: check no circular inclusions
793
# TODO: create a nested progress bar
794
for version in range(self.num_versions()):
795
inclusions = list(self._parents[version])
798
if inclusions[-1] >= version:
799
raise WeaveFormatError("invalid included version %d for index %d"
800
% (inclusions[-1], version))
802
# try extracting all versions; parallel extraction is used
803
nv = self.num_versions()
808
# For creating the ancestry, IntSet is much faster (3.7s vs 0.17s)
809
# The problem is that set membership is much more expensive
810
name = self._idx_to_name(i)
813
new_inc = set([name])
814
for p in self._parents[i]:
815
new_inc.update(inclusions[self._idx_to_name(p)])
817
if set(new_inc) != set(self.get_ancestry(name)):
818
raise AssertionError(
820
% (set(new_inc), set(self.get_ancestry(name))))
821
inclusions[name] = new_inc
823
nlines = len(self._weave)
825
update_text = 'checking weave'
827
short_name = os.path.basename(self._weave_name)
828
update_text = 'checking %s' % (short_name,)
829
update_text = update_text[:25]
831
for lineno, insert, deleteset, line in self._walk_internal():
833
progress_bar.update(update_text, lineno, nlines)
835
for name, name_inclusions in inclusions.items():
836
# The active inclusion must be an ancestor,
837
# and no ancestors must have deleted this line,
838
# because we don't support resurrection.
839
if (insert in name_inclusions) and not (deleteset & name_inclusions):
840
sha1s[name].update(line)
843
version = self._idx_to_name(i)
844
hd = sha1s[version].hexdigest()
845
expected = self._sha1s[i]
847
raise errors.WeaveInvalidChecksum(
848
"mismatched sha1 for version %s: "
849
"got %s, expected %s"
850
% (version, hd, expected))
852
# TODO: check insertions are properly nested, that there are
853
# no lines outside of insertion blocks, that deletions are
854
# properly paired, etc.
784
if other.numversions() == 0:
785
return # nothing to update, easy
786
# two loops so that we do not change ourselves before verifying it
788
# work through in index order to make sure we get all dependencies
789
for other_idx, name in enumerate(other._names):
790
if self._check_version_consistent(other, other_idx, name):
792
for other_idx, name in enumerate(other._names):
793
# TODO: If all the parents of the other version are already
794
# present then we can avoid some work by just taking the delta
795
# and adjusting the offsets.
796
new_parents = self._imported_parents(other, other_idx)
797
lines = other.get_lines(other_idx)
798
sha1 = other._sha1s[other_idx]
799
self.add(name, new_parents, lines, sha1)
856
802
def _imported_parents(self, other, other_idx):
857
803
"""Return list of parents in self corresponding to indexes in other."""
859
805
for parent_idx in other._parents[other_idx]:
860
806
parent_name = other._names[parent_idx]
861
if parent_name not in self._name_map:
807
if parent_name not in self._names:
862
808
# should not be possible
863
raise WeaveError("missing parent {%s} of {%s} in %r"
809
raise WeaveError("missing parent {%s} of {%s} in %r"
864
810
% (parent_name, other._name_map[other_idx], self))
865
811
new_parents.append(self._name_map[parent_name])
866
812
return new_parents
896
def _reweave(self, other, pb, msg):
897
"""Reweave self with other - internal helper for join().
899
:param other: The other weave to merge
900
:param pb: An optional progress bar, indicating how far done we are
901
:param msg: An optional message for the progress
903
new_weave = _reweave(self, other, pb=pb, msg=msg)
904
self._copy_weave_content(new_weave)
906
def _copy_weave_content(self, otherweave):
907
"""adsorb the content from otherweave."""
846
def reweave(self, other):
847
"""Reweave self with other."""
848
new_weave = reweave(self, other)
908
849
for attr in self.__slots__:
909
if attr != '_weave_name':
910
setattr(self, attr, copy(getattr(otherweave, attr)))
913
class WeaveFile(Weave):
914
"""A WeaveFile represents a Weave on disk and writes on change."""
916
WEAVE_SUFFIX = '.weave'
918
def __init__(self, name, transport, filemode=None, create=False, access_mode='w', get_scope=None):
919
"""Create a WeaveFile.
921
:param create: If not True, only open an existing knit.
923
super(WeaveFile, self).__init__(name, access_mode, get_scope=get_scope,
924
allow_reserved=False)
925
self._transport = transport
926
self._filemode = filemode
928
_read_weave_v5(self._transport.get(name + WeaveFile.WEAVE_SUFFIX), self)
929
except errors.NoSuchFile:
935
def _add_lines(self, version_id, parents, lines, parent_texts,
936
left_matching_blocks, nostore_sha, random_id, check_content):
937
"""Add a version and save the weave."""
938
self.check_not_reserved_id(version_id)
939
result = super(WeaveFile, self)._add_lines(version_id, parents, lines,
940
parent_texts, left_matching_blocks, nostore_sha, random_id,
945
def copy_to(self, name, transport):
946
"""See VersionedFile.copy_to()."""
947
# as we are all in memory always, just serialise to the new place.
949
write_weave_v5(self, sio)
951
transport.put_file(name + WeaveFile.WEAVE_SUFFIX, sio, self._filemode)
954
"""Save the weave."""
955
self._check_write_ok()
957
write_weave_v5(self, sio)
959
bytes = sio.getvalue()
960
path = self._weave_name + WeaveFile.WEAVE_SUFFIX
962
self._transport.put_bytes(path, bytes, self._filemode)
963
except errors.NoSuchFile:
964
self._transport.mkdir(dirname(path))
965
self._transport.put_bytes(path, bytes, self._filemode)
969
"""See VersionedFile.get_suffixes()."""
970
return [WeaveFile.WEAVE_SUFFIX]
972
def insert_record_stream(self, stream):
973
super(WeaveFile, self).insert_record_stream(stream)
977
def _reweave(wa, wb, pb=None, msg=None):
850
setattr(self, attr, getattr(new_weave, attr))
978
854
"""Combine two weaves and return the result.
980
This works even if a revision R has different parents in
856
This works even if a revision R has different parents in
981
857
wa and wb. In the resulting weave all the parents are given.
983
This is done by just building up a new weave, maintaining ordering
859
This is done by just building up a new weave, maintaining ordering
984
860
of the versions in the two inputs. More efficient approaches
985
might be possible but it should only be necessary to do
986
this operation rarely, when a new previously ghost version is
861
might be possible but it should only be necessary to do
862
this operation rarely, when a new previously ghost version is
989
:param pb: An optional progress bar, indicating how far done we are
990
:param msg: An optional message for the progress
994
queue_a = range(wa.num_versions())
995
queue_b = range(wb.num_versions())
867
queue_a = range(wa.numversions())
868
queue_b = range(wb.numversions())
996
869
# first determine combined parents of all versions
997
870
# map from version name -> all parent names
998
871
combined_parents = _reweave_parent_graphs(wa, wb)
999
872
mutter("combined parents: %r", combined_parents)
1000
order = tsort.topo_sort(combined_parents.iteritems())
873
order = _make_reweave_order(wa._names, wb._names, combined_parents)
1001
874
mutter("order to reweave: %r", order)
1006
for idx, name in enumerate(order):
1008
pb.update(msg, idx, len(order))
1009
876
if name in wa._name_map:
1010
877
lines = wa.get_lines(name)
1011
878
if name in wb._name_map:
1012
lines_b = wb.get_lines(name)
1013
if lines != lines_b:
1014
mutter('Weaves differ on content. rev_id {%s}', name)
1015
mutter('weaves: %s, %s', wa._weave_name, wb._weave_name)
1017
lines = list(difflib.unified_diff(lines, lines_b,
1018
wa._weave_name, wb._weave_name))
1019
mutter('lines:\n%s', ''.join(lines))
1020
raise errors.WeaveTextDiffers(name, wa, wb)
879
assert lines == wb.get_lines(name)
1022
881
lines = wb.get_lines(name)
1023
wr._add(name, lines, [wr._lookup(i) for i in combined_parents[name]])
882
wr.add(name, combined_parents[name], lines)
1027
886
def _reweave_parent_graphs(wa, wb):
1028
887
"""Return combined parent ancestry for two weaves.
1030
889
Returned as a list of (version_name, set(parent_names))"""
1032
891
for weave in [wa, wb]:
1033
892
for idx, name in enumerate(weave._names):
1034
893
p = combined.setdefault(name, set())
1035
p.update(map(weave._idx_to_name, weave._parents[idx]))
894
p.update(map(weave.idx_to_name, weave._parents[idx]))
898
def _make_reweave_order(wa_order, wb_order, combined_parents):
899
"""Return an order to reweave versions respecting parents."""
903
next_a = next_b = None
904
len_a = len(wa_order)
905
len_b = len(wb_order)
906
while ia < len(wa_order) or ib < len(wb_order):
908
next_a = wa_order[ia]
912
if combined_parents[next_a].issubset(done):
914
result.append(next_a)
918
next_b = wb_order[ib]
922
elif combined_parents[next_b].issubset(done):
924
result.append(next_b)
927
raise WeaveError("don't know how to reweave at {%s} and {%s}"
933
"""Show the weave's table-of-contents"""
934
print '%6s %50s %10s %10s' % ('ver', 'name', 'sha1', 'parents')
935
for i in (6, 50, 10, 10):
938
for i in range(w.numversions()):
941
parent_str = ' '.join(map(str, w._parents[i]))
942
print '%6d %-50.50s %10.10s %s' % (i, name, sha1, parent_str)
946
def weave_stats(weave_file, pb):
947
from bzrlib.weavefile import read_weave
949
wf = file(weave_file, 'rb')
951
# FIXME: doesn't work on pipes
952
weave_size = wf.tell()
956
for i in range(vers):
957
pb.update('checking sizes', i, vers)
958
for origin, lineno, line in w._extract([i]):
963
print 'versions %9d' % vers
964
print 'weave file %9d bytes' % weave_size
965
print 'total contents %9d bytes' % total
966
print 'compression ratio %9.2fx' % (float(total) / float(weave_size))
969
print 'average size %9d bytes' % avg
970
print 'relative size %9.2fx' % (float(weave_size) / float(avg))
974
print """bzr weave tool
976
Experimental tool for weave algorithm.
980
Create an empty weave file
981
weave get WEAVEFILE VERSION
982
Write out specified version.
983
weave check WEAVEFILE
984
Check consistency of all versions.
986
Display table of contents.
987
weave add WEAVEFILE NAME [BASE...] < NEWTEXT
988
Add NEWTEXT, with specified parent versions.
989
weave annotate WEAVEFILE VERSION
990
Display origin of each line.
991
weave mash WEAVEFILE VERSION...
992
Display composite of all selected versions.
993
weave merge WEAVEFILE VERSION1 VERSION2 > OUT
994
Auto-merge two versions and display conflicts.
995
weave diff WEAVEFILE VERSION1 VERSION2
996
Show differences between two versions.
1000
% weave init foo.weave
1002
% weave add foo.weave ver0 < foo.txt
1005
(create updated version)
1007
% weave get foo.weave 0 | diff -u - foo.txt
1008
% weave add foo.weave ver1 0 < foo.txt
1011
% weave get foo.weave 0 > foo.txt (create forked version)
1013
% weave add foo.weave ver2 0 < foo.txt
1016
% weave merge foo.weave 1 2 > foo.txt (merge them)
1017
% vi foo.txt (resolve conflicts)
1018
% weave add foo.weave merged 1 2 < foo.txt (commit merged version)
1030
# in case we're run directly from the subdirectory
1031
sys.path.append('..')
1033
from bzrlib.weavefile import write_weave, read_weave
1034
from bzrlib.progress import ProgressBar
1049
return read_weave(file(argv[2], 'rb'))
1055
# at the moment, based on everything in the file
1057
parents = map(int, argv[4:])
1058
lines = sys.stdin.readlines()
1059
ver = w.add(name, parents, lines)
1060
write_weave(w, file(argv[2], 'wb'))
1061
print 'added version %r %d' % (name, ver)
1064
if os.path.exists(fn):
1065
raise IOError("file exists")
1067
write_weave(w, file(fn, 'wb'))
1068
elif cmd == 'get': # get one version
1070
sys.stdout.writelines(w.get_iter(int(argv[3])))
1072
elif cmd == 'mash': # get composite
1074
sys.stdout.writelines(w.mash_iter(map(int, argv[3:])))
1077
from difflib import unified_diff
1080
v1, v2 = map(int, argv[3:5])
1083
diff_gen = unified_diff(lines1, lines2,
1084
'%s version %d' % (fn, v1),
1085
'%s version %d' % (fn, v2))
1086
sys.stdout.writelines(diff_gen)
1088
elif cmd == 'annotate':
1090
# newline is added to all lines regardless; too hard to get
1091
# reasonable formatting otherwise
1093
for origin, text in w.annotate(int(argv[3])):
1094
text = text.rstrip('\r\n')
1096
print ' | %s' % (text)
1098
print '%5d | %s' % (origin, text)
1104
elif cmd == 'stats':
1105
weave_stats(argv[2], ProgressBar())
1107
elif cmd == 'check':
1112
print '%d versions ok' % w.numversions()
1114
elif cmd == 'inclusions':
1116
print ' '.join(map(str, w.inclusions([int(argv[3])])))
1118
elif cmd == 'parents':
1120
print ' '.join(map(str, w._parents[int(argv[3])]))
1122
elif cmd == 'plan-merge':
1124
for state, line in w.plan_merge(int(argv[3]), int(argv[4])):
1126
print '%14s | %s' % (state, line),
1128
elif cmd == 'merge':
1130
p = w.plan_merge(int(argv[3]), int(argv[4]))
1131
sys.stdout.writelines(w.weave_merge(p))
1133
elif cmd == 'mash-merge':
1139
v1, v2 = map(int, argv[3:5])
1141
basis = w.inclusions([v1]).intersection(w.inclusions([v2]))
1143
base_lines = list(w.mash_iter(basis))
1144
a_lines = list(w.get(v1))
1145
b_lines = list(w.get(v2))
1147
from bzrlib.merge3 import Merge3
1148
m3 = Merge3(base_lines, a_lines, b_lines)
1150
name_a = 'version %d' % v1
1151
name_b = 'version %d' % v2
1152
sys.stdout.writelines(m3.merge_lines(name_a=name_a, name_b=name_b))
1154
raise ValueError('unknown command %r' % cmd)
1158
def profile_main(argv):
1159
import tempfile, hotshot, hotshot.stats
1161
prof_f = tempfile.NamedTemporaryFile()
1163
prof = hotshot.Profile(prof_f.name)
1165
ret = prof.runcall(main, argv)
1168
stats = hotshot.stats.load(prof_f.name)
1170
stats.sort_stats('cumulative')
1171
## XXX: Might like to write to stderr or the trace file instead but
1172
## print_stats seems hardcoded to stdout
1173
stats.print_stats(20)
1178
if __name__ == '__main__':
1180
if '--profile' in sys.argv:
1182
args.remove('--profile')
1183
sys.exit(profile_main(args))
1185
sys.exit(main(sys.argv))
added
removed
bzrlib/weave.py
