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
73
from ..lazy_import import lazy_import
74
lazy_import(globals(), """
75
from breezy import tsort
81
from ..errors import (
82
RevisionAlreadyPresent,
84
UnavailableRepresentation,
86
from ..osutils import dirname, sha, sha_strings, split_lines
87
from ..revision import NULL_REVISION
88
from ..sixish import (
91
from ..trace import mutter
92
from .versionedfile import (
99
from .weavefile import _read_weave_v5, write_weave_v5
102
class WeaveError(errors.BzrError):
104
_fmt = "Error in processing weave: %(msg)s"
106
def __init__(self, msg=None):
107
errors.BzrError.__init__(self)
111
class WeaveRevisionAlreadyPresent(WeaveError):
113
_fmt = "Revision {%(revision_id)s} already present in %(weave)s"
115
def __init__(self, revision_id, weave):
117
WeaveError.__init__(self)
118
self.revision_id = revision_id
122
class WeaveRevisionNotPresent(WeaveError):
124
_fmt = "Revision {%(revision_id)s} not present in %(weave)s"
126
def __init__(self, revision_id, weave):
127
WeaveError.__init__(self)
128
self.revision_id = revision_id
71
from difflib import SequenceMatcher
73
from bzrlib.trace import mutter
76
class WeaveError(Exception):
77
"""Exception in processing weave"""
132
80
class WeaveFormatError(WeaveError):
134
_fmt = "Weave invariant violated: %(what)s"
136
def __init__(self, what):
137
WeaveError.__init__(self)
81
"""Weave invariant violated"""
141
84
class WeaveParentMismatch(WeaveError):
143
_fmt = "Parents are mismatched between two revisions. %(msg)s"
146
class WeaveInvalidChecksum(WeaveError):
148
_fmt = "Text did not match its checksum: %(msg)s"
151
class WeaveTextDiffers(WeaveError):
153
_fmt = ("Weaves differ on text content. Revision:"
154
" {%(revision_id)s}, %(weave_a)s, %(weave_b)s")
156
def __init__(self, revision_id, weave_a, weave_b):
157
WeaveError.__init__(self)
158
self.revision_id = revision_id
159
self.weave_a = weave_a
160
self.weave_b = weave_b
163
class WeaveContentFactory(ContentFactory):
164
"""Content factory for streaming from weaves.
166
:seealso ContentFactory:
169
def __init__(self, version, weave):
170
"""Create a WeaveContentFactory for version from weave."""
171
ContentFactory.__init__(self)
172
self.sha1 = weave.get_sha1s([version])[version]
173
self.key = (version,)
174
parents = weave.get_parent_map([version])[version]
175
self.parents = tuple((parent,) for parent in parents)
176
self.storage_kind = 'fulltext'
179
def get_bytes_as(self, storage_kind):
180
if storage_kind == 'fulltext':
181
return self._weave.get_text(self.key[-1])
182
elif storage_kind in ('chunked', 'lines'):
183
return self._weave.get_lines(self.key[-1])
185
raise UnavailableRepresentation(self.key, storage_kind, 'fulltext')
187
def iter_bytes_as(self, storage_kind):
188
if storage_kind in ('chunked', 'lines'):
189
return iter(self._weave.get_lines(self.key[-1]))
191
raise UnavailableRepresentation(self.key, storage_kind, 'fulltext')
194
class Weave(VersionedFile):
85
"""Parents are mismatched between two revisions."""
195
89
"""weave - versioned text file storage.
197
91
A Weave manages versions of line-based text files, keeping track
198
92
of the originating version for each line.
337
203
if not isinstance(other, Weave):
339
205
return self._parents == other._parents \
340
and self._weave == other._weave \
341
and self._sha1s == other._sha1s
206
and self._weave == other._weave \
207
and self._sha1s == other._sha1s
343
210
def __ne__(self, other):
344
211
return not self.__eq__(other)
346
def _idx_to_name(self, version):
347
return self._names[version]
349
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):
350
223
"""Convert symbolic version name to index."""
351
if not self._allow_reserved:
352
self.check_not_reserved_id(name)
354
225
return self._name_map[name]
356
raise RevisionNotPresent(name, self._weave_name)
359
"""See VersionedFile.versions."""
360
return self._names[:]
362
def has_version(self, version_id):
363
"""See VersionedFile.has_version."""
364
return (version_id in self._name_map)
366
__contains__ = has_version
368
def get_record_stream(self, versions, ordering, include_delta_closure):
369
"""Get a stream of records for versions.
371
:param versions: The versions to include. Each version is a tuple
373
:param ordering: Either 'unordered' or 'topological'. A topologically
374
sorted stream has compression parents strictly before their
376
:param include_delta_closure: If True then the closure across any
377
compression parents will be included (in the opaque data).
378
:return: An iterator of ContentFactory objects, each of which is only
379
valid until the iterator is advanced.
381
versions = [version[-1] for version in versions]
382
if ordering == 'topological':
383
parents = self.get_parent_map(versions)
384
new_versions = tsort.topo_sort(parents)
385
new_versions.extend(set(versions).difference(set(parents)))
386
versions = new_versions
387
elif ordering == 'groupcompress':
388
parents = self.get_parent_map(versions)
389
new_versions = sort_groupcompress(parents)
390
new_versions.extend(set(versions).difference(set(parents)))
391
versions = new_versions
392
for version in versions:
394
yield WeaveContentFactory(version, self)
396
yield AbsentContentFactory((version,))
398
def get_parent_map(self, version_ids):
399
"""See VersionedFile.get_parent_map."""
401
for version_id in version_ids:
402
if version_id == NULL_REVISION:
407
map(self._idx_to_name,
408
self._parents[self._lookup(version_id)]))
409
except RevisionNotPresent:
411
result[version_id] = parents
414
def get_parents_with_ghosts(self, version_id):
415
raise NotImplementedError(self.get_parents_with_ghosts)
417
def insert_record_stream(self, stream):
418
"""Insert a record stream into this versioned file.
420
:param stream: A stream of records to insert.
422
:seealso VersionedFile.get_record_stream:
425
for record in stream:
426
# Raise an error when a record is missing.
427
if record.storage_kind == 'absent':
428
raise RevisionNotPresent([record.key[0]], self)
429
# adapt to non-tuple interface
430
parents = [parent[0] for parent in record.parents]
431
if record.storage_kind in ('fulltext', 'chunked', 'lines'):
433
record.key[0], parents,
434
record.get_bytes_as('lines'))
436
adapter_key = record.storage_kind, 'lines'
438
adapter = adapters[adapter_key]
440
adapter_factory = adapter_registry.get(adapter_key)
441
adapter = adapter_factory(self)
442
adapters[adapter_key] = adapter
443
lines = adapter.get_bytes(record, 'lines')
445
self.add_lines(record.key[0], parents, lines)
446
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]
449
239
def _check_repeated_add(self, name, parents, text, sha1):
450
240
"""Check that a duplicated add is OK.
452
242
If it is, return the (old) index; otherwise raise an exception.
454
idx = self._lookup(name)
455
if sorted(self._parents[idx]) != sorted(parents) \
456
or sha1 != self._sha1s[idx]:
457
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)
460
def _add_lines(self, version_id, parents, lines, parent_texts,
461
left_matching_blocks, nostore_sha, random_id,
463
"""See VersionedFile.add_lines."""
464
idx = self._add(version_id, lines, list(map(self._lookup, parents)),
465
nostore_sha=nostore_sha)
466
return sha_strings(lines), sum(map(len, lines)), idx
468
def _add(self, version_id, lines, parents, sha1=None, nostore_sha=None):
255
def add(self, name, parents, text, sha1=None):
469
256
"""Add a single text on top of the weave.
471
258
Returns the index number of the newly added version.
474
261
Symbolic name for this version.
475
262
(Typically the revision-id of the revision that added it.)
476
If None, a name will be allocated based on the hash. (sha1:SHAHASH)
479
265
List or set of direct parent version numbers.
482
268
Sequence of lines to be added in the new version.
484
:param nostore_sha: See VersionedFile.add_lines.
270
sha -- SHA-1 of the file, if known. This is trusted to be
486
self._check_lines_not_unicode(lines)
487
self._check_lines_are_lines(lines)
489
sha1 = sha_strings(lines)
490
if sha1 == nostore_sha:
491
raise errors.ExistingContent
492
if version_id is None:
493
version_id = b"sha1:" + sha1
494
if version_id in self._name_map:
495
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)
497
282
self._check_versions(parents)
283
## self._check_lines(text)
498
284
new_version = len(self._parents)
500
# if we abort after here the (in-memory) weave will be corrupt because
501
# only some fields are updated
502
# XXX: FIXME implement a succeed-or-fail of the rest of this routine.
503
# - Robert Collins 20060226
287
# if we abort after here the (in-memory) weave will be corrupt because only
288
# some fields are updated
504
289
self._parents.append(parents[:])
505
290
self._sha1s.append(sha1)
506
self._names.append(version_id)
507
self._name_map[version_id] = new_version
291
self._names.append(name)
292
self._name_map[name] = new_version
510
296
# special case; adding with no parents revision; can do
511
297
# this more quickly by just appending unconditionally.
512
298
# even more specially, if we're adding an empty text we
513
299
# need do nothing at all.
515
self._weave.append((b'{', new_version))
516
self._weave.extend(lines)
517
self._weave.append((b'}', None))
301
self._weave.append(('{', new_version))
302
self._weave.extend(text)
303
self._weave.append(('}', None))
518
305
return new_version
520
307
if len(parents) == 1:
608
445
except IndexError:
609
446
raise IndexError("invalid version number %r" % i)
611
def _compatible_parents(self, my_parents, other_parents):
612
"""During join check that other_parents are joinable with my_parents.
614
Joinable is defined as 'is a subset of' - supersets may require
615
regeneration of diffs, but subsets do not.
617
return len(other_parents.difference(my_parents)) == 0
619
def annotate(self, version_id):
620
"""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.
622
456
The index indicates when the line originated in the weave."""
623
incls = [self._lookup(version_id)]
624
return [(self._idx_to_name(origin), text) for origin, lineno, text in
625
self._extract(incls)]
627
def iter_lines_added_or_present_in_versions(self, version_ids=None,
629
"""See VersionedFile.iter_lines_added_or_present_in_versions()."""
630
if version_ids is None:
631
version_ids = self.versions()
632
version_ids = set(version_ids)
633
for lineno, inserted, deletes, line in self._walk_internal(
635
if inserted not in version_ids:
637
if not line.endswith(b'\n'):
638
yield line + b'\n', inserted
642
def _walk_internal(self, version_ids=None):
643
"""Helper method for weave actions."""
648
lineno = 0 # line of weave, 0-based
650
for l in self._weave:
651
if l.__class__ == tuple:
654
istack.append(self._names[v])
658
dset.add(self._names[v])
660
dset.remove(self._names[v])
662
raise WeaveFormatError('unexpected instruction %r' % v)
664
yield lineno, istack[-1], frozenset(dset), l
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.
473
lineno = 0 # line of weave, 0-based
475
for l in self._weave:
476
if isinstance(l, tuple):
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))
668
raise WeaveFormatError("unclosed insertion blocks "
669
"at end of weave: %s" % istack)
552
raise WFE("unclosed insertion blocks at end of weave",
671
raise WeaveFormatError(
672
"unclosed deletion blocks at end of weave: %s" % dset)
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()
674
677
def plan_merge(self, ver_a, ver_b):
675
678
"""Return pseudo-annotation indicating how the two versions merge.
716
718
# not in either revision
717
719
yield 'irrelevant', line
719
def _extract(self, versions):
720
"""Yield annotation of lines in included set.
722
Yields a sequence of tuples (origin, lineno, text), where
723
origin is the origin version, lineno the index in the weave,
724
and text the text of the line.
726
The set typically but not necessarily corresponds to a version.
729
if not isinstance(i, int):
732
included = self._inclusions(versions)
738
lineno = 0 # line of weave, 0-based
745
# 449 0 4474.6820 2356.5590 breezy.weave:556(_extract)
746
# +285282 0 1676.8040 1676.8040 +<isinstance>
747
# 1.6 seconds in 'isinstance'.
748
# changing the first isinstance:
749
# 449 0 2814.2660 1577.1760 breezy.weave:556(_extract)
750
# +140414 0 762.8050 762.8050 +<isinstance>
751
# note that the inline time actually dropped (less function calls)
752
# and total processing time was halved.
753
# we're still spending ~1/4 of the method in isinstance though.
754
# so lets hard code the acceptable string classes we expect:
755
# 449 0 1202.9420 786.2930 breezy.weave:556(_extract)
756
# +71352 0 377.5560 377.5560 +<method 'append' of 'list'
758
# yay, down to ~1/4 the initial extract time, and our inline time
759
# has shrunk again, with isinstance no longer dominating.
760
# tweaking the stack inclusion test to use a set gives:
761
# 449 0 1122.8030 713.0080 breezy.weave:556(_extract)
762
# +71352 0 354.9980 354.9980 +<method 'append' of 'list'
764
# - a 5% win, or possibly just noise. However with large istacks that
765
# 'in' test could dominate, so I'm leaving this change in place - when
766
# its fast enough to consider profiling big datasets we can review.
768
for l in self._weave:
769
if l.__class__ == tuple:
776
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
784
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':
787
isactive = (not dset) and istack and (
788
istack[-1] in included)
790
result.append((istack[-1], lineno, l))
793
raise WeaveFormatError("unclosed insertion blocks "
794
"at end of weave: %s" % istack)
796
raise WeaveFormatError(
797
"unclosed deletion blocks at end of weave: %s" % dset)
800
def _maybe_lookup(self, name_or_index):
801
"""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().
805
# GZ 2017-04-01: This used to check for long as well, but I don't think
806
# there are python implementations with sys.maxsize > sys.maxint
807
if isinstance(name_or_index, int):
810
return self._lookup(name_or_index)
812
def get_lines(self, version_id):
813
"""See VersionedFile.get_lines()."""
814
int_index = self._maybe_lookup(version_id)
815
result = [line for (origin, lineno, line)
816
in self._extract([int_index])]
817
expected_sha1 = self._sha1s[int_index]
818
measured_sha1 = sha_strings(result)
819
if measured_sha1 != expected_sha1:
820
raise WeaveInvalidChecksum(
821
'file %s, revision %s, expected: %s, measured %s'
822
% (self._weave_name, version_id,
823
expected_sha1, measured_sha1))
826
def get_sha1s(self, version_ids):
827
"""See VersionedFile.get_sha1s()."""
829
for v in version_ids:
830
result[v] = self._sha1s[self._lookup(v)]
833
def num_versions(self):
834
"""How many versions are in this weave?"""
835
return len(self._parents)
837
__len__ = num_versions
839
def check(self, progress_bar=None):
840
# TODO evaluate performance hit of using string sets in this routine.
841
# TODO: check no circular inclusions
842
# TODO: create a nested progress bar
843
for version in range(self.num_versions()):
844
inclusions = list(self._parents[version])
847
if inclusions[-1] >= version:
848
raise WeaveFormatError(
849
"invalid included version %d for index %d"
850
% (inclusions[-1], version))
852
# try extracting all versions; parallel extraction is used
853
nv = self.num_versions()
858
# For creating the ancestry, IntSet is much faster (3.7s vs 0.17s)
859
# The problem is that set membership is much more expensive
860
name = self._idx_to_name(i)
864
for p in self._parents[i]:
865
new_inc.update(inclusions[self._idx_to_name(p)])
867
if set(new_inc) != set(self.get_ancestry(name)):
868
raise AssertionError(
870
% (set(new_inc), set(self.get_ancestry(name))))
871
inclusions[name] = new_inc
873
nlines = len(self._weave)
875
update_text = 'checking weave'
877
short_name = os.path.basename(self._weave_name)
878
update_text = 'checking %s' % (short_name,)
879
update_text = update_text[:25]
881
for lineno, insert, deleteset, line in self._walk_internal():
883
progress_bar.update(update_text, lineno, nlines)
885
for name, name_inclusions in inclusions.items():
886
# The active inclusion must be an ancestor,
887
# and no ancestors must have deleted this line,
888
# because we don't support resurrection.
889
if ((insert in name_inclusions) and
890
not (deleteset & name_inclusions)):
891
sha1s[name].update(line)
894
version = self._idx_to_name(i)
895
hd = sha1s[version].hexdigest().encode()
896
expected = self._sha1s[i]
898
raise WeaveInvalidChecksum(
899
"mismatched sha1 for version %s: "
900
"got %s, expected %s"
901
% (version, hd, expected))
903
# TODO: check insertions are properly nested, that there are
904
# no lines outside of insertion blocks, that deletions are
905
# 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)
907
802
def _imported_parents(self, other, other_idx):
908
803
"""Return list of parents in self corresponding to indexes in other."""
910
805
for parent_idx in other._parents[other_idx]:
911
806
parent_name = other._names[parent_idx]
912
if parent_name not in self._name_map:
807
if parent_name not in self._names:
913
808
# should not be possible
914
raise WeaveError("missing parent {%s} of {%s} in %r"
809
raise WeaveError("missing parent {%s} of {%s} in %r"
915
810
% (parent_name, other._name_map[other_idx], self))
916
811
new_parents.append(self._name_map[parent_name])
917
812
return new_parents
925
820
* the same direct parents (by name, not index, and disregarding
928
823
If present & correct return True;
929
if not present in self return False;
824
if not present in self return False;
930
825
if inconsistent raise error."""
931
826
this_idx = self._name_map.get(name, -1)
932
827
if this_idx != -1:
933
828
if self._sha1s[this_idx] != other._sha1s[other_idx]:
934
raise WeaveTextDiffers(name, self, other)
829
raise WeaveError("inconsistent texts for version {%s} "
830
"when joining weaves"
935
832
self_parents = self._parents[this_idx]
936
833
other_parents = other._parents[other_idx]
937
n1 = {self._names[i] for i in self_parents}
938
n2 = {other._names[i] for i in other_parents}
939
if not self._compatible_parents(n1, n2):
940
raise WeaveParentMismatch(
941
"inconsistent parents "
834
n1 = [self._names[i] for i in self_parents]
835
n2 = [other._names[i] for i in other_parents]
839
raise WeaveParentMismatch("inconsistent parents "
942
840
"for version {%s}: %s vs %s" % (name, n1, n2))
944
842
return True # ok!
948
def _reweave(self, other, pb, msg):
949
"""Reweave self with other - internal helper for join().
951
:param other: The other weave to merge
952
:param pb: An optional progress bar, indicating how far done we are
953
:param msg: An optional message for the progress
955
new_weave = _reweave(self, other, pb=pb, msg=msg)
956
self._copy_weave_content(new_weave)
958
def _copy_weave_content(self, otherweave):
959
"""adsorb the content from otherweave."""
846
def reweave(self, other):
847
"""Reweave self with other."""
848
new_weave = reweave(self, other)
960
849
for attr in self.__slots__:
961
if attr != '_weave_name':
962
setattr(self, attr, copy(getattr(otherweave, attr)))
965
class WeaveFile(Weave):
966
"""A WeaveFile represents a Weave on disk and writes on change."""
968
WEAVE_SUFFIX = '.weave'
970
def __init__(self, name, transport, filemode=None, create=False,
971
access_mode='w', get_scope=None):
972
"""Create a WeaveFile.
974
:param create: If not True, only open an existing knit.
976
super(WeaveFile, self).__init__(name, access_mode, get_scope=get_scope,
977
allow_reserved=False)
978
self._transport = transport
979
self._filemode = filemode
981
f = self._transport.get(name + WeaveFile.WEAVE_SUFFIX)
982
_read_weave_v5(BytesIO(f.read()), self)
983
except errors.NoSuchFile:
989
def _add_lines(self, version_id, parents, lines, parent_texts,
990
left_matching_blocks, nostore_sha, random_id,
992
"""Add a version and save the weave."""
993
self.check_not_reserved_id(version_id)
994
result = super(WeaveFile, self)._add_lines(
995
version_id, parents, lines, parent_texts, left_matching_blocks,
996
nostore_sha, random_id, check_content)
1000
def copy_to(self, name, transport):
1001
"""See VersionedFile.copy_to()."""
1002
# as we are all in memory always, just serialise to the new place.
1004
write_weave_v5(self, sio)
1006
transport.put_file(name + WeaveFile.WEAVE_SUFFIX, sio, self._filemode)
1009
"""Save the weave."""
1010
self._check_write_ok()
1012
write_weave_v5(self, sio)
1014
bytes = sio.getvalue()
1015
path = self._weave_name + WeaveFile.WEAVE_SUFFIX
1017
self._transport.put_bytes(path, bytes, self._filemode)
1018
except errors.NoSuchFile:
1019
self._transport.mkdir(dirname(path))
1020
self._transport.put_bytes(path, bytes, self._filemode)
1024
"""See VersionedFile.get_suffixes()."""
1025
return [WeaveFile.WEAVE_SUFFIX]
1027
def insert_record_stream(self, stream):
1028
super(WeaveFile, self).insert_record_stream(stream)
1032
def _reweave(wa, wb, pb=None, msg=None):
850
setattr(self, attr, getattr(new_weave, attr))
1033
854
"""Combine two weaves and return the result.
1035
This works even if a revision R has different parents in
856
This works even if a revision R has different parents in
1036
857
wa and wb. In the resulting weave all the parents are given.
1038
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
1039
860
of the versions in the two inputs. More efficient approaches
1040
might be possible but it should only be necessary to do
1041
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
1044
:param pb: An optional progress bar, indicating how far done we are
1045
:param msg: An optional message for the progress
867
queue_a = range(wa.numversions())
868
queue_b = range(wb.numversions())
1048
869
# first determine combined parents of all versions
1049
870
# map from version name -> all parent names
1050
871
combined_parents = _reweave_parent_graphs(wa, wb)
1051
872
mutter("combined parents: %r", combined_parents)
1052
order = tsort.topo_sort(combined_parents.items())
873
order = _make_reweave_order(wa._names, wb._names, combined_parents)
1053
874
mutter("order to reweave: %r", order)
1058
for idx, name in enumerate(order):
1060
pb.update(msg, idx, len(order))
1061
876
if name in wa._name_map:
1062
877
lines = wa.get_lines(name)
1063
878
if name in wb._name_map:
1064
lines_b = wb.get_lines(name)
1065
if lines != lines_b:
1066
mutter('Weaves differ on content. rev_id {%s}', name)
1067
mutter('weaves: %s, %s', wa._weave_name, wb._weave_name)
1069
lines = list(difflib.unified_diff(lines, lines_b,
1070
wa._weave_name, wb._weave_name))
1071
mutter('lines:\n%s', ''.join(lines))
1072
raise WeaveTextDiffers(name, wa, wb)
879
assert lines == wb.get_lines(name)
1074
881
lines = wb.get_lines(name)
1075
wr._add(name, lines, [wr._lookup(i) for i in combined_parents[name]])
882
wr.add(name, combined_parents[name], lines)
1079
886
def _reweave_parent_graphs(wa, wb):
1080
887
"""Return combined parent ancestry for two weaves.
1082
889
Returned as a list of (version_name, set(parent_names))"""
1084
891
for weave in [wa, wb]:
1085
892
for idx, name in enumerate(weave._names):
1086
893
p = combined.setdefault(name, set())
1087
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
