61
59
# where the basis and destination are unchanged.
63
61
# FIXME: Sometimes we will be given a parents list for a revision
64
# that includes some redundant parents (i.e. already a parent of
65
# something in the list.) We should eliminate them. This can
62
# that includes some redundant parents (i.e. already a parent of
63
# something in the list.) We should eliminate them. This can
66
64
# be done fairly efficiently because the sequence numbers constrain
67
65
# the possible relationships.
71
from difflib import SequenceMatcher
67
# FIXME: the conflict markers should be *7* characters
70
from cStringIO import StringIO
73
from bzrlib.lazy_import import lazy_import
74
lazy_import(globals(), """
75
from bzrlib import tsort
81
from bzrlib.errors import (WeaveError, WeaveFormatError, WeaveParentMismatch,
82
RevisionAlreadyPresent,
84
UnavailableRepresentation,
86
from bzrlib.osutils import dirname, sha, sha_strings, split_lines
87
import bzrlib.patiencediff
88
from bzrlib.revision import NULL_REVISION
89
from bzrlib.symbol_versioning import *
73
90
from bzrlib.trace import mutter
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."""
91
from bzrlib.versionedfile import (
98
from bzrlib.weavefile import _read_weave_v5, write_weave_v5
101
class WeaveContentFactory(ContentFactory):
102
"""Content factory for streaming from weaves.
104
:seealso ContentFactory:
107
def __init__(self, version, weave):
108
"""Create a WeaveContentFactory for version from weave."""
109
ContentFactory.__init__(self)
110
self.sha1 = weave.get_sha1s([version])[version]
111
self.key = (version,)
112
parents = weave.get_parent_map([version])[version]
113
self.parents = tuple((parent,) for parent in parents)
114
self.storage_kind = 'fulltext'
117
def get_bytes_as(self, storage_kind):
118
if storage_kind == 'fulltext':
119
return self._weave.get_text(self.key[-1])
120
elif storage_kind == 'chunked':
121
return self._weave.get_lines(self.key[-1])
123
raise UnavailableRepresentation(self.key, storage_kind, 'fulltext')
126
class Weave(VersionedFile):
89
127
"""weave - versioned text file storage.
91
129
A Weave manages versions of line-based text files, keeping track
92
130
of the originating version for each line.
205
270
return self._parents == other._parents \
206
271
and self._weave == other._weave \
207
and self._sha1s == other._sha1s
272
and self._sha1s == other._sha1s
210
274
def __ne__(self, other):
211
275
return not self.__eq__(other)
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):
277
def _idx_to_name(self, version):
278
return self._names[version]
280
def _lookup(self, name):
223
281
"""Convert symbolic version name to index."""
282
if not self._allow_reserved:
283
self.check_not_reserved_id(name)
225
285
return self._name_map[name]
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]
287
raise RevisionNotPresent(name, self._weave_name)
290
"""See VersionedFile.versions."""
291
return self._names[:]
293
def has_version(self, version_id):
294
"""See VersionedFile.has_version."""
295
return (version_id in self._name_map)
297
__contains__ = has_version
299
def get_record_stream(self, versions, ordering, include_delta_closure):
300
"""Get a stream of records for versions.
302
:param versions: The versions to include. Each version is a tuple
304
:param ordering: Either 'unordered' or 'topological'. A topologically
305
sorted stream has compression parents strictly before their
307
:param include_delta_closure: If True then the closure across any
308
compression parents will be included (in the opaque data).
309
:return: An iterator of ContentFactory objects, each of which is only
310
valid until the iterator is advanced.
312
versions = [version[-1] for version in versions]
313
if ordering == 'topological':
314
parents = self.get_parent_map(versions)
315
new_versions = tsort.topo_sort(parents)
316
new_versions.extend(set(versions).difference(set(parents)))
317
versions = new_versions
318
elif ordering == 'groupcompress':
319
parents = self.get_parent_map(versions)
320
new_versions = sort_groupcompress(parents)
321
new_versions.extend(set(versions).difference(set(parents)))
322
versions = new_versions
323
for version in versions:
325
yield WeaveContentFactory(version, self)
327
yield AbsentContentFactory((version,))
329
def get_parent_map(self, version_ids):
330
"""See VersionedFile.get_parent_map."""
332
for version_id in version_ids:
333
if version_id == NULL_REVISION:
338
map(self._idx_to_name,
339
self._parents[self._lookup(version_id)]))
340
except RevisionNotPresent:
342
result[version_id] = parents
345
def get_parents_with_ghosts(self, version_id):
346
raise NotImplementedError(self.get_parents_with_ghosts)
348
def insert_record_stream(self, stream):
349
"""Insert a record stream into this versioned file.
351
:param stream: A stream of records to insert.
353
:seealso VersionedFile.get_record_stream:
356
for record in stream:
357
# Raise an error when a record is missing.
358
if record.storage_kind == 'absent':
359
raise RevisionNotPresent([record.key[0]], self)
360
# adapt to non-tuple interface
361
parents = [parent[0] for parent in record.parents]
362
if (record.storage_kind == 'fulltext'
363
or record.storage_kind == 'chunked'):
364
self.add_lines(record.key[0], parents,
365
osutils.chunks_to_lines(record.get_bytes_as('chunked')))
367
adapter_key = record.storage_kind, 'fulltext'
369
adapter = adapters[adapter_key]
371
adapter_factory = adapter_registry.get(adapter_key)
372
adapter = adapter_factory(self)
373
adapters[adapter_key] = adapter
374
lines = split_lines(adapter.get_bytes(record))
376
self.add_lines(record.key[0], parents, lines)
377
except RevisionAlreadyPresent:
239
380
def _check_repeated_add(self, name, parents, text, sha1):
240
381
"""Check that a duplicated add is OK.
242
383
If it is, return the (old) index; otherwise raise an exception.
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)
385
idx = self._lookup(name)
386
if sorted(self._parents[idx]) != sorted(parents) \
387
or sha1 != self._sha1s[idx]:
388
raise RevisionAlreadyPresent(name, self._weave_name)
255
def add(self, name, parents, text, sha1=None):
391
def _add_lines(self, version_id, parents, lines, parent_texts,
392
left_matching_blocks, nostore_sha, random_id, check_content):
393
"""See VersionedFile.add_lines."""
394
idx = self._add(version_id, lines, map(self._lookup, parents),
395
nostore_sha=nostore_sha)
396
return sha_strings(lines), sum(map(len, lines)), idx
398
def _add(self, version_id, lines, parents, sha1=None, nostore_sha=None):
256
399
"""Add a single text on top of the weave.
258
401
Returns the index number of the newly added version.
261
404
Symbolic name for this version.
262
405
(Typically the revision-id of the revision that added it.)
406
If None, a name will be allocated based on the hash. (sha1:SHAHASH)
265
409
List or set of direct parent version numbers.
268
412
Sequence of lines to be added in the new version.
270
sha -- SHA-1 of the file, if known. This is trusted to be
414
:param nostore_sha: See VersionedFile.add_lines.
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)
416
self._check_lines_not_unicode(lines)
417
self._check_lines_are_lines(lines)
419
sha1 = sha_strings(lines)
420
if sha1 == nostore_sha:
421
raise errors.ExistingContent
422
if version_id is None:
423
version_id = "sha1:" + sha1
424
if version_id in self._name_map:
425
return self._check_repeated_add(version_id, parents, lines, sha1)
282
427
self._check_versions(parents)
283
## self._check_lines(text)
428
## self._check_lines(lines)
284
429
new_version = len(self._parents)
287
431
# if we abort after here the (in-memory) weave will be corrupt because only
288
432
# some fields are updated
433
# XXX: FIXME implement a succeed-or-fail of the rest of this routine.
434
# - Robert Collins 20060226
289
435
self._parents.append(parents[:])
290
436
self._sha1s.append(sha1)
291
self._names.append(name)
292
self._name_map[name] = new_version
437
self._names.append(version_id)
438
self._name_map[version_id] = new_version
296
442
# special case; adding with no parents revision; can do
297
443
# this more quickly by just appending unconditionally.
298
444
# even more specially, if we're adding an empty text we
299
445
# need do nothing at all.
301
447
self._weave.append(('{', new_version))
302
self._weave.extend(text)
448
self._weave.extend(lines)
303
449
self._weave.append(('}', None))
305
450
return new_version
307
452
if len(parents) == 1:
445
554
except IndexError:
446
555
raise IndexError("invalid version number %r" % i)
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.
557
def _compatible_parents(self, my_parents, other_parents):
558
"""During join check that other_parents are joinable with my_parents.
560
Joinable is defined as 'is a subset of' - supersets may require
561
regeneration of diffs, but subsets do not.
563
return len(other_parents.difference(my_parents)) == 0
565
def annotate(self, version_id):
566
"""Return a list of (version-id, line) tuples for version_id.
456
568
The index indicates when the line originated in the weave."""
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.
569
incls = [self._lookup(version_id)]
570
return [(self._idx_to_name(origin), text) for origin, lineno, text in
571
self._extract(incls)]
573
def iter_lines_added_or_present_in_versions(self, version_ids=None,
575
"""See VersionedFile.iter_lines_added_or_present_in_versions()."""
576
if version_ids is None:
577
version_ids = self.versions()
578
version_ids = set(version_ids)
579
for lineno, inserted, deletes, line in self._walk_internal(version_ids):
580
if inserted not in version_ids: continue
582
yield line + '\n', inserted
586
def _walk_internal(self, version_ids=None):
587
"""Helper method for weave actions."""
473
592
lineno = 0 # line of weave, 0-based
475
594
for l in self._weave:
476
if isinstance(l, tuple):
595
if l.__class__ == tuple:
599
istack.append(self._names[v])
603
dset.add(self._names[v])
605
dset.remove(self._names[v])
489
raise WeaveFormatError('unexpected instruction %r'
607
raise WeaveFormatError('unexpected instruction %r' % v)
492
assert isinstance(l, basestring)
494
yield lineno, istack[-1], dset, l
609
yield lineno, istack[-1], frozenset(dset), l
613
raise WeaveFormatError("unclosed insertion blocks "
614
"at end of weave: %s" % istack)
616
raise WeaveFormatError("unclosed deletion blocks at end of weave: %s"
619
def plan_merge(self, ver_a, ver_b):
620
"""Return pseudo-annotation indicating how the two versions merge.
622
This is computed between versions a and b and their common
625
Weave lines present in none of them are skipped entirely.
627
inc_a = set(self.get_ancestry([ver_a]))
628
inc_b = set(self.get_ancestry([ver_b]))
629
inc_c = inc_a & inc_b
631
for lineno, insert, deleteset, line in self._walk_internal([ver_a, ver_b]):
632
if deleteset & inc_c:
633
# killed in parent; can't be in either a or b
634
# not relevant to our work
635
yield 'killed-base', line
636
elif insert in inc_c:
637
# was inserted in base
638
killed_a = bool(deleteset & inc_a)
639
killed_b = bool(deleteset & inc_b)
640
if killed_a and killed_b:
641
yield 'killed-both', line
643
yield 'killed-a', line
645
yield 'killed-b', line
647
yield 'unchanged', line
648
elif insert in inc_a:
649
if deleteset & inc_a:
650
yield 'ghost-a', line
654
elif insert in inc_b:
655
if deleteset & inc_b:
656
yield 'ghost-b', line
660
# not in either revision
661
yield 'irrelevant', line
499
663
def _extract(self, versions):
500
664
"""Yield annotation of lines in included set.
523
688
WFE = WeaveFormatError
691
# 449 0 4474.6820 2356.5590 bzrlib.weave:556(_extract)
692
# +285282 0 1676.8040 1676.8040 +<isinstance>
693
# 1.6 seconds in 'isinstance'.
694
# changing the first isinstance:
695
# 449 0 2814.2660 1577.1760 bzrlib.weave:556(_extract)
696
# +140414 0 762.8050 762.8050 +<isinstance>
697
# note that the inline time actually dropped (less function calls)
698
# and total processing time was halved.
699
# we're still spending ~1/4 of the method in isinstance though.
700
# so lets hard code the acceptable string classes we expect:
701
# 449 0 1202.9420 786.2930 bzrlib.weave:556(_extract)
702
# +71352 0 377.5560 377.5560 +<method 'append' of 'list'
704
# yay, down to ~1/4 the initial extract time, and our inline time
705
# has shrunk again, with isinstance no longer dominating.
706
# tweaking the stack inclusion test to use a set gives:
707
# 449 0 1122.8030 713.0080 bzrlib.weave:556(_extract)
708
# +71352 0 354.9980 354.9980 +<method 'append' of 'list'
710
# - a 5% win, or possibly just noise. However with large istacks that
711
# 'in' test could dominate, so I'm leaving this change in place -
712
# when its fast enough to consider profiling big datasets we can review.
525
717
for l in self._weave:
526
if isinstance(l, tuple):
718
if l.__class__ == tuple:
530
assert v not in istack
725
iset.remove(istack.pop())
535
727
if v in included:
540
730
if v in included:
733
raise AssertionError()
544
assert isinstance(l, basestring)
545
735
if isactive is None:
546
736
isactive = (not dset) and istack and (istack[-1] in included)
548
738
result.append((istack[-1], lineno, l))
552
raise WFE("unclosed insertion blocks at end of weave",
741
raise WeaveFormatError("unclosed insertion blocks "
742
"at end of weave: %s" % istack)
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):
744
raise WeaveFormatError("unclosed deletion blocks at end of weave: %s"
748
def _maybe_lookup(self, name_or_index):
749
"""Convert possible symbolic name to index, or pass through indexes.
753
if isinstance(name_or_index, (int, long)):
756
return self._lookup(name_or_index)
758
def get_lines(self, version_id):
759
"""See VersionedFile.get_lines()."""
760
int_index = self._maybe_lookup(version_id)
761
result = [line for (origin, lineno, line) in self._extract([int_index])]
762
expected_sha1 = self._sha1s[int_index]
763
measured_sha1 = sha_strings(result)
764
if measured_sha1 != expected_sha1:
765
raise errors.WeaveInvalidChecksum(
766
'file %s, revision %s, expected: %s, measured %s'
767
% (self._weave_name, version_id,
768
expected_sha1, measured_sha1))
771
def get_sha1s(self, version_ids):
772
"""See VersionedFile.get_sha1s()."""
774
for v in version_ids:
775
result[v] = self._sha1s[self._lookup(v)]
778
def num_versions(self):
779
"""How many versions are in this weave?"""
598
780
l = len(self._parents)
599
assert l == len(self._sha1s)
604
return self.numversions()
783
__len__ = num_versions
607
785
def check(self, progress_bar=None):
608
# check no circular inclusions
609
for version in range(self.numversions()):
786
# TODO evaluate performance hit of using string sets in this routine.
787
# TODO: check no circular inclusions
788
# TODO: create a nested progress bar
789
for version in range(self.num_versions()):
610
790
inclusions = list(self._parents[version])
612
792
inclusions.sort()
614
794
raise WeaveFormatError("invalid included version %d for index %d"
615
795
% (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):
797
# try extracting all versions; parallel extraction is used
798
nv = self.num_versions()
803
# For creating the ancestry, IntSet is much faster (3.7s vs 0.17s)
804
# The problem is that set membership is much more expensive
805
name = self._idx_to_name(i)
808
new_inc = set([name])
809
for p in self._parents[i]:
810
new_inc.update(inclusions[self._idx_to_name(p)])
812
if set(new_inc) != set(self.get_ancestry(name)):
813
raise AssertionError(
815
% (set(new_inc), set(self.get_ancestry(name))))
816
inclusions[name] = new_inc
818
nlines = len(self._weave)
820
update_text = 'checking weave'
822
short_name = os.path.basename(self._weave_name)
823
update_text = 'checking %s' % (short_name,)
824
update_text = update_text[:25]
826
for lineno, insert, deleteset, line in self._walk_internal():
622
progress_bar.update('checking text', version, nv)
624
for l in self.get_iter(version):
627
expected = self._sha1s[version]
828
progress_bar.update(update_text, lineno, nlines)
830
for name, name_inclusions in inclusions.items():
831
# The active inclusion must be an ancestor,
832
# and no ancestors must have deleted this line,
833
# because we don't support resurrection.
834
if (insert in name_inclusions) and not (deleteset & name_inclusions):
835
sha1s[name].update(line)
838
version = self._idx_to_name(i)
839
hd = sha1s[version].hexdigest()
840
expected = self._sha1s[i]
628
841
if hd != expected:
629
raise WeaveError("mismatched sha1 for version %d; "
630
"got %s, expected %s"
631
% (version, hd, expected))
842
raise errors.WeaveInvalidChecksum(
843
"mismatched sha1 for version %s: "
844
"got %s, expected %s"
845
% (version, hd, expected))
633
847
# TODO: check insertions are properly nested, that there are
634
848
# no lines outside of insertion blocks, that deletions are
635
849
# 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()
677
def plan_merge(self, ver_a, ver_b):
678
"""Return pseudo-annotation indicating how the two versions merge.
680
This is computed between versions a and b and their common
683
Weave lines present in none of them are skipped entirely.
685
inc_a = self.inclusions([ver_a])
686
inc_b = self.inclusions([ver_b])
687
inc_c = inc_a & inc_b
689
for lineno, insert, deleteset, line in self._walk():
690
if deleteset & inc_c:
691
# killed in parent; can't be in either a or b
692
# not relevant to our work
693
yield 'killed-base', line
694
elif insert in inc_c:
695
# was inserted in base
696
killed_a = bool(deleteset & inc_a)
697
killed_b = bool(deleteset & inc_b)
698
if killed_a and killed_b:
699
yield 'killed-both', line
701
yield 'killed-a', line
703
yield 'killed-b', line
705
yield 'unchanged', line
706
elif insert in inc_a:
707
if deleteset & inc_a:
708
yield 'ghost-a', line
712
elif insert in inc_b:
713
if deleteset & inc_b:
714
yield 'ghost-b', line
718
# not in either revision
719
yield 'irrelevant', line
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
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':
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().
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)
802
851
def _imported_parents(self, other, other_idx):
803
852
"""Return list of parents in self corresponding to indexes in other."""
805
854
for parent_idx in other._parents[other_idx]:
806
855
parent_name = other._names[parent_idx]
807
if parent_name not in self._names:
856
if parent_name not in self._name_map:
808
857
# should not be possible
809
raise WeaveError("missing parent {%s} of {%s} in %r"
858
raise WeaveError("missing parent {%s} of {%s} in %r"
810
859
% (parent_name, other._name_map[other_idx], self))
811
860
new_parents.append(self._name_map[parent_name])
812
861
return new_parents
846
def reweave(self, other):
847
"""Reweave self with other."""
848
new_weave = reweave(self, other)
891
def _reweave(self, other, pb, msg):
892
"""Reweave self with other - internal helper for join().
894
:param other: The other weave to merge
895
:param pb: An optional progress bar, indicating how far done we are
896
:param msg: An optional message for the progress
898
new_weave = _reweave(self, other, pb=pb, msg=msg)
899
self._copy_weave_content(new_weave)
901
def _copy_weave_content(self, otherweave):
902
"""adsorb the content from otherweave."""
849
903
for attr in self.__slots__:
850
setattr(self, attr, getattr(new_weave, attr))
904
if attr != '_weave_name':
905
setattr(self, attr, copy(getattr(otherweave, attr)))
908
class WeaveFile(Weave):
909
"""A WeaveFile represents a Weave on disk and writes on change."""
911
WEAVE_SUFFIX = '.weave'
913
def __init__(self, name, transport, filemode=None, create=False, access_mode='w', get_scope=None):
914
"""Create a WeaveFile.
916
:param create: If not True, only open an existing knit.
918
super(WeaveFile, self).__init__(name, access_mode, get_scope=get_scope,
919
allow_reserved=False)
920
self._transport = transport
921
self._filemode = filemode
923
f = self._transport.get(name + WeaveFile.WEAVE_SUFFIX)
924
_read_weave_v5(StringIO(f.read()), self)
925
except errors.NoSuchFile:
931
def _add_lines(self, version_id, parents, lines, parent_texts,
932
left_matching_blocks, nostore_sha, random_id, check_content):
933
"""Add a version and save the weave."""
934
self.check_not_reserved_id(version_id)
935
result = super(WeaveFile, self)._add_lines(version_id, parents, lines,
936
parent_texts, left_matching_blocks, nostore_sha, random_id,
941
def copy_to(self, name, transport):
942
"""See VersionedFile.copy_to()."""
943
# as we are all in memory always, just serialise to the new place.
945
write_weave_v5(self, sio)
947
transport.put_file(name + WeaveFile.WEAVE_SUFFIX, sio, self._filemode)
950
"""Save the weave."""
951
self._check_write_ok()
953
write_weave_v5(self, sio)
955
bytes = sio.getvalue()
956
path = self._weave_name + WeaveFile.WEAVE_SUFFIX
958
self._transport.put_bytes(path, bytes, self._filemode)
959
except errors.NoSuchFile:
960
self._transport.mkdir(dirname(path))
961
self._transport.put_bytes(path, bytes, self._filemode)
965
"""See VersionedFile.get_suffixes()."""
966
return [WeaveFile.WEAVE_SUFFIX]
968
def insert_record_stream(self, stream):
969
super(WeaveFile, self).insert_record_stream(stream)
973
def _reweave(wa, wb, pb=None, msg=None):
854
974
"""Combine two weaves and return the result.
856
This works even if a revision R has different parents in
976
This works even if a revision R has different parents in
857
977
wa and wb. In the resulting weave all the parents are given.
859
This is done by just building up a new weave, maintaining ordering
979
This is done by just building up a new weave, maintaining ordering
860
980
of the versions in the two inputs. More efficient approaches
861
might be possible but it should only be necessary to do
862
this operation rarely, when a new previously ghost version is
981
might be possible but it should only be necessary to do
982
this operation rarely, when a new previously ghost version is
985
:param pb: An optional progress bar, indicating how far done we are
986
:param msg: An optional message for the progress
867
queue_a = range(wa.numversions())
868
queue_b = range(wb.numversions())
990
queue_a = range(wa.num_versions())
991
queue_b = range(wb.num_versions())
869
992
# first determine combined parents of all versions
870
993
# map from version name -> all parent names
871
994
combined_parents = _reweave_parent_graphs(wa, wb)
872
995
mutter("combined parents: %r", combined_parents)
873
order = _make_reweave_order(wa._names, wb._names, combined_parents)
996
order = tsort.topo_sort(combined_parents.iteritems())
874
997
mutter("order to reweave: %r", order)
1002
for idx, name in enumerate(order):
1004
pb.update(msg, idx, len(order))
876
1005
if name in wa._name_map:
877
1006
lines = wa.get_lines(name)
878
1007
if name in wb._name_map:
879
assert lines == wb.get_lines(name)
1008
lines_b = wb.get_lines(name)
1009
if lines != lines_b:
1010
mutter('Weaves differ on content. rev_id {%s}', name)
1011
mutter('weaves: %s, %s', wa._weave_name, wb._weave_name)
1013
lines = list(difflib.unified_diff(lines, lines_b,
1014
wa._weave_name, wb._weave_name))
1015
mutter('lines:\n%s', ''.join(lines))
1016
raise errors.WeaveTextDiffers(name, wa, wb)
881
1018
lines = wb.get_lines(name)
882
wr.add(name, combined_parents[name], lines)
1019
wr._add(name, lines, [wr._lookup(i) for i in combined_parents[name]])
886
1023
def _reweave_parent_graphs(wa, wb):
887
1024
"""Return combined parent ancestry for two weaves.
889
1026
Returned as a list of (version_name, set(parent_names))"""
891
1028
for weave in [wa, wb]:
892
1029
for idx, name in enumerate(weave._names):
893
1030
p = combined.setdefault(name, set())
894
p.update(map(weave.idx_to_name, weave._parents[idx]))
1031
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
