3
# Copyright (C) 2005 Canonical Ltd
5
# This program is free software; you can redistribute it and/or modify
6
# it under the terms of the GNU General Public License as published by
7
# the Free Software Foundation; either version 2 of the License, or
8
# (at your option) any later version.
10
# This program is distributed in the hope that it will be useful,
11
# but WITHOUT ANY WARRANTY; without even the implied warranty of
12
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13
# GNU General Public License for more details.
15
# You should have received a copy of the GNU General Public License
16
# along with this program; if not, write to the Free Software
17
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
19
# Author: Martin Pool <mbp@canonical.com>
22
"""Weave - storage of related text file versions"""
25
# XXX: If we do weaves this way, will a merge still behave the same
26
# way if it's done in a different order? That's a pretty desirable
29
# TODO: Nothing here so far assumes the lines are really \n newlines,
30
# rather than being split up in some other way. We could accomodate
31
# binaries, perhaps by naively splitting on \n or perhaps using
32
# something like a rolling checksum.
34
# TODO: End marker for each version so we can stop reading?
36
# TODO: Check that no insertion occurs inside a deletion that was
37
# active in the version of the insertion.
39
# TODO: In addition to the SHA-1 check, perhaps have some code that
40
# checks structural constraints of the weave: ie that insertions are
41
# properly nested, that there is no text outside of an insertion, that
42
# insertions or deletions are not repeated, etc.
44
# TODO: Parallel-extract that passes back each line along with a
45
# description of which revisions include it. Nice for checking all
46
# shas or calculating stats in parallel.
48
# TODO: Using a single _extract routine and then processing the output
49
# is probably inefficient. It's simple enough that we can afford to
50
# have slight specializations for different ways its used: annotate,
51
# basis for add, get, etc.
53
# TODO: Probably the API should work only in names to hide the integer
54
# indexes from the user.
56
# TODO: Is there any potential performance win by having an add()
57
# variant that is passed a pre-cooked version of the single basis
60
# TODO: Reweave can possibly be made faster by remembering diffs
61
# where the basis and destination are unchanged.
63
# 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
66
# be done fairly efficiently because the sequence numbers constrain
67
# the possible relationships.
73
from difflib import SequenceMatcher
75
from bzrlib.trace import mutter
76
from bzrlib.errors import (WeaveError, WeaveFormatError, WeaveParentMismatch,
77
RevisionAlreadyPresent,
79
WeaveRevisionAlreadyPresent,
80
WeaveRevisionNotPresent,
82
import bzrlib.errors as errors
83
from bzrlib.symbol_versioning import *
84
from bzrlib.tsort import topo_sort
85
from bzrlib.versionedfile import VersionedFile
88
class Weave(VersionedFile):
89
"""weave - versioned text file storage.
91
A Weave manages versions of line-based text files, keeping track
92
of the originating version for each line.
94
To clients the "lines" of the file are represented as a list of strings.
95
These strings will typically have terminal newline characters, but
96
this is not required. In particular files commonly do not have a newline
97
at the end of the file.
99
Texts can be identified in either of two ways:
101
* a nonnegative index number.
103
* a version-id string.
105
Typically the index number will be valid only inside this weave and
106
the version-id is used to reference it in the larger world.
108
The weave is represented as a list mixing edit instructions and
109
literal text. Each entry in _weave can be either a string (or
110
unicode), or a tuple. If a string, it means that the given line
111
should be output in the currently active revisions.
113
If a tuple, it gives a processing instruction saying in which
114
revisions the enclosed lines are active. The tuple has the form
115
(instruction, version).
117
The instruction can be '{' or '}' for an insertion block, and '['
118
and ']' for a deletion block respectively. The version is the
119
integer version index. There is no replace operator, only deletes
120
and inserts. For '}', the end of an insertion, there is no
121
version parameter because it always closes the most recently
126
* A later version can delete lines that were introduced by any
127
number of ancestor versions; this implies that deletion
128
instructions can span insertion blocks without regard to the
129
insertion block's nesting.
131
* Similarly, deletions need not be properly nested with regard to
132
each other, because they might have been generated by
133
independent revisions.
135
* Insertions are always made by inserting a new bracketed block
136
into a single point in the previous weave. This implies they
137
can nest but not overlap, and the nesting must always have later
138
insertions on the inside.
140
* It doesn't seem very useful to have an active insertion
141
inside an inactive insertion, but it might happen.
143
* Therefore, all instructions are always"considered"; that
144
is passed onto and off the stack. An outer inactive block
145
doesn't disable an inner block.
147
* Lines are enabled if the most recent enclosing insertion is
148
active and none of the enclosing deletions are active.
150
* There is no point having a deletion directly inside its own
151
insertion; you might as well just not write it. And there
152
should be no way to get an earlier version deleting a later
156
Text of the weave; list of control instruction tuples and strings.
159
List of parents, indexed by version number.
160
It is only necessary to store the minimal set of parents for
161
each version; the parent's parents are implied.
164
List of hex SHA-1 of each version.
167
List of symbolic names for each version. Each should be unique.
170
For each name, the version number.
173
Descriptive name of this weave; typically the filename if known.
177
__slots__ = ['_weave', '_parents', '_sha1s', '_names', '_name_map',
180
def __init__(self, weave_name=None):
186
self._weave_name = weave_name
189
return "Weave(%r)" % self._weave_name
193
"""Return a deep copy of self.
195
The copy can be modified without affecting the original weave."""
197
other._weave = self._weave[:]
198
other._parents = self._parents[:]
199
other._sha1s = self._sha1s[:]
200
other._names = self._names[:]
201
other._name_map = self._name_map.copy()
202
other._weave_name = self._weave_name
205
def __eq__(self, other):
206
if not isinstance(other, Weave):
208
return self._parents == other._parents \
209
and self._weave == other._weave \
210
and self._sha1s == other._sha1s
213
def __ne__(self, other):
214
return not self.__eq__(other)
216
@deprecated_method(zero_eight)
217
def idx_to_name(self, index):
218
"""Old public interface, the public interface is all names now."""
221
def _idx_to_name(self, version):
222
return self._names[version]
224
@deprecated_method(zero_eight)
225
def lookup(self, name):
226
"""Backwards compatability thunk:
228
Return name, as name is valid in the api now, and spew deprecation
233
def _lookup(self, name):
234
"""Convert symbolic version name to index."""
236
return self._name_map[name]
238
raise RevisionNotPresent(name, self._weave_name)
240
@deprecated_method(zero_eight)
241
def iter_names(self):
242
"""Deprecated convenience function, please see VersionedFile.names()."""
243
return iter(self.names())
245
@deprecated_method(zero_eight)
247
"""See Weave.versions for the current api."""
248
return self.versions()
251
"""See VersionedFile.versions."""
252
return self._names[:]
254
def has_version(self, version_id):
255
"""See VersionedFile.has_version."""
256
return self._name_map.has_key(version_id)
258
__contains__ = has_version
260
@deprecated_method(zero_eight)
261
def parent_names(self, version):
262
"""Return version names for parents of a version.
264
See get_parents for the current api.
266
return self.get_parents(version)
268
def get_parents(self, version_id):
269
"""See VersionedFile.get_parent."""
270
return map(self._idx_to_name, self._parents[self._lookup(version_id)])
272
def _check_repeated_add(self, name, parents, text, sha1):
273
"""Check that a duplicated add is OK.
275
If it is, return the (old) index; otherwise raise an exception.
277
idx = self._lookup(name)
278
if sorted(self._parents[idx]) != sorted(parents) \
279
or sha1 != self._sha1s[idx]:
280
raise RevisionAlreadyPresent(name, self._weave_name)
283
@deprecated_method(zero_eight)
284
def add_identical(self, old_rev_id, new_rev_id, parents):
285
"""Please use Weave.clone_text now."""
286
return self.clone_text(new_rev_id, old_rev_id, parents, None)
288
def add_lines(self, version_id, parents, lines):
289
"""See VersionedFile.add_lines."""
290
return self._add(version_id, lines, map(self._lookup, parents))
292
@deprecated_method(zero_eight)
293
def add(self, name, parents, text, sha1=None):
294
"""See VersionedFile.add_lines for the non deprecated api."""
295
return self._add(name, text, map(self._maybe_lookup, parents), sha1)
297
def _add(self, version_id, lines, parents, sha1=None):
298
"""Add a single text on top of the weave.
300
Returns the index number of the newly added version.
303
Symbolic name for this version.
304
(Typically the revision-id of the revision that added it.)
307
List or set of direct parent version numbers.
310
Sequence of lines to be added in the new version.
312
from bzrlib.osutils import sha_strings
314
assert isinstance(version_id, basestring)
316
sha1 = sha_strings(lines)
317
if version_id in self._name_map:
318
return self._check_repeated_add(version_id, parents, lines, sha1)
320
self._check_versions(parents)
321
## self._check_lines(lines)
322
new_version = len(self._parents)
324
# if we abort after here the (in-memory) weave will be corrupt because only
325
# some fields are updated
326
self._parents.append(parents[:])
327
self._sha1s.append(sha1)
328
self._names.append(version_id)
329
self._name_map[version_id] = new_version
333
# special case; adding with no parents revision; can do
334
# this more quickly by just appending unconditionally.
335
# even more specially, if we're adding an empty text we
336
# need do nothing at all.
338
self._weave.append(('{', new_version))
339
self._weave.extend(lines)
340
self._weave.append(('}', None))
343
if len(parents) == 1:
344
pv = list(parents)[0]
345
if sha1 == self._sha1s[pv]:
346
# special case: same as the single parent
350
ancestors = self._inclusions(parents)
354
# basis a list of (origin, lineno, line)
357
for origin, lineno, line in self._extract(ancestors):
358
basis_lineno.append(lineno)
359
basis_lines.append(line)
361
# another small special case: a merge, producing the same text
363
if lines == basis_lines:
366
# add a sentinal, because we can also match against the final line
367
basis_lineno.append(len(self._weave))
369
# XXX: which line of the weave should we really consider
370
# matches the end of the file? the current code says it's the
371
# last line of the weave?
373
#print 'basis_lines:', basis_lines
374
#print 'new_lines: ', lines
376
s = SequenceMatcher(None, basis_lines, lines)
378
# offset gives the number of lines that have been inserted
379
# into the weave up to the current point; if the original edit instruction
380
# says to change line A then we actually change (A+offset)
383
for tag, i1, i2, j1, j2 in s.get_opcodes():
384
# i1,i2 are given in offsets within basis_lines; we need to map them
385
# back to offsets within the entire weave
386
#print 'raw match', tag, i1, i2, j1, j2
390
i1 = basis_lineno[i1]
391
i2 = basis_lineno[i2]
393
assert 0 <= j1 <= j2 <= len(lines)
395
#print tag, i1, i2, j1, j2
397
# the deletion and insertion are handled separately.
398
# first delete the region.
400
self._weave.insert(i1+offset, ('[', new_version))
401
self._weave.insert(i2+offset+1, (']', new_version))
405
# there may have been a deletion spanning up to
406
# i2; we want to insert after this region to make sure
407
# we don't destroy ourselves
409
self._weave[i:i] = ([('{', new_version)]
412
offset += 2 + (j2 - j1)
415
def clone_text(self, new_version_id, old_version_id, parents,
417
"""See VersionedFile.clone_text."""
418
old_lines = self.get_text(old_version_id)
419
self.add_lines(new_version_id, parents, old_lines)
421
def _inclusions(self, versions):
422
"""Return set of all ancestors of given version(s)."""
424
for v in xrange(max(versions), 0, -1):
426
# include all its parents
427
i.update(self._parents[v])
429
## except IndexError:
430
## raise ValueError("version %d not present in weave" % v)
432
@deprecated_method(zero_eight)
433
def inclusions(self, version_ids):
434
"""Deprecated - see VersionedFile.get_ancestry for the replacement."""
437
if isinstance(version_ids[0], int):
438
return [self._idx_to_name(v) for v in self._inclusions(version_ids)]
440
return self.get_ancestry(version_ids)
442
def get_ancestry(self, version_ids):
443
"""See VersionedFile.get_ancestry."""
444
if isinstance(version_ids, basestring):
445
version_ids = [version_ids]
446
i = self._inclusions([self._lookup(v) for v in version_ids])
447
return [self._idx_to_name(v) for v in i]
449
def _check_lines(self, text):
450
if not isinstance(text, list):
451
raise ValueError("text should be a list, not %s" % type(text))
454
if not isinstance(l, basestring):
455
raise ValueError("text line should be a string or unicode, not %s"
460
def _check_versions(self, indexes):
461
"""Check everything in the sequence of indexes is valid"""
466
raise IndexError("invalid version number %r" % i)
468
def annotate(self, version_id):
469
if isinstance(version_id, int):
470
warn('Weave.annotate(int) is deprecated. Please use version names'
471
' in all circumstances as of 0.8',
476
for origin, lineno, text in self._extract([version_id]):
477
result.append((origin, text))
480
return super(Weave, self).annotate(version_id)
482
def annotate_iter(self, version_id):
483
"""Yield list of (version-id, line) pairs for the specified version.
485
The index indicates when the line originated in the weave."""
486
incls = [self._lookup(version_id)]
487
for origin, lineno, text in self._extract(incls):
488
yield self._idx_to_name(origin), text
490
@deprecated_method(zero_eight)
492
"""_walk has become walk, a supported api."""
495
def walk(self, version_ids=None):
496
"""See VersionedFile.walk."""
501
lineno = 0 # line of weave, 0-based
503
for l in self._weave:
504
if isinstance(l, tuple):
508
istack.append(self._idx_to_name(v))
512
assert self._idx_to_name(v) not in dset
513
dset.add(self._idx_to_name(v))
515
dset.remove(self._idx_to_name(v))
517
raise WeaveFormatError('unexpected instruction %r' % v)
519
assert isinstance(l, basestring)
521
yield lineno, istack[-1], dset.copy(), l
525
raise WeaveFormatError("unclosed insertion blocks "
526
"at end of weave: %s" % istack)
528
raise WeaveFormatError("unclosed deletion blocks at end of weave: %s"
531
def _extract(self, versions):
532
"""Yield annotation of lines in included set.
534
Yields a sequence of tuples (origin, lineno, text), where
535
origin is the origin version, lineno the index in the weave,
536
and text the text of the line.
538
The set typically but not necessarily corresponds to a version.
541
if not isinstance(i, int):
544
included = self._inclusions(versions)
549
lineno = 0 # line of weave, 0-based
555
WFE = WeaveFormatError
557
for l in self._weave:
558
if isinstance(l, tuple):
562
assert v not in istack
576
assert isinstance(l, basestring)
578
isactive = (not dset) and istack and (istack[-1] in included)
580
result.append((istack[-1], lineno, l))
583
raise WeaveFormatError("unclosed insertion blocks "
584
"at end of weave: %s" % istack)
586
raise WeaveFormatError("unclosed deletion blocks at end of weave: %s"
590
@deprecated_method(zero_eight)
591
def get_iter(self, name_or_index):
592
"""Deprecated, please do not use. Lookups are not not needed."""
593
return self._get_iter(self._maybe_lookup(name_or_index))
595
@deprecated_method(zero_eight)
596
def maybe_lookup(self, name_or_index):
597
"""Deprecated, please do not use. Lookups are not not needed."""
598
return self._maybe_lookup(name_or_index)
600
def _maybe_lookup(self, name_or_index):
601
"""Convert possible symbolic name to index, or pass through indexes.
605
if isinstance(name_or_index, (int, long)):
608
return self._lookup(name_or_index)
610
def _get_iter(self, version_id):
611
"""Yield lines for the specified version."""
612
incls = [self._maybe_lookup(version_id)]
617
# We don't have sha1 sums for multiple entries
619
for origin, lineno, line in self._extract(incls):
624
expected_sha1 = self._sha1s[index]
625
measured_sha1 = cur_sha.hexdigest()
626
if measured_sha1 != expected_sha1:
627
raise errors.WeaveInvalidChecksum(
628
'file %s, revision %s, expected: %s, measured %s'
629
% (self._weave_name, self._names[index],
630
expected_sha1, measured_sha1))
632
@deprecated_method(zero_eight)
633
def get(self, version_id):
634
"""Please use either Weave.get_text or Weave.get_lines as desired."""
635
return self.get_lines(version_id)
637
def get_lines(self, version_id):
638
"""See VersionedFile.get_lines()."""
639
return list(self._get_iter(version_id))
641
def get_sha1(self, name):
642
"""Get the stored sha1 sum for the given revision.
644
:param name: The name of the version to lookup
646
return self._sha1s[self._lookup(name)]
648
def numversions(self):
649
l = len(self._parents)
650
assert l == len(self._sha1s)
653
__len__ = numversions
655
def check(self, progress_bar=None):
656
# TODO evaluate performance hit of using string sets in this routine.
657
# check no circular inclusions
658
for version in range(self.numversions()):
659
inclusions = list(self._parents[version])
662
if inclusions[-1] >= version:
663
raise WeaveFormatError("invalid included version %d for index %d"
664
% (inclusions[-1], version))
666
# try extracting all versions; parallel extraction is used
667
nv = self.numversions()
672
# For creating the ancestry, IntSet is much faster (3.7s vs 0.17s)
673
# The problem is that set membership is much more expensive
674
name = self._idx_to_name(i)
675
sha1s[name] = sha.new()
677
new_inc = set([name])
678
for p in self._parents[i]:
679
new_inc.update(inclusions[self._idx_to_name(p)])
681
assert set(new_inc) == set(self.get_ancestry(name)), \
682
'failed %s != %s' % (set(new_inc), set(self.get_ancestry(name)))
683
inclusions[name] = new_inc
685
nlines = len(self._weave)
687
update_text = 'checking weave'
689
short_name = os.path.basename(self._weave_name)
690
update_text = 'checking %s' % (short_name,)
691
update_text = update_text[:25]
693
for lineno, insert, deleteset, line in self.walk():
695
progress_bar.update(update_text, lineno, nlines)
697
for name, name_inclusions in inclusions.items():
698
# The active inclusion must be an ancestor,
699
# and no ancestors must have deleted this line,
700
# because we don't support resurrection.
701
if (insert in name_inclusions) and not (deleteset & name_inclusions):
702
sha1s[name].update(line)
705
version = self._idx_to_name(i)
706
hd = sha1s[version].hexdigest()
707
expected = self._sha1s[i]
709
raise errors.WeaveInvalidChecksum(
710
"mismatched sha1 for version %s: "
711
"got %s, expected %s"
712
% (version, hd, expected))
714
# TODO: check insertions are properly nested, that there are
715
# no lines outside of insertion blocks, that deletions are
716
# properly paired, etc.
719
def join(self, other, pb=None, msg=None, version_ids=None):
721
"""Integrate versions from other into this weave.
723
The resulting weave contains all the history of both weaves;
724
any version you could retrieve from either self or other can be
725
retrieved from self after this call.
727
It is illegal for the two weaves to contain different values
728
or different parents for any version. See also reweave().
730
:param other: The other weave to pull into this one
731
:param pb: An optional progress bar
732
:param msg: An optional message to display for progress
734
if not other.versions():
735
return # nothing to update, easy
738
for version_id in version_ids:
739
if not self.has_version(version_id):
740
raise RevisionNotPresent(version_id, self._weave_name)
741
assert version_ids == None
743
# two loops so that we do not change ourselves before verifying it
745
# work through in index order to make sure we get all dependencies
748
for other_idx, name in enumerate(other._names):
749
self._check_version_consistent(other, other_idx, name)
750
sha1 = other._sha1s[other_idx]
754
if name in self._name_map:
755
idx = self._lookup(name)
756
n1 = set(map(other._idx_to_name, other._parents[other_idx]))
757
n2 = set(map(self._idx_to_name, self._parents[idx]))
758
if sha1 == self._sha1s[idx] and n1 == n2:
761
names_to_join.append((other_idx, name))
768
for other_idx, name in names_to_join:
769
# TODO: If all the parents of the other version are already
770
# present then we can avoid some work by just taking the delta
771
# and adjusting the offsets.
772
new_parents = self._imported_parents(other, other_idx)
773
sha1 = other._sha1s[other_idx]
778
pb.update(msg, merged, len(names_to_join))
780
lines = other.get_lines(other_idx)
781
self._add(name, lines, new_parents, sha1)
783
mutter("merged = %d, processed = %d, file_id=%s; deltat=%d"%(
784
merged, processed, self._weave_name, time.time()-time0))
786
def _imported_parents(self, other, other_idx):
787
"""Return list of parents in self corresponding to indexes in other."""
789
for parent_idx in other._parents[other_idx]:
790
parent_name = other._names[parent_idx]
791
if parent_name not in self._names:
792
# should not be possible
793
raise WeaveError("missing parent {%s} of {%s} in %r"
794
% (parent_name, other._name_map[other_idx], self))
795
new_parents.append(self._name_map[parent_name])
798
def _check_version_consistent(self, other, other_idx, name):
799
"""Check if a version in consistent in this and other.
801
To be consistent it must have:
804
* the same direct parents (by name, not index, and disregarding
807
If present & correct return True;
808
if not present in self return False;
809
if inconsistent raise error."""
810
this_idx = self._name_map.get(name, -1)
812
if self._sha1s[this_idx] != other._sha1s[other_idx]:
813
raise WeaveError("inconsistent texts for version {%s} "
814
"when joining weaves"
816
self_parents = self._parents[this_idx]
817
other_parents = other._parents[other_idx]
818
n1 = set([self._names[i] for i in self_parents])
819
n2 = set([other._names[i] for i in other_parents])
821
raise WeaveParentMismatch("inconsistent parents "
822
"for version {%s}: %s vs %s" % (name, n1, n2))
828
def reweave(self, other, pb=None, msg=None):
829
"""Reweave self with other.
831
:param other: The other weave to merge
832
:param pb: An optional progress bar, indicating how far done we are
833
:param msg: An optional message for the progress
835
new_weave = reweave(self, other, pb=pb, msg=msg)
836
for attr in self.__slots__:
837
setattr(self, attr, getattr(new_weave, attr))
840
def reweave(wa, wb, pb=None, msg=None):
841
"""Combine two weaves and return the result.
843
This works even if a revision R has different parents in
844
wa and wb. In the resulting weave all the parents are given.
846
This is done by just building up a new weave, maintaining ordering
847
of the versions in the two inputs. More efficient approaches
848
might be possible but it should only be necessary to do
849
this operation rarely, when a new previously ghost version is
852
:param pb: An optional progress bar, indicating how far done we are
853
:param msg: An optional message for the progress
857
queue_a = range(wa.numversions())
858
queue_b = range(wb.numversions())
859
# first determine combined parents of all versions
860
# map from version name -> all parent names
861
combined_parents = _reweave_parent_graphs(wa, wb)
862
mutter("combined parents: %r", combined_parents)
863
order = topo_sort(combined_parents.iteritems())
864
mutter("order to reweave: %r", order)
869
for idx, name in enumerate(order):
871
pb.update(msg, idx, len(order))
872
if name in wa._name_map:
873
lines = wa.get_lines(name)
874
if name in wb._name_map:
875
lines_b = wb.get_lines(name)
877
mutter('Weaves differ on content. rev_id {%s}', name)
878
mutter('weaves: %s, %s', wa._weave_name, wb._weave_name)
880
lines = list(difflib.unified_diff(lines, lines_b,
881
wa._weave_name, wb._weave_name))
882
mutter('lines:\n%s', ''.join(lines))
883
raise errors.WeaveTextDiffers(name, wa, wb)
885
lines = wb.get_lines(name)
886
wr._add(name, lines, [wr._lookup(i) for i in combined_parents[name]])
889
def _reweave_parent_graphs(wa, wb):
890
"""Return combined parent ancestry for two weaves.
892
Returned as a list of (version_name, set(parent_names))"""
894
for weave in [wa, wb]:
895
for idx, name in enumerate(weave._names):
896
p = combined.setdefault(name, set())
897
p.update(map(weave._idx_to_name, weave._parents[idx]))
902
"""Show the weave's table-of-contents"""
903
print '%6s %50s %10s %10s' % ('ver', 'name', 'sha1', 'parents')
904
for i in (6, 50, 10, 10):
907
for i in range(w.numversions()):
910
parent_str = ' '.join(map(str, w._parents[i]))
911
print '%6d %-50.50s %10.10s %s' % (i, name, sha1, parent_str)
915
def weave_stats(weave_file, pb):
916
from bzrlib.weavefile import read_weave
918
wf = file(weave_file, 'rb')
919
w = read_weave(wf, WeaveVersionedFile)
920
# FIXME: doesn't work on pipes
921
weave_size = wf.tell()
925
for i in range(vers):
926
pb.update('checking sizes', i, vers)
927
for origin, lineno, line in w._extract([i]):
932
print 'versions %9d' % vers
933
print 'weave file %9d bytes' % weave_size
934
print 'total contents %9d bytes' % total
935
print 'compression ratio %9.2fx' % (float(total) / float(weave_size))
938
print 'average size %9d bytes' % avg
939
print 'relative size %9.2fx' % (float(weave_size) / float(avg))
943
print """bzr weave tool
945
Experimental tool for weave algorithm.
949
Create an empty weave file
950
weave get WEAVEFILE VERSION
951
Write out specified version.
952
weave check WEAVEFILE
953
Check consistency of all versions.
955
Display table of contents.
956
weave add WEAVEFILE NAME [BASE...] < NEWTEXT
957
Add NEWTEXT, with specified parent versions.
958
weave annotate WEAVEFILE VERSION
959
Display origin of each line.
960
weave merge WEAVEFILE VERSION1 VERSION2 > OUT
961
Auto-merge two versions and display conflicts.
962
weave diff WEAVEFILE VERSION1 VERSION2
963
Show differences between two versions.
967
% weave init foo.weave
969
% weave add foo.weave ver0 < foo.txt
972
(create updated version)
974
% weave get foo.weave 0 | diff -u - foo.txt
975
% weave add foo.weave ver1 0 < foo.txt
978
% weave get foo.weave 0 > foo.txt (create forked version)
980
% weave add foo.weave ver2 0 < foo.txt
983
% weave merge foo.weave 1 2 > foo.txt (merge them)
984
% vi foo.txt (resolve conflicts)
985
% weave add foo.weave merged 1 2 < foo.txt (commit merged version)
997
# in case we're run directly from the subdirectory
998
sys.path.append('..')
1000
from bzrlib.weavefile import write_weave, read_weave
1001
from bzrlib.progress import ProgressBar
1016
return read_weave(file(argv[2], 'rb'))
1022
# at the moment, based on everything in the file
1024
parents = map(int, argv[4:])
1025
lines = sys.stdin.readlines()
1026
ver = w.add(name, parents, lines)
1027
write_weave(w, file(argv[2], 'wb'))
1028
print 'added version %r %d' % (name, ver)
1031
if os.path.exists(fn):
1032
raise IOError("file exists")
1034
write_weave(w, file(fn, 'wb'))
1035
elif cmd == 'get': # get one version
1037
sys.stdout.writelines(w.get_iter(int(argv[3])))
1040
from difflib import unified_diff
1043
v1, v2 = map(int, argv[3:5])
1046
diff_gen = unified_diff(lines1, lines2,
1047
'%s version %d' % (fn, v1),
1048
'%s version %d' % (fn, v2))
1049
sys.stdout.writelines(diff_gen)
1051
elif cmd == 'annotate':
1053
# newline is added to all lines regardless; too hard to get
1054
# reasonable formatting otherwise
1056
for origin, text in w.annotate(int(argv[3])):
1057
text = text.rstrip('\r\n')
1059
print ' | %s' % (text)
1061
print '%5d | %s' % (origin, text)
1067
elif cmd == 'stats':
1068
weave_stats(argv[2], ProgressBar())
1070
elif cmd == 'check':
1075
print '%d versions ok' % w.numversions()
1077
elif cmd == 'inclusions':
1079
print ' '.join(map(str, w.inclusions([int(argv[3])])))
1081
elif cmd == 'parents':
1083
print ' '.join(map(str, w._parents[int(argv[3])]))
1085
elif cmd == 'plan-merge':
1087
for state, line in w.plan_merge(int(argv[3]), int(argv[4])):
1089
print '%14s | %s' % (state, line),
1091
elif cmd == 'merge':
1093
p = w.plan_merge(int(argv[3]), int(argv[4]))
1094
sys.stdout.writelines(w.weave_merge(p))
1097
raise ValueError('unknown command %r' % cmd)
1101
def profile_main(argv):
1102
import tempfile, hotshot, hotshot.stats
1104
prof_f = tempfile.NamedTemporaryFile()
1106
prof = hotshot.Profile(prof_f.name)
1108
ret = prof.runcall(main, argv)
1111
stats = hotshot.stats.load(prof_f.name)
1113
stats.sort_stats('cumulative')
1114
## XXX: Might like to write to stderr or the trace file instead but
1115
## print_stats seems hardcoded to stdout
1116
stats.print_stats(20)
1121
def lsprofile_main(argv):
1122
from bzrlib.lsprof import profile
1123
ret,stats = profile(main, argv)
1129
if __name__ == '__main__':
1131
if '--profile' in sys.argv:
1133
args.remove('--profile')
1134
sys.exit(profile_main(args))
1135
elif '--lsprof' in sys.argv:
1137
args.remove('--lsprof')
1138
sys.exit(lsprofile_main(args))
1140
sys.exit(main(sys.argv))
added
removed
bzrlib/weave.py
