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"""
24
# before intset (r923) 2000 versions in 41.5s
25
# with intset (r926) 2000 versions in 93s !!!
26
# better to just use plain sets.
28
# making _extract build and return a list, rather than being a generator
31
# with python -O, r923 does 2000 versions in 36.87s
33
# with optimizations to avoid mutating lists - 35.75! I guess copying
34
# all the elements every time costs more than the small manipulations.
35
# a surprisingly small change.
37
# r931, which avoids using a generator for extract, does 36.98s
39
# with memoized inclusions, takes 41.49s; not very good
41
# with slots, takes 37.35s; without takes 39.16, a bit surprising
43
# with the delta calculation mixed in with the add method, rather than
44
# separated, takes 36.78s
46
# with delta folded in and mutation of the list, 36.13s
48
# with all this and simplification of add code, 33s
54
# TODO: Perhaps have copy method for Weave instances?
56
# XXX: If we do weaves this way, will a merge still behave the same
57
# way if it's done in a different order? That's a pretty desirable
60
# TODO: Nothing here so far assumes the lines are really \n newlines,
61
# rather than being split up in some other way. We could accomodate
62
# binaries, perhaps by naively splitting on \n or perhaps using
63
# something like a rolling checksum.
65
# TODO: End marker for each version so we can stop reading?
67
# TODO: Check that no insertion occurs inside a deletion that was
68
# active in the version of the insertion.
70
# TODO: In addition to the SHA-1 check, perhaps have some code that
71
# checks structural constraints of the weave: ie that insertions are
72
# properly nested, that there is no text outside of an insertion, that
73
# insertions or deletions are not repeated, etc.
75
# TODO: Parallel-extract that passes back each line along with a
76
# description of which revisions include it. Nice for checking all
79
# TODO: Using a single _extract routine and then processing the output
80
# is probably inefficient. It's simple enough that we can afford to
81
# have slight specializations for different ways its used: annotate,
82
# basis for add, get, etc.
84
# TODO: Perhaps the API should work only in names to hide the integer
85
# indexes from the user?
91
from cStringIO import StringIO
93
from bzrlib.osutils import sha_strings
96
class WeaveError(Exception):
97
"""Exception in processing weave"""
100
class WeaveFormatError(WeaveError):
101
"""Weave invariant violated"""
105
"""weave - versioned text file storage.
107
A Weave manages versions of line-based text files, keeping track
108
of the originating version for each line.
110
To clients the "lines" of the file are represented as a list of strings.
111
These strings will typically have terminal newline characters, but
112
this is not required. In particular files commonly do not have a newline
113
at the end of the file.
115
Texts can be identified in either of two ways:
117
* a nonnegative index number.
119
* a version-id string. (not implemented yet)
121
Typically the index number will be valid only inside this weave and
122
the version-id is used to reference it in the larger world.
124
The weave is represented as a list mixing edit instructions and
125
literal text. Each entry in _weave can be either a string (or
126
unicode), or a tuple. If a string, it means that the given line
127
should be output in the currently active revisions.
129
If a tuple, it gives a processing instruction saying in which
130
revisions the enclosed lines are active. The tuple has the form
131
(instruction, version).
133
The instruction can be '{' or '}' for an insertion block, and '['
134
and ']' for a deletion block respectively. The version is the
135
integer version index. There is no replace operator, only deletes
136
and inserts. For '}', the end of an insertion, there is no
137
version parameter because it always closes the most recently
142
* A later version can delete lines that were introduced by any
143
number of ancestor versions; this implies that deletion
144
instructions can span insertion blocks without regard to the
145
insertion block's nesting.
147
* Similarly, deletions need not be properly nested with regard to
148
each other, because they might have been generated by
149
independent revisions.
151
* Insertions are always made by inserting a new bracketed block
152
into a single point in the previous weave. This implies they
153
can nest but not overlap, and the nesting must always have later
154
insertions on the inside.
156
* It doesn't seem very useful to have an active insertion
157
inside an inactive insertion, but it might happen.
159
* Therefore, all instructions are always"considered"; that
160
is passed onto and off the stack. An outer inactive block
161
doesn't disable an inner block.
163
* Lines are enabled if the most recent enclosing insertion is
164
active and none of the enclosing deletions are active.
166
* There is no point having a deletion directly inside its own
167
insertion; you might as well just not write it. And there
168
should be no way to get an earlier version deleting a later
172
Text of the weave; list of control instruction tuples and strings.
175
List of parents, indexed by version number.
176
It is only necessary to store the minimal set of parents for
177
each version; the parent's parents are implied.
180
List of hex SHA-1 of each version.
183
List of symbolic names for each version. Each should be unique.
186
For each name, the version number.
189
Descriptive name of this weave; typically the filename if known.
193
__slots__ = ['_weave', '_parents', '_sha1s', '_names', '_name_map',
196
def __init__(self, weave_name=None):
202
self._weave_name = 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)
217
def lookup(self, name):
218
assert isinstance(name, basestring), type(name)
220
return self._name_map[name]
222
raise WeaveError("name %r not present in weave %r" %
223
(name, self._weave_name))
226
def idx_to_name(self, version):
227
return self._names[version]
230
def _check_repeated_add(self, name, parents, text):
231
"""Check that a duplicated add is OK.
233
If it is, return the (old) index; otherwise raise an exception.
235
idx = self.lookup(name)
236
if sorted(self._parents[idx]) != sorted(parents):
237
raise WeaveError("name \"%s\" already present in weave "
238
"with different parents" % name)
239
new_sha1 = sha_strings(text)
240
if new_sha1 != self._sha1s[idx]:
241
raise WeaveError("name \"%s\" already present in weave "
242
"with different text" % name)
247
def add(self, name, parents, text):
248
"""Add a single text on top of the weave.
250
Returns the index number of the newly added version.
253
Symbolic name for this version.
254
(Typically the revision-id of the revision that added it.)
257
List or set of direct parent version numbers.
260
Sequence of lines to be added in the new version."""
262
assert isinstance(name, basestring)
263
if name in self._name_map:
264
return self._check_repeated_add(name, parents, text)
266
self._check_versions(parents)
267
## self._check_lines(text)
268
new_version = len(self._parents)
270
sha1 = sha_strings(text)
272
# if we abort after here the (in-memory) weave will be corrupt because only
273
# some fields are updated
274
self._parents.append(parents[:])
275
self._sha1s.append(sha1)
276
self._names.append(name)
277
self._name_map[name] = new_version
281
# special case; adding with no parents revision; can do
282
# this more quickly by just appending unconditionally.
283
# even more specially, if we're adding an empty text we
284
# need do nothing at all.
286
self._weave.append(('{', new_version))
287
self._weave.extend(text)
288
self._weave.append(('}', None))
292
if len(parents) == 1:
293
pv = list(parents)[0]
294
if sha1 == self._sha1s[pv]:
295
# special case: same as the single parent
299
ancestors = self.inclusions(parents)
303
# basis a list of (origin, lineno, line)
306
for origin, lineno, line in self._extract(ancestors):
307
basis_lineno.append(lineno)
308
basis_lines.append(line)
310
# another small special case: a merge, producing the same text
312
if text == basis_lines:
315
# add a sentinal, because we can also match against the final line
316
basis_lineno.append(len(self._weave))
318
# XXX: which line of the weave should we really consider
319
# matches the end of the file? the current code says it's the
320
# last line of the weave?
322
#print 'basis_lines:', basis_lines
323
#print 'new_lines: ', lines
325
from difflib import SequenceMatcher
326
s = SequenceMatcher(None, basis_lines, text)
328
# offset gives the number of lines that have been inserted
329
# into the weave up to the current point; if the original edit instruction
330
# says to change line A then we actually change (A+offset)
333
for tag, i1, i2, j1, j2 in s.get_opcodes():
334
# i1,i2 are given in offsets within basis_lines; we need to map them
335
# back to offsets within the entire weave
336
#print 'raw match', tag, i1, i2, j1, j2
340
i1 = basis_lineno[i1]
341
i2 = basis_lineno[i2]
343
assert 0 <= j1 <= j2 <= len(text)
345
#print tag, i1, i2, j1, j2
347
# the deletion and insertion are handled separately.
348
# first delete the region.
350
self._weave.insert(i1+offset, ('[', new_version))
351
self._weave.insert(i2+offset+1, (']', new_version))
355
# there may have been a deletion spanning up to
356
# i2; we want to insert after this region to make sure
357
# we don't destroy ourselves
359
self._weave[i:i] = ([('{', new_version)]
362
offset += 2 + (j2 - j1)
367
def inclusions(self, versions):
368
"""Return set of all ancestors of given version(s)."""
374
# include all its parents
375
i.update(self._parents[v])
379
raise ValueError("version %d not present in weave" % v)
382
def parents(self, version):
383
return self._parents[version]
386
def minimal_parents(self, version):
387
"""Find the minimal set of parents for the version."""
388
included = self._parents[version]
393
li.sort(reverse=True)
401
gotit.update(self.inclusions(pv))
403
assert mininc[0] >= 0
404
assert mininc[-1] < version
409
def _check_lines(self, text):
410
if not isinstance(text, list):
411
raise ValueError("text should be a list, not %s" % type(text))
414
if not isinstance(l, basestring):
415
raise ValueError("text line should be a string or unicode, not %s"
420
def _check_versions(self, indexes):
421
"""Check everything in the sequence of indexes is valid"""
426
raise IndexError("invalid version number %r" % i)
429
def annotate(self, index):
430
return list(self.annotate_iter(index))
433
def annotate_iter(self, version):
434
"""Yield list of (index-id, line) pairs for the specified version.
436
The index indicates when the line originated in the weave."""
437
for origin, lineno, text in self._extract([version]):
445
(lineno, insert, deletes, text)
446
for each literal line.
452
lineno = 0 # line of weave, 0-based
454
for l in self._weave:
455
if isinstance(l, tuple):
468
raise WeaveFormatError('unexpected instruction %r'
471
assert isinstance(l, basestring)
473
yield lineno, istack[-1], dset, l
478
def _extract(self, versions):
479
"""Yield annotation of lines in included set.
481
Yields a sequence of tuples (origin, lineno, text), where
482
origin is the origin version, lineno the index in the weave,
483
and text the text of the line.
485
The set typically but not necessarily corresponds to a version.
488
if not isinstance(i, int):
491
included = self.inclusions(versions)
496
lineno = 0 # line of weave, 0-based
502
WFE = WeaveFormatError
504
for l in self._weave:
505
if isinstance(l, tuple):
509
assert v not in istack
523
assert isinstance(l, basestring)
525
isactive = (not dset) and istack and (istack[-1] in included)
527
result.append((istack[-1], lineno, l))
531
raise WFE("unclosed insertion blocks at end of weave",
534
raise WFE("unclosed deletion blocks at end of weave",
541
def get_iter(self, version):
542
"""Yield lines for the specified version."""
543
for origin, lineno, line in self._extract([version]):
547
def get_text(self, version):
548
assert isinstance(version, int)
550
s.writelines(self.get_iter(version))
554
def get(self, index):
555
return list(self.get_iter(index))
558
def mash_iter(self, included):
559
"""Return composed version of multiple included versions."""
560
for origin, lineno, text in self._extract(included):
564
def dump(self, to_file):
565
from pprint import pprint
566
print >>to_file, "Weave._weave = ",
567
pprint(self._weave, to_file)
568
print >>to_file, "Weave._parents = ",
569
pprint(self._parents, to_file)
573
def numversions(self):
574
l = len(self._parents)
575
assert l == len(self._sha1s)
580
return self.numversions()
583
def check(self, progress_bar=None):
584
# check no circular inclusions
585
for version in range(self.numversions()):
586
inclusions = list(self._parents[version])
589
if inclusions[-1] >= version:
590
raise WeaveFormatError("invalid included version %d for index %d"
591
% (inclusions[-1], version))
593
# try extracting all versions; this is a bit slow and parallel
594
# extraction could be used
595
nv = self.numversions()
596
for version in range(nv):
598
progress_bar.update('checking text', version, nv)
600
for l in self.get_iter(version):
603
expected = self._sha1s[version]
605
raise WeaveError("mismatched sha1 for version %d; "
606
"got %s, expected %s"
607
% (version, hd, expected))
609
# TODO: check insertions are properly nested, that there are
610
# no lines outside of insertion blocks, that deletions are
611
# properly paired, etc.
615
def merge(self, merge_versions):
616
"""Automerge and mark conflicts between versions.
618
This returns a sequence, each entry describing alternatives
619
for a chunk of the file. Each of the alternatives is given as
622
If there is a chunk of the file where there's no diagreement,
623
only one alternative is given.
626
# approach: find the included versions common to all the
628
raise NotImplementedError()
632
def _delta(self, included, lines):
633
"""Return changes from basis to new revision.
635
The old text for comparison is the union of included revisions.
637
This is used in inserting a new text.
639
Delta is returned as a sequence of
640
(weave1, weave2, newlines).
642
This indicates that weave1:weave2 of the old weave should be
643
replaced by the sequence of lines in newlines. Note that
644
these line numbers are positions in the total weave and don't
645
correspond to the lines in any extracted version, or even the
646
extracted union of included versions.
648
If line1=line2, this is a pure insert; if newlines=[] this is a
649
pure delete. (Similar to difflib.)
654
def plan_merge(self, ver_a, ver_b):
655
"""Return pseudo-annotation indicating how the two versions merge.
657
This is computed between versions a and b and their common
660
Weave lines present in none of them are skipped entirely.
662
inc_a = self.inclusions([ver_a])
663
inc_b = self.inclusions([ver_b])
664
inc_c = inc_a & inc_b
666
for lineno, insert, deleteset, line in self._walk():
667
if deleteset & inc_c:
668
# killed in parent; can't be in either a or b
669
# not relevant to our work
670
yield 'killed-base', line
671
elif insert in inc_c:
672
# was inserted in base
673
killed_a = bool(deleteset & inc_a)
674
killed_b = bool(deleteset & inc_b)
675
if killed_a and killed_b:
676
yield 'killed-both', line
678
yield 'killed-a', line
680
yield 'killed-b', line
682
yield 'unchanged', line
683
elif insert in inc_a:
684
if deleteset & inc_a:
685
yield 'ghost-a', line
689
elif insert in inc_b:
690
if deleteset & inc_b:
691
yield 'ghost-b', line
695
# not in either revision
696
yield 'irrelevant', line
698
yield 'unchanged', '' # terminator
702
def weave_merge(self, plan):
707
for state, line in plan:
708
if state == 'unchanged' or state == 'killed-both':
709
# resync and flush queued conflicts changes if any
710
if not lines_a and not lines_b:
712
elif ch_a and not ch_b:
714
for l in lines_a: yield l
715
elif ch_b and not ch_a:
716
for l in lines_b: yield l
717
elif lines_a == lines_b:
718
for l in lines_a: yield l
721
for l in lines_a: yield l
723
for l in lines_b: yield l
730
if state == 'unchanged':
733
elif state == 'killed-a':
736
elif state == 'killed-b':
739
elif state == 'new-a':
742
elif state == 'new-b':
746
assert state in ('irrelevant', 'ghost-a', 'ghost-b', 'killed-base',
757
"""Show the weave's table-of-contents"""
758
print '%6s %50s %10s %10s' % ('ver', 'name', 'sha1', 'parents')
759
for i in (6, 50, 10, 10):
762
for i in range(w.numversions()):
765
parent_str = ' '.join(map(str, w._parents[i]))
766
print '%6d %-50.50s %10.10s %s' % (i, name, sha1, parent_str)
770
def weave_stats(weave_file):
771
from bzrlib.progress import ProgressBar
772
from bzrlib.weavefile import read_weave
776
wf = file(weave_file, 'rb')
778
# FIXME: doesn't work on pipes
779
weave_size = wf.tell()
783
for i in range(vers):
784
pb.update('checking sizes', i, vers)
785
for line in w.get_iter(i):
790
print 'versions %9d' % vers
791
print 'weave file %9d bytes' % weave_size
792
print 'total contents %9d bytes' % total
793
print 'compression ratio %9.2fx' % (float(total) / float(weave_size))
796
print 'average size %9d bytes' % avg
797
print 'relative size %9.2fx' % (float(weave_size) / float(avg))
801
print """bzr weave tool
803
Experimental tool for weave algorithm.
807
Create an empty weave file
808
weave get WEAVEFILE VERSION
809
Write out specified version.
810
weave check WEAVEFILE
811
Check consistency of all versions.
813
Display table of contents.
814
weave add WEAVEFILE NAME [BASE...] < NEWTEXT
815
Add NEWTEXT, with specified parent versions.
816
weave annotate WEAVEFILE VERSION
817
Display origin of each line.
818
weave mash WEAVEFILE VERSION...
819
Display composite of all selected versions.
820
weave merge WEAVEFILE VERSION1 VERSION2 > OUT
821
Auto-merge two versions and display conflicts.
825
% weave init foo.weave
827
% weave add foo.weave ver0 < foo.txt
830
(create updated version)
832
% weave get foo.weave 0 | diff -u - foo.txt
833
% weave add foo.weave ver1 0 < foo.txt
836
% weave get foo.weave 0 > foo.txt (create forked version)
838
% weave add foo.weave ver2 0 < foo.txt
841
% weave merge foo.weave 1 2 > foo.txt (merge them)
842
% vi foo.txt (resolve conflicts)
843
% weave add foo.weave merged 1 2 < foo.txt (commit merged version)
852
from weavefile import write_weave, read_weave
853
from bzrlib.progress import ProgressBar
868
return read_weave(file(argv[2], 'rb'))
874
# at the moment, based on everything in the file
876
parents = map(int, argv[4:])
877
lines = sys.stdin.readlines()
878
ver = w.add(name, parents, lines)
879
write_weave(w, file(argv[2], 'wb'))
880
print 'added version %r %d' % (name, ver)
883
if os.path.exists(fn):
884
raise IOError("file exists")
886
write_weave(w, file(fn, 'wb'))
887
elif cmd == 'get': # get one version
889
sys.stdout.writelines(w.get_iter(int(argv[3])))
891
elif cmd == 'mash': # get composite
893
sys.stdout.writelines(w.mash_iter(map(int, argv[3:])))
895
elif cmd == 'annotate':
897
# newline is added to all lines regardless; too hard to get
898
# reasonable formatting otherwise
900
for origin, text in w.annotate(int(argv[3])):
901
text = text.rstrip('\r\n')
903
print ' | %s' % (text)
905
print '%5d | %s' % (origin, text)
919
print '%d versions ok' % w.numversions()
921
elif cmd == 'inclusions':
923
print ' '.join(map(str, w.inclusions([int(argv[3])])))
925
elif cmd == 'parents':
927
print ' '.join(map(str, w._parents[int(argv[3])]))
929
elif cmd == 'plan-merge':
931
for state, line in w.plan_merge(int(argv[3]), int(argv[4])):
933
print '%14s | %s' % (state, line),
937
p = w.plan_merge(int(argv[3]), int(argv[4]))
938
sys.stdout.writelines(w.weave_merge(p))
940
elif cmd == 'mash-merge':
946
v1, v2 = map(int, argv[3:5])
948
basis = w.inclusions([v1]).intersection(w.inclusions([v2]))
950
base_lines = list(w.mash_iter(basis))
951
a_lines = list(w.get(v1))
952
b_lines = list(w.get(v2))
954
from bzrlib.merge3 import Merge3
955
m3 = Merge3(base_lines, a_lines, b_lines)
957
name_a = 'version %d' % v1
958
name_b = 'version %d' % v2
959
sys.stdout.writelines(m3.merge_lines(name_a=name_a, name_b=name_b))
961
raise ValueError('unknown command %r' % cmd)
965
def profile_main(argv):
966
import tempfile, hotshot, hotshot.stats
968
prof_f = tempfile.NamedTemporaryFile()
970
prof = hotshot.Profile(prof_f.name)
972
ret = prof.runcall(main, argv)
975
stats = hotshot.stats.load(prof_f.name)
977
stats.sort_stats('cumulative')
978
## XXX: Might like to write to stderr or the trace file instead but
979
## print_stats seems hardcoded to stdout
980
stats.print_stats(20)
985
if __name__ == '__main__':
987
if '--profile' in sys.argv:
989
args.remove('--profile')
990
sys.exit(profile_main(args))
992
sys.exit(main(sys.argv))