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
# TODO: Perhaps have copy method for Weave instances?
26
# XXX: If we do weaves this way, will a merge still behave the same
27
# way if it's done in a different order? That's a pretty desirable
30
# TODO: How to write these to disk? One option is cPickle, which
31
# would be fast but less friendly to C, and perhaps not portable. Another is
33
# TODO: Nothing here so far assumes the lines are really \n newlines,
34
# rather than being split up in some other way. We could accomodate
35
# binaries, perhaps by naively splitting on \n or perhaps using
36
# something like a rolling checksum.
38
# TODO: Perhaps track SHA-1 in the header for protection? This would
39
# be redundant with it being stored in the inventory, but perhaps
42
# TODO: Track version names as well as indexes.
44
# TODO: Probably do transitive expansion when specifying parents?
46
# TODO: Separate out some code to read and write weaves.
48
# TODO: End marker for each version so we can stop reading?
50
# TODO: Check that no insertion occurs inside a deletion that was
51
# active in the version of the insertion.
53
# TODO: Perhaps a special slower check() method that verifies more
54
# nesting constraints and the MD5 of each version?
62
from sets import Set, ImmutableSet
64
frozenset = ImmutableSet
68
class WeaveError(Exception):
69
"""Exception in processing weave"""
72
class WeaveFormatError(WeaveError):
73
"""Weave invariant violated"""
77
"""weave - versioned text file storage.
79
A Weave manages versions of line-based text files, keeping track
80
of the originating version for each line.
82
To clients the "lines" of the file are represented as a list of strings.
83
These strings will typically have terminal newline characters, but
84
this is not required. In particular files commonly do not have a newline
85
at the end of the file.
87
Texts can be identified in either of two ways:
89
* a nonnegative index number.
91
* a version-id string.
93
Typically the index number will be valid only inside this weave and
94
the version-id is used to reference it in the larger world.
96
The weave is represented as a list mixing edit instructions and
97
literal text. Each entry in _l can be either a string (or
98
unicode), or a tuple. If a string, it means that the given line
99
should be output in the currently active revisions.
101
If a tuple, it gives a processing instruction saying in which
102
revisions the enclosed lines are active. The tuple has the form
103
(instruction, version).
105
The instruction can be '{' or '}' for an insertion block, and '['
106
and ']' for a deletion block respectively. The version is the
107
integer version index. There is no replace operator, only deletes
112
* A later version can delete lines that were introduced by any
113
number of ancestor versions; this implies that deletion
114
instructions can span insertion blocks without regard to the
115
insertion block's nesting.
117
* Similarly, deletions need not be properly nested with regard to
118
each other, because they might have been generated by
119
independent revisions.
121
* Insertions are always made by inserting a new bracketed block
122
into a single point in the previous weave. This implies they
123
can nest but not overlap, and the nesting must always have later
124
insertions on the inside.
126
* It doesn't seem very useful to have an active insertion
127
inside an inactive insertion, but it might happen.
129
* Therefore, all instructions are always"considered"; that
130
is passed onto and off the stack. An outer inactive block
131
doesn't disable an inner block.
133
* Lines are enabled if the most recent enclosing insertion is
134
active and none of the enclosing deletions are active.
136
* There is no point having a deletion directly inside its own
137
insertion; you might as well just not write it. And there
138
should be no way to get an earlier version deleting a later
145
List of versions, indexed by index number.
147
For each version we store the set (included_versions), which
148
lists the previous versions also considered active; the
149
versions included in those versions are included transitively.
150
So new versions created from nothing list []; most versions
151
have a single entry; some have more.
154
List of hex SHA-1 of each version, or None if not recorded.
162
def __eq__(self, other):
163
if not isinstance(other, Weave):
165
return self._v == other._v \
166
and self._l == other._l
169
def __ne__(self, other):
170
return not self.__eq__(other)
173
def add(self, parents, text):
174
"""Add a single text on top of the weave.
176
Returns the index number of the newly added version.
179
List or set of parent version numbers. This must normally include
180
the parents and the parent's parents, or wierd things might happen.
183
Sequence of lines to be added in the new version."""
184
## self._check_versions(parents)
185
## self._check_lines(text)
196
delta = self._delta(self.inclusions(parents), text)
198
# offset gives the number of lines that have been inserted
199
# into the weave up to the current point; if the original edit instruction
200
# says to change line A then we actually change (A+offset)
203
for i1, i2, newlines in delta:
206
assert i2 <= len(self._l)
208
# the deletion and insertion are handled separately.
209
# first delete the region.
211
self._l.insert(i1+offset, ('[', idx))
212
self._l.insert(i2+offset+1, (']', idx))
217
# there may have been a deletion spanning up to
218
# i2; we want to insert after this region to make sure
219
# we don't destroy ourselves
221
self._l[i:i] = [('{', idx)] \
224
offset += 2 + len(newlines)
226
self._addversion(parents)
228
# special case; adding with no parents revision; can do this
229
# more quickly by just appending unconditionally
230
self._l.append(('{', idx))
232
self._l.append(('}', idx))
234
self._addversion(None)
236
self._sha1s.append(sha1)
241
def inclusions(self, versions):
242
"""Expand out everything included by versions."""
248
raise ValueError("version %d not present in weave" % v)
252
def minimal_parents(self, version):
253
"""Find the minimal set of parents for the version."""
254
included = self._v[version]
268
gotit.update(self._v[pv])
270
assert mininc[0] >= 0
271
assert mininc[-1] < version
275
def _addversion(self, parents):
277
self._v.append(frozenset(parents))
279
self._v.append(frozenset())
282
def _check_lines(self, text):
283
if not isinstance(text, list):
284
raise ValueError("text should be a list, not %s" % type(text))
287
if not isinstance(l, basestring):
288
raise ValueError("text line should be a string or unicode, not %s"
293
def _check_versions(self, indexes):
294
"""Check everything in the sequence of indexes is valid"""
299
raise IndexError("invalid version number %r" % i)
302
def annotate(self, index):
303
return list(self.annotate_iter(index))
306
def annotate_iter(self, version):
307
"""Yield list of (index-id, line) pairs for the specified version.
309
The index indicates when the line originated in the weave."""
310
included = self.inclusions([version])
311
for origin, lineno, text in self._extract(included):
315
def _extract(self, included):
316
"""Yield annotation of lines in included set.
318
Yields a sequence of tuples (origin, lineno, text), where
319
origin is the origin version, lineno the index in the weave,
320
and text the text of the line.
322
The set typically but not necessarily corresponds to a version.
328
lineno = 0 # line of weave, 0-based
331
WFE = WeaveFormatError
334
if isinstance(l, tuple):
336
if v in included: # only active blocks are interesting
338
assert v not in istack
344
isactive = istack and not dset
353
isactive = istack and not dset
355
assert isinstance(l, basestring)
357
yield istack[-1], lineno, l
361
raise WFE("unclosed insertion blocks at end of weave",
364
raise WFE("unclosed deletion blocks at end of weave",
368
def get_iter(self, version):
369
"""Yield lines for the specified version."""
370
for origin, lineno, line in self._extract(self.inclusions([version])):
374
def get(self, index):
375
return list(self.get_iter(index))
378
def mash_iter(self, included):
379
"""Return composed version of multiple included versions."""
380
included = frozenset(included)
381
for origin, lineno, text in self._extract(included):
385
def dump(self, to_file):
386
from pprint import pprint
387
print >>to_file, "Weave._l = ",
388
pprint(self._l, to_file)
389
print >>to_file, "Weave._v = ",
390
pprint(self._v, to_file)
394
def numversions(self):
396
assert l == len(self._sha1s)
401
# check no circular inclusions
402
for version in range(self.numversions()):
403
inclusions = list(self._v[version])
406
if inclusions[-1] >= version:
407
raise WeaveFormatError("invalid included version %d for index %d"
408
% (inclusions[-1], version))
410
# try extracting all versions; this is a bit slow and parallel
411
# extraction could be used
413
for version in range(self.numversions()):
415
for l in self.get_iter(version):
418
expected = self._sha1s[version]
420
raise WeaveError("mismatched sha1 for version %d; "
421
"got %s, expected %s"
422
% (version, hd, expected))
424
# TODO: check insertions are properly nested, that there are
425
# no lines outside of insertion blocks, that deletions are
426
# properly paired, etc.
430
def merge(self, merge_versions):
431
"""Automerge and mark conflicts between versions.
433
This returns a sequence, each entry describing alternatives
434
for a chunk of the file. Each of the alternatives is given as
437
If there is a chunk of the file where there's no diagreement,
438
only one alternative is given.
441
# approach: find the included versions common to all the
443
raise NotImplementedError()
447
def _delta(self, included, lines):
448
"""Return changes from basis to new revision.
450
The old text for comparison is the union of included revisions.
452
This is used in inserting a new text.
454
Delta is returned as a sequence of
455
(weave1, weave2, newlines).
457
This indicates that weave1:weave2 of the old weave should be
458
replaced by the sequence of lines in newlines. Note that
459
these line numbers are positions in the total weave and don't
460
correspond to the lines in any extracted version, or even the
461
extracted union of included versions.
463
If line1=line2, this is a pure insert; if newlines=[] this is a
464
pure delete. (Similar to difflib.)
466
# basis a list of (origin, lineno, line)
469
for origin, lineno, line in self._extract(included):
470
basis_lineno.append(lineno)
471
basis_lines.append(line)
473
# add a sentinal, because we can also match against the final line
474
basis_lineno.append(len(self._l))
476
# XXX: which line of the weave should we really consider
477
# matches the end of the file? the current code says it's the
478
# last line of the weave?
480
from difflib import SequenceMatcher
481
s = SequenceMatcher(None, basis_lines, lines)
483
# TODO: Perhaps return line numbers from composed weave as well?
485
for tag, i1, i2, j1, j2 in s.get_opcodes():
486
##print tag, i1, i2, j1, j2
491
# i1,i2 are given in offsets within basis_lines; we need to map them
492
# back to offsets within the entire weave
493
real_i1 = basis_lineno[i1]
494
real_i2 = basis_lineno[i2]
498
assert j2 <= len(lines)
500
yield real_i1, real_i2, lines[j1:j2]
504
def weave_info(filename, out):
505
"""Show some text information about the weave."""
506
from weavefile import read_weave
507
wf = file(filename, 'rb')
509
# FIXME: doesn't work on pipes
510
weave_size = wf.tell()
511
print >>out, "weave file size %d bytes" % weave_size
512
print >>out, "weave contains %d versions" % len(w._v)
515
print '%6s %6s %8s %40s %20s' % ('ver', 'lines', 'bytes', 'sha1', 'parents')
516
for i in (6, 6, 8, 40, 20):
519
for i in range(len(w._v)):
522
bytes = sum((len(a) for a in text))
524
print '%6d %6d %8d %40s' % (i, lines, bytes, sha1),
525
print ', '.join(map(str, w.minimal_parents(i)))
528
print >>out, "versions total %d bytes" % total
529
print >>out, "compression ratio %.3f" % (float(total)/float(weave_size))
533
print """bzr weave tool
535
Experimental tool for weave algorithm.
539
Create an empty weave file
540
weave get WEAVEFILE VERSION
541
Write out specified version.
542
weave check WEAVEFILE
543
Check consistency of all versions.
545
Display table of contents.
546
weave add WEAVEFILE [BASE...] < NEWTEXT
547
Add NEWTEXT, with specified parent versions.
548
weave annotate WEAVEFILE VERSION
549
Display origin of each line.
550
weave mash WEAVEFILE VERSION...
551
Display composite of all selected versions.
552
weave merge WEAVEFILE VERSION1 VERSION2 > OUT
553
Auto-merge two versions and display conflicts.
557
% weave init foo.weave
559
% weave add foo.weave < foo.txt
562
(create updated version)
564
% weave get foo.weave 0 | diff -u - foo.txt
565
% weave add foo.weave 0 < foo.txt
568
% weave get foo.weave 0 > foo.txt (create forked version)
570
% weave add foo.weave 0 < foo.txt
573
% weave merge foo.weave 1 2 > foo.txt (merge them)
574
% vi foo.txt (resolve conflicts)
575
% weave add foo.weave 1 2 < foo.txt (commit merged version)
584
from weavefile import write_weave, read_weave
588
return read_weave(file(argv[2], 'rb'))
594
# at the moment, based on everything in the file
595
parents = map(int, argv[3:])
596
lines = sys.stdin.readlines()
597
ver = w.add(parents, lines)
598
write_weave(w, file(argv[2], 'wb'))
599
print 'added version %d' % ver
602
if os.path.exists(fn):
603
raise IOError("file exists")
605
write_weave(w, file(fn, 'wb'))
606
elif cmd == 'get': # get one version
608
sys.stdout.writelines(w.get_iter(int(argv[3])))
610
elif cmd == 'mash': # get composite
612
sys.stdout.writelines(w.mash_iter(map(int, argv[3:])))
614
elif cmd == 'annotate':
616
# newline is added to all lines regardless; too hard to get
617
# reasonable formatting otherwise
619
for origin, text in w.annotate(int(argv[3])):
620
text = text.rstrip('\r\n')
622
print ' | %s' % (text)
624
print '%5d | %s' % (origin, text)
628
weave_info(argv[2], sys.stdout)
640
v1, v2 = map(int, argv[3:5])
642
basis = w.inclusions([v1]).intersection(w.inclusions([v2]))
644
base_lines = list(w.mash_iter(basis))
645
a_lines = list(w.get(v1))
646
b_lines = list(w.get(v2))
648
from bzrlib.merge3 import Merge3
649
m3 = Merge3(base_lines, a_lines, b_lines)
651
name_a = 'version %d' % v1
652
name_b = 'version %d' % v2
653
sys.stdout.writelines(m3.merge_lines(name_a=name_a, name_b=name_b))
655
raise ValueError('unknown command %r' % cmd)
658
if __name__ == '__main__':
660
sys.exit(main(sys.argv))
added
removed
bzrlib/weave.py
