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."""
249
def _addversion(self, parents):
251
self._v.append(frozenset(parents))
253
self._v.append(frozenset())
256
def _check_lines(self, text):
257
if not isinstance(text, list):
258
raise ValueError("text should be a list, not %s" % type(text))
261
if not isinstance(l, basestring):
262
raise ValueError("text line should be a string or unicode, not %s" % type(l))
266
def _check_versions(self, indexes):
267
"""Check everything in the sequence of indexes is valid"""
272
raise IndexError("invalid version number %r" % i)
275
def annotate(self, index):
276
return list(self.annotate_iter(index))
279
def annotate_iter(self, version):
280
"""Yield list of (index-id, line) pairs for the specified version.
282
The index indicates when the line originated in the weave."""
283
included = self.inclusions([version])
284
for origin, lineno, text in self._extract(included):
288
def _extract(self, included):
289
"""Yield annotation of lines in included set.
291
Yields a sequence of tuples (origin, lineno, text), where
292
origin is the origin version, lineno the index in the weave,
293
and text the text of the line.
295
The set typically but not necessarily corresponds to a version.
297
istack = [] # versions for which an insertion block is current
299
dset = set() # versions for which a deletion block is current
303
lineno = 0 # line of weave, 0-based
305
# TODO: Probably only need to put included revisions in the istack
307
# TODO: Could split this into two functions, one that updates
308
# the stack and the other that processes the results -- but
309
# I'm not sure it's really needed.
311
# TODO: In fact, I think we only need to store the *count* of
312
# active insertions and deletions, and we can maintain that by
313
# just by just counting as we go along.
315
WFE = WeaveFormatError
318
if isinstance(l, tuple):
319
isactive = None # recalculate
322
if istack and (istack[-1] >= v):
323
raise WFE("improperly nested insertions %d>=%d on line %d"
324
% (istack[-1], v, lineno))
330
raise WFE("unmatched close of insertion %d on line %d"
333
raise WFE("mismatched close of insertion %d!=%d on line %d"
338
raise WFE("repeated deletion marker for version %d on line %d"
342
raise WFE("version %d deletes own text on line %d"
350
raise WFE("unmatched close of deletion %d on line %d"
353
raise WFE("invalid processing instruction %r on line %d"
356
assert isinstance(l, basestring)
358
raise WFE("literal at top level on line %d"
361
isactive = (istack[-1] in included) \
362
and not included.intersection(dset)
365
yield origin, lineno, l
369
raise WFE("unclosed insertion blocks at end of weave",
372
raise WFE("unclosed deletion blocks at end of weave",
376
def get_iter(self, version):
377
"""Yield lines for the specified version."""
378
for origin, lineno, line in self._extract(self.inclusions([version])):
382
def get(self, index):
383
return list(self.get_iter(index))
386
def mash_iter(self, included):
387
"""Return composed version of multiple included versions."""
388
included = frozenset(included)
389
for origin, lineno, text in self._extract(included):
393
def dump(self, to_file):
394
from pprint import pprint
395
print >>to_file, "Weave._l = ",
396
pprint(self._l, to_file)
397
print >>to_file, "Weave._v = ",
398
pprint(self._v, to_file)
402
def numversions(self):
404
assert l == len(self._sha1s)
409
# check no circular inclusions
410
for version in range(self.numversions()):
411
inclusions = list(self._v[version])
414
if inclusions[-1] >= version:
415
raise WeaveFormatError("invalid included version %d for index %d"
416
% (inclusions[-1], version))
418
# try extracting all versions; this is a bit slow and parallel
419
# extraction could be used
421
for version in range(self.numversions()):
423
for l in self.get_iter(version):
426
expected = self._sha1s[version]
428
raise WeaveError("mismatched sha1 for version %d; "
429
"got %s, expected %s"
430
% (version, hd, expected))
434
def merge(self, merge_versions):
435
"""Automerge and mark conflicts between versions.
437
This returns a sequence, each entry describing alternatives
438
for a chunk of the file. Each of the alternatives is given as
441
If there is a chunk of the file where there's no diagreement,
442
only one alternative is given.
445
# approach: find the included versions common to all the
447
raise NotImplementedError()
451
def _delta(self, included, lines):
452
"""Return changes from basis to new revision.
454
The old text for comparison is the union of included revisions.
456
This is used in inserting a new text.
458
Delta is returned as a sequence of
459
(weave1, weave2, newlines).
461
This indicates that weave1:weave2 of the old weave should be
462
replaced by the sequence of lines in newlines. Note that
463
these line numbers are positions in the total weave and don't
464
correspond to the lines in any extracted version, or even the
465
extracted union of included versions.
467
If line1=line2, this is a pure insert; if newlines=[] this is a
468
pure delete. (Similar to difflib.)
470
# basis a list of (origin, lineno, line)
473
for origin, lineno, line in self._extract(included):
474
basis_lineno.append(lineno)
475
basis_lines.append(line)
477
# add a sentinal, because we can also match against the final line
478
basis_lineno.append(len(self._l))
480
# XXX: which line of the weave should we really consider
481
# matches the end of the file? the current code says it's the
482
# last line of the weave?
484
from difflib import SequenceMatcher
485
s = SequenceMatcher(None, basis_lines, lines)
487
# TODO: Perhaps return line numbers from composed weave as well?
489
for tag, i1, i2, j1, j2 in s.get_opcodes():
490
##print tag, i1, i2, j1, j2
495
# i1,i2 are given in offsets within basis_lines; we need to map them
496
# back to offsets within the entire weave
497
real_i1 = basis_lineno[i1]
498
real_i2 = basis_lineno[i2]
502
assert j2 <= len(lines)
504
yield real_i1, real_i2, lines[j1:j2]
508
def weave_info(filename, out):
509
"""Show some text information about the weave."""
510
from weavefile import read_weave
511
wf = file(filename, 'rb')
513
# FIXME: doesn't work on pipes
514
weave_size = wf.tell()
515
print >>out, "weave file size %d bytes" % weave_size
516
print >>out, "weave contains %d versions" % len(w._v)
519
print ' %8s %8s %8s %s' % ('version', 'lines', 'bytes', 'sha1')
520
print ' -------- -------- -------- ----------------------------------------'
521
for i in range(len(w._v)):
524
bytes = sum((len(a) for a in text))
526
print ' %8d %8d %8d %s' % (i, lines, bytes, sha1)
529
print >>out, "versions total %d bytes" % total
530
print >>out, "compression ratio %.3f" % (float(total)/float(weave_size))
537
from weavefile import write_weave_v1, read_weave
540
w = read_weave(file(argv[2], 'rb'))
541
# at the moment, based on everything in the file
542
parents = set(range(len(w._v)))
543
lines = sys.stdin.readlines()
544
ver = w.add(parents, lines)
545
write_weave_v1(w, file(argv[2], 'wb'))
546
print 'added %d' % ver
549
if os.path.exists(fn):
550
raise IOError("file exists")
552
write_weave_v1(w, file(fn, 'wb'))
554
w = read_weave(file(argv[2], 'rb'))
555
sys.stdout.writelines(w.get_iter(int(argv[3])))
556
elif cmd == 'annotate':
557
w = read_weave(file(argv[2], 'rb'))
558
# newline is added to all lines regardless; too hard to get
559
# reasonable formatting otherwise
561
for origin, text in w.annotate(int(argv[3])):
562
text = text.rstrip('\r\n')
564
print ' | %s' % (text)
566
print '%5d | %s' % (origin, text)
569
weave_info(argv[2], sys.stdout)
571
w = read_weave(file(argv[2], 'rb'))
574
raise ValueError('unknown command %r' % cmd)
577
if __name__ == '__main__':
579
sys.exit(main(sys.argv))
added
removed
bzrlib/weave.py
