1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
|
# Copyright (C) 2004, 2005 by Canonical Ltd
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
# mbp: "you know that thing where cvs gives you conflict markers?"
# s: "i hate that."
from difflib import SequenceMatcher
from bzrlib.errors import CantReprocessAndShowBase
def intersect(ra, rb):
"""Given two ranges return the range where they intersect or None.
>>> intersect((0, 10), (0, 6))
(0, 6)
>>> intersect((0, 10), (5, 15))
(5, 10)
>>> intersect((0, 10), (10, 15))
>>> intersect((0, 9), (10, 15))
>>> intersect((0, 9), (7, 15))
(7, 9)
"""
assert ra[0] <= ra[1]
assert rb[0] <= rb[1]
sa = max(ra[0], rb[0])
sb = min(ra[1], rb[1])
if sa < sb:
return sa, sb
else:
return None
def compare_range(a, astart, aend, b, bstart, bend):
"""Compare a[astart:aend] == b[bstart:bend], without slicing.
"""
if (aend-astart) != (bend-bstart):
return False
for ia, ib in zip(xrange(astart, aend), xrange(bstart, bend)):
if a[ia] != b[ib]:
return False
else:
return True
class Merge3(object):
"""3-way merge of texts.
Given BASE, OTHER, THIS, tries to produce a combined text
incorporating the changes from both BASE->OTHER and BASE->THIS.
All three will typically be sequences of lines."""
def __init__(self, base, a, b):
self.base = base
self.a = a
self.b = b
def merge_lines(self,
name_a=None,
name_b=None,
name_base=None,
start_marker='<<<<<<<',
mid_marker='=======',
end_marker='>>>>>>>',
base_marker=None,
reprocess=False):
"""Return merge in cvs-like form.
"""
if base_marker and reprocess:
raise CantReprocessAndShowBase()
if name_a:
start_marker = start_marker + ' ' + name_a
if name_b:
end_marker = end_marker + ' ' + name_b
if name_base and base_marker:
base_marker = base_marker + ' ' + name_base
merge_regions = self.merge_regions()
if reprocess is True:
merge_regions = self.reprocess_merge_regions(merge_regions)
for t in merge_regions:
what = t[0]
if what == 'unchanged':
for i in range(t[1], t[2]):
yield self.base[i]
elif what == 'a' or what == 'same':
for i in range(t[1], t[2]):
yield self.a[i]
elif what == 'b':
for i in range(t[1], t[2]):
yield self.b[i]
elif what == 'conflict':
yield start_marker + '\n'
for i in range(t[3], t[4]):
yield self.a[i]
if base_marker is not None:
yield base_marker + '\n'
for i in range(t[1], t[2]):
yield self.base[i]
yield mid_marker + '\n'
for i in range(t[5], t[6]):
yield self.b[i]
yield end_marker + '\n'
else:
raise ValueError(what)
def merge_annotated(self):
"""Return merge with conflicts, showing origin of lines.
Most useful for debugging merge.
"""
for t in self.merge_regions():
what = t[0]
if what == 'unchanged':
for i in range(t[1], t[2]):
yield 'u | ' + self.base[i]
elif what == 'a' or what == 'same':
for i in range(t[1], t[2]):
yield what[0] + ' | ' + self.a[i]
elif what == 'b':
for i in range(t[1], t[2]):
yield 'b | ' + self.b[i]
elif what == 'conflict':
yield '<<<<\n'
for i in range(t[3], t[4]):
yield 'A | ' + self.a[i]
yield '----\n'
for i in range(t[5], t[6]):
yield 'B | ' + self.b[i]
yield '>>>>\n'
else:
raise ValueError(what)
def merge_groups(self):
"""Yield sequence of line groups. Each one is a tuple:
'unchanged', lines
Lines unchanged from base
'a', lines
Lines taken from a
'same', lines
Lines taken from a (and equal to b)
'b', lines
Lines taken from b
'conflict', base_lines, a_lines, b_lines
Lines from base were changed to either a or b and conflict.
"""
for t in self.merge_regions():
what = t[0]
if what == 'unchanged':
yield what, self.base[t[1]:t[2]]
elif what == 'a' or what == 'same':
yield what, self.a[t[1]:t[2]]
elif what == 'b':
yield what, self.b[t[1]:t[2]]
elif what == 'conflict':
yield (what,
self.base[t[1]:t[2]],
self.a[t[3]:t[4]],
self.b[t[5]:t[6]])
else:
raise ValueError(what)
def merge_regions(self):
"""Return sequences of matching and conflicting regions.
This returns tuples, where the first value says what kind we
have:
'unchanged', start, end
Take a region of base[start:end]
'same', astart, aend
b and a are different from base but give the same result
'a', start, end
Non-clashing insertion from a[start:end]
Method is as follows:
The two sequences align only on regions which match the base
and both descendents. These are found by doing a two-way diff
of each one against the base, and then finding the
intersections between those regions. These "sync regions"
are by definition unchanged in both and easily dealt with.
The regions in between can be in any of three cases:
conflicted, or changed on only one side.
"""
# section a[0:ia] has been disposed of, etc
iz = ia = ib = 0
for zmatch, zend, amatch, aend, bmatch, bend in self.find_sync_regions():
#print 'match base [%d:%d]' % (zmatch, zend)
matchlen = zend - zmatch
assert matchlen >= 0
assert matchlen == (aend - amatch)
assert matchlen == (bend - bmatch)
len_a = amatch - ia
len_b = bmatch - ib
len_base = zmatch - iz
assert len_a >= 0
assert len_b >= 0
assert len_base >= 0
#print 'unmatched a=%d, b=%d' % (len_a, len_b)
if len_a or len_b:
# try to avoid actually slicing the lists
equal_a = compare_range(self.a, ia, amatch,
self.base, iz, zmatch)
equal_b = compare_range(self.b, ib, bmatch,
self.base, iz, zmatch)
same = compare_range(self.a, ia, amatch,
self.b, ib, bmatch)
if same:
yield 'same', ia, amatch
elif equal_a and not equal_b:
yield 'b', ib, bmatch
elif equal_b and not equal_a:
yield 'a', ia, amatch
elif not equal_a and not equal_b:
yield 'conflict', iz, zmatch, ia, amatch, ib, bmatch
else:
raise AssertionError("can't handle a=b=base but unmatched")
ia = amatch
ib = bmatch
iz = zmatch
# if the same part of the base was deleted on both sides
# that's OK, we can just skip it.
if matchlen > 0:
assert ia == amatch
assert ib == bmatch
assert iz == zmatch
yield 'unchanged', zmatch, zend
iz = zend
ia = aend
ib = bend
def reprocess_merge_regions(self, merge_regions):
"""Where there are conflict regions, remove the agreed lines.
Lines where both A and B have made the same changes are
eliminated.
"""
for region in merge_regions:
if region[0] != "conflict":
yield region
continue
type, iz, zmatch, ia, amatch, ib, bmatch = region
a_region = self.a[ia:amatch]
b_region = self.b[ib:bmatch]
matches = SequenceMatcher(None, a_region,
b_region).get_matching_blocks()
next_a = ia
next_b = ib
for region_ia, region_ib, region_len in matches[:-1]:
region_ia += ia
region_ib += ib
reg = self.mismatch_region(next_a, region_ia, next_b,
region_ib)
if reg is not None:
yield reg
yield 'same', region_ia, region_len+region_ia
next_a = region_ia + region_len
next_b = region_ib + region_len
reg = self.mismatch_region(next_a, amatch, next_b, bmatch)
if reg is not None:
yield reg
@staticmethod
def mismatch_region(next_a, region_ia, next_b, region_ib):
if next_a < region_ia or next_b < region_ib:
return 'conflict', None, None, next_a, region_ia, next_b, region_ib
def find_sync_regions(self):
"""Return a list of sync regions, where both descendents match the base.
Generates a list of (base1, base2, a1, a2, b1, b2). There is
always a zero-length sync region at the end of all the files.
"""
ia = ib = 0
amatches = SequenceMatcher(None, self.base, self.a).get_matching_blocks()
bmatches = SequenceMatcher(None, self.base, self.b).get_matching_blocks()
len_a = len(amatches)
len_b = len(bmatches)
sl = []
while ia < len_a and ib < len_b:
abase, amatch, alen = amatches[ia]
bbase, bmatch, blen = bmatches[ib]
# there is an unconflicted block at i; how long does it
# extend? until whichever one ends earlier.
i = intersect((abase, abase+alen), (bbase, bbase+blen))
if i:
intbase = i[0]
intend = i[1]
intlen = intend - intbase
# found a match of base[i[0], i[1]]; this may be less than
# the region that matches in either one
assert intlen <= alen
assert intlen <= blen
assert abase <= intbase
assert bbase <= intbase
asub = amatch + (intbase - abase)
bsub = bmatch + (intbase - bbase)
aend = asub + intlen
bend = bsub + intlen
assert self.base[intbase:intend] == self.a[asub:aend], \
(self.base[intbase:intend], self.a[asub:aend])
assert self.base[intbase:intend] == self.b[bsub:bend]
sl.append((intbase, intend,
asub, aend,
bsub, bend))
# advance whichever one ends first in the base text
if (abase + alen) < (bbase + blen):
ia += 1
else:
ib += 1
intbase = len(self.base)
abase = len(self.a)
bbase = len(self.b)
sl.append((intbase, intbase, abase, abase, bbase, bbase))
return sl
def find_unconflicted(self):
"""Return a list of ranges in base that are not conflicted."""
import re
# don't sync-up on lines containing only blanks or pounds
junk_re = re.compile(r'^[ \t#]*$')
am = SequenceMatcher(junk_re.match, self.base, self.a).get_matching_blocks()
bm = SequenceMatcher(junk_re.match, self.base, self.b).get_matching_blocks()
unc = []
while am and bm:
# there is an unconflicted block at i; how long does it
# extend? until whichever one ends earlier.
a1 = am[0][0]
a2 = a1 + am[0][2]
b1 = bm[0][0]
b2 = b1 + bm[0][2]
i = intersect((a1, a2), (b1, b2))
if i:
unc.append(i)
if a2 < b2:
del am[0]
else:
del bm[0]
return unc
def main(argv):
# as for diff3 and meld the syntax is "MINE BASE OTHER"
a = file(argv[1], 'rt').readlines()
base = file(argv[2], 'rt').readlines()
b = file(argv[3], 'rt').readlines()
m3 = Merge3(base, a, b)
#for sr in m3.find_sync_regions():
# print sr
# sys.stdout.writelines(m3.merge_lines(name_a=argv[1], name_b=argv[3]))
sys.stdout.writelines(m3.merge_annotated())
if __name__ == '__main__':
import sys
sys.exit(main(sys.argv))
|