Show More
@@ -1,477 +1,471 b'' | |||
|
1 | 1 | # Copyright (C) 2004, 2005 Canonical Ltd |
|
2 | 2 | # |
|
3 | 3 | # This program is free software; you can redistribute it and/or modify |
|
4 | 4 | # it under the terms of the GNU General Public License as published by |
|
5 | 5 | # the Free Software Foundation; either version 2 of the License, or |
|
6 | 6 | # (at your option) any later version. |
|
7 | 7 | # |
|
8 | 8 | # This program is distributed in the hope that it will be useful, |
|
9 | 9 | # but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
10 | 10 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
11 | 11 | # GNU General Public License for more details. |
|
12 | 12 | # |
|
13 | 13 | # You should have received a copy of the GNU General Public License |
|
14 | 14 | # along with this program; if not, see <http://www.gnu.org/licenses/>. |
|
15 | 15 | |
|
16 | 16 | # mbp: "you know that thing where cvs gives you conflict markers?" |
|
17 | 17 | # s: "i hate that." |
|
18 | 18 | |
|
19 | 19 | from __future__ import absolute_import |
|
20 | 20 | |
|
21 | 21 | from .i18n import _ |
|
22 | 22 | from . import ( |
|
23 | 23 | error, |
|
24 | 24 | mdiff, |
|
25 | 25 | pycompat, |
|
26 | 26 | util, |
|
27 | 27 | ) |
|
28 | 28 | |
|
29 | 29 | class CantReprocessAndShowBase(Exception): |
|
30 | 30 | pass |
|
31 | 31 | |
|
32 | 32 | def intersect(ra, rb): |
|
33 | 33 | """Given two ranges return the range where they intersect or None. |
|
34 | 34 | |
|
35 | 35 | >>> intersect((0, 10), (0, 6)) |
|
36 | 36 | (0, 6) |
|
37 | 37 | >>> intersect((0, 10), (5, 15)) |
|
38 | 38 | (5, 10) |
|
39 | 39 | >>> intersect((0, 10), (10, 15)) |
|
40 | 40 | >>> intersect((0, 9), (10, 15)) |
|
41 | 41 | >>> intersect((0, 9), (7, 15)) |
|
42 | 42 | (7, 9) |
|
43 | 43 | """ |
|
44 | 44 | assert ra[0] <= ra[1] |
|
45 | 45 | assert rb[0] <= rb[1] |
|
46 | 46 | |
|
47 | 47 | sa = max(ra[0], rb[0]) |
|
48 | 48 | sb = min(ra[1], rb[1]) |
|
49 | 49 | if sa < sb: |
|
50 | 50 | return sa, sb |
|
51 | 51 | else: |
|
52 | 52 | return None |
|
53 | 53 | |
|
54 | 54 | def compare_range(a, astart, aend, b, bstart, bend): |
|
55 | 55 | """Compare a[astart:aend] == b[bstart:bend], without slicing. |
|
56 | 56 | """ |
|
57 | 57 | if (aend - astart) != (bend - bstart): |
|
58 | 58 | return False |
|
59 | 59 | for ia, ib in zip(xrange(astart, aend), xrange(bstart, bend)): |
|
60 | 60 | if a[ia] != b[ib]: |
|
61 | 61 | return False |
|
62 | 62 | else: |
|
63 | 63 | return True |
|
64 | 64 | |
|
65 | 65 | class Merge3Text(object): |
|
66 | 66 | """3-way merge of texts. |
|
67 | 67 | |
|
68 | 68 | Given strings BASE, OTHER, THIS, tries to produce a combined text |
|
69 | 69 | incorporating the changes from both BASE->OTHER and BASE->THIS.""" |
|
70 | 70 | def __init__(self, basetext, atext, btext, base=None, a=None, b=None): |
|
71 | 71 | self.basetext = basetext |
|
72 | 72 | self.atext = atext |
|
73 | 73 | self.btext = btext |
|
74 | 74 | if base is None: |
|
75 | 75 | base = mdiff.splitnewlines(basetext) |
|
76 | 76 | if a is None: |
|
77 | 77 | a = mdiff.splitnewlines(atext) |
|
78 | 78 | if b is None: |
|
79 | 79 | b = mdiff.splitnewlines(btext) |
|
80 | 80 | self.base = base |
|
81 | 81 | self.a = a |
|
82 | 82 | self.b = b |
|
83 | 83 | |
|
84 | 84 | def merge_lines(self, |
|
85 | 85 | name_a=None, |
|
86 | 86 | name_b=None, |
|
87 | 87 | name_base=None, |
|
88 | 88 | start_marker='<<<<<<<', |
|
89 | 89 | mid_marker='=======', |
|
90 | 90 | end_marker='>>>>>>>', |
|
91 | 91 | base_marker=None, |
|
92 | 92 | localorother=None, |
|
93 | 93 | minimize=False): |
|
94 | 94 | """Return merge in cvs-like form. |
|
95 | 95 | """ |
|
96 | 96 | self.conflicts = False |
|
97 | 97 | newline = '\n' |
|
98 | 98 | if len(self.a) > 0: |
|
99 | 99 | if self.a[0].endswith('\r\n'): |
|
100 | 100 | newline = '\r\n' |
|
101 | 101 | elif self.a[0].endswith('\r'): |
|
102 | 102 | newline = '\r' |
|
103 | 103 | if name_a and start_marker: |
|
104 | 104 | start_marker = start_marker + ' ' + name_a |
|
105 | 105 | if name_b and end_marker: |
|
106 | 106 | end_marker = end_marker + ' ' + name_b |
|
107 | 107 | if name_base and base_marker: |
|
108 | 108 | base_marker = base_marker + ' ' + name_base |
|
109 | 109 | merge_regions = self.merge_regions() |
|
110 | 110 | if minimize: |
|
111 | 111 | merge_regions = self.minimize(merge_regions) |
|
112 | 112 | for t in merge_regions: |
|
113 | 113 | what = t[0] |
|
114 | 114 | if what == 'unchanged': |
|
115 | 115 | for i in range(t[1], t[2]): |
|
116 | 116 | yield self.base[i] |
|
117 | 117 | elif what == 'a' or what == 'same': |
|
118 | 118 | for i in range(t[1], t[2]): |
|
119 | 119 | yield self.a[i] |
|
120 | 120 | elif what == 'b': |
|
121 | 121 | for i in range(t[1], t[2]): |
|
122 | 122 | yield self.b[i] |
|
123 | 123 | elif what == 'conflict': |
|
124 | 124 | if localorother == 'local': |
|
125 | 125 | for i in range(t[3], t[4]): |
|
126 | 126 | yield self.a[i] |
|
127 | 127 | elif localorother == 'other': |
|
128 | 128 | for i in range(t[5], t[6]): |
|
129 | 129 | yield self.b[i] |
|
130 | 130 | else: |
|
131 | 131 | self.conflicts = True |
|
132 | 132 | if start_marker is not None: |
|
133 | 133 | yield start_marker + newline |
|
134 | 134 | for i in range(t[3], t[4]): |
|
135 | 135 | yield self.a[i] |
|
136 | 136 | if base_marker is not None: |
|
137 | 137 | yield base_marker + newline |
|
138 | 138 | for i in range(t[1], t[2]): |
|
139 | 139 | yield self.base[i] |
|
140 | 140 | if mid_marker is not None: |
|
141 | 141 | yield mid_marker + newline |
|
142 | 142 | for i in range(t[5], t[6]): |
|
143 | 143 | yield self.b[i] |
|
144 | 144 | if end_marker is not None: |
|
145 | 145 | yield end_marker + newline |
|
146 | 146 | else: |
|
147 | 147 | raise ValueError(what) |
|
148 | 148 | |
|
149 | 149 | def merge_groups(self): |
|
150 | 150 | """Yield sequence of line groups. Each one is a tuple: |
|
151 | 151 | |
|
152 | 152 | 'unchanged', lines |
|
153 | 153 | Lines unchanged from base |
|
154 | 154 | |
|
155 | 155 | 'a', lines |
|
156 | 156 | Lines taken from a |
|
157 | 157 | |
|
158 | 158 | 'same', lines |
|
159 | 159 | Lines taken from a (and equal to b) |
|
160 | 160 | |
|
161 | 161 | 'b', lines |
|
162 | 162 | Lines taken from b |
|
163 | 163 | |
|
164 | 164 | 'conflict', base_lines, a_lines, b_lines |
|
165 | 165 | Lines from base were changed to either a or b and conflict. |
|
166 | 166 | """ |
|
167 | 167 | for t in self.merge_regions(): |
|
168 | 168 | what = t[0] |
|
169 | 169 | if what == 'unchanged': |
|
170 | 170 | yield what, self.base[t[1]:t[2]] |
|
171 | 171 | elif what == 'a' or what == 'same': |
|
172 | 172 | yield what, self.a[t[1]:t[2]] |
|
173 | 173 | elif what == 'b': |
|
174 | 174 | yield what, self.b[t[1]:t[2]] |
|
175 | 175 | elif what == 'conflict': |
|
176 | 176 | yield (what, |
|
177 | 177 | self.base[t[1]:t[2]], |
|
178 | 178 | self.a[t[3]:t[4]], |
|
179 | 179 | self.b[t[5]:t[6]]) |
|
180 | 180 | else: |
|
181 | 181 | raise ValueError(what) |
|
182 | 182 | |
|
183 | 183 | def merge_regions(self): |
|
184 | 184 | """Return sequences of matching and conflicting regions. |
|
185 | 185 | |
|
186 | 186 | This returns tuples, where the first value says what kind we |
|
187 | 187 | have: |
|
188 | 188 | |
|
189 | 189 | 'unchanged', start, end |
|
190 | 190 | Take a region of base[start:end] |
|
191 | 191 | |
|
192 | 192 | 'same', astart, aend |
|
193 | 193 | b and a are different from base but give the same result |
|
194 | 194 | |
|
195 | 195 | 'a', start, end |
|
196 | 196 | Non-clashing insertion from a[start:end] |
|
197 | 197 | |
|
198 | 198 | 'conflict', zstart, zend, astart, aend, bstart, bend |
|
199 | 199 | Conflict between a and b, with z as common ancestor |
|
200 | 200 | |
|
201 | 201 | Method is as follows: |
|
202 | 202 | |
|
203 | 203 | The two sequences align only on regions which match the base |
|
204 | 204 | and both descendants. These are found by doing a two-way diff |
|
205 | 205 | of each one against the base, and then finding the |
|
206 | 206 | intersections between those regions. These "sync regions" |
|
207 | 207 | are by definition unchanged in both and easily dealt with. |
|
208 | 208 | |
|
209 | 209 | The regions in between can be in any of three cases: |
|
210 | 210 | conflicted, or changed on only one side. |
|
211 | 211 | """ |
|
212 | 212 | |
|
213 | 213 | # section a[0:ia] has been disposed of, etc |
|
214 | 214 | iz = ia = ib = 0 |
|
215 | 215 | |
|
216 | 216 | for region in self.find_sync_regions(): |
|
217 | 217 | zmatch, zend, amatch, aend, bmatch, bend = region |
|
218 | 218 | #print 'match base [%d:%d]' % (zmatch, zend) |
|
219 | 219 | |
|
220 | 220 | matchlen = zend - zmatch |
|
221 | 221 | assert matchlen >= 0 |
|
222 | 222 | assert matchlen == (aend - amatch) |
|
223 | 223 | assert matchlen == (bend - bmatch) |
|
224 | 224 | |
|
225 | 225 | len_a = amatch - ia |
|
226 | 226 | len_b = bmatch - ib |
|
227 | 227 | len_base = zmatch - iz |
|
228 | 228 | assert len_a >= 0 |
|
229 | 229 | assert len_b >= 0 |
|
230 | 230 | assert len_base >= 0 |
|
231 | 231 | |
|
232 | 232 | #print 'unmatched a=%d, b=%d' % (len_a, len_b) |
|
233 | 233 | |
|
234 | 234 | if len_a or len_b: |
|
235 | 235 | # try to avoid actually slicing the lists |
|
236 | 236 | equal_a = compare_range(self.a, ia, amatch, |
|
237 | 237 | self.base, iz, zmatch) |
|
238 | 238 | equal_b = compare_range(self.b, ib, bmatch, |
|
239 | 239 | self.base, iz, zmatch) |
|
240 | 240 | same = compare_range(self.a, ia, amatch, |
|
241 | 241 | self.b, ib, bmatch) |
|
242 | 242 | |
|
243 | 243 | if same: |
|
244 | 244 | yield 'same', ia, amatch |
|
245 | 245 | elif equal_a and not equal_b: |
|
246 | 246 | yield 'b', ib, bmatch |
|
247 | 247 | elif equal_b and not equal_a: |
|
248 | 248 | yield 'a', ia, amatch |
|
249 | 249 | elif not equal_a and not equal_b: |
|
250 | 250 | yield 'conflict', iz, zmatch, ia, amatch, ib, bmatch |
|
251 | 251 | else: |
|
252 | 252 | raise AssertionError("can't handle a=b=base but unmatched") |
|
253 | 253 | |
|
254 | 254 | ia = amatch |
|
255 | 255 | ib = bmatch |
|
256 | 256 | iz = zmatch |
|
257 | 257 | |
|
258 | 258 | # if the same part of the base was deleted on both sides |
|
259 | 259 | # that's OK, we can just skip it. |
|
260 | 260 | |
|
261 | 261 | |
|
262 | 262 | if matchlen > 0: |
|
263 | 263 | assert ia == amatch |
|
264 | 264 | assert ib == bmatch |
|
265 | 265 | assert iz == zmatch |
|
266 | 266 | |
|
267 | 267 | yield 'unchanged', zmatch, zend |
|
268 | 268 | iz = zend |
|
269 | 269 | ia = aend |
|
270 | 270 | ib = bend |
|
271 | 271 | |
|
272 | 272 | def minimize(self, merge_regions): |
|
273 | 273 | """Trim conflict regions of lines where A and B sides match. |
|
274 | 274 | |
|
275 | 275 | Lines where both A and B have made the same changes at the beginning |
|
276 | 276 | or the end of each merge region are eliminated from the conflict |
|
277 | 277 | region and are instead considered the same. |
|
278 | 278 | """ |
|
279 | 279 | for region in merge_regions: |
|
280 | 280 | if region[0] != "conflict": |
|
281 | 281 | yield region |
|
282 | 282 | continue |
|
283 | 283 | issue, z1, z2, a1, a2, b1, b2 = region |
|
284 | 284 | alen = a2 - a1 |
|
285 | 285 | blen = b2 - b1 |
|
286 | 286 | |
|
287 | 287 | # find matches at the front |
|
288 | 288 | ii = 0 |
|
289 | 289 | while ii < alen and ii < blen and \ |
|
290 | 290 | self.a[a1 + ii] == self.b[b1 + ii]: |
|
291 | 291 | ii += 1 |
|
292 | 292 | startmatches = ii |
|
293 | 293 | |
|
294 | 294 | # find matches at the end |
|
295 | 295 | ii = 0 |
|
296 | 296 | while ii < alen and ii < blen and \ |
|
297 | 297 | self.a[a2 - ii - 1] == self.b[b2 - ii - 1]: |
|
298 | 298 | ii += 1 |
|
299 | 299 | endmatches = ii |
|
300 | 300 | |
|
301 | 301 | if startmatches > 0: |
|
302 | 302 | yield 'same', a1, a1 + startmatches |
|
303 | 303 | |
|
304 | 304 | yield ('conflict', z1, z2, |
|
305 | 305 | a1 + startmatches, a2 - endmatches, |
|
306 | 306 | b1 + startmatches, b2 - endmatches) |
|
307 | 307 | |
|
308 | 308 | if endmatches > 0: |
|
309 | 309 | yield 'same', a2 - endmatches, a2 |
|
310 | 310 | |
|
311 | 311 | def find_sync_regions(self): |
|
312 | 312 | """Return a list of sync regions, where both descendants match the base. |
|
313 | 313 | |
|
314 | 314 | Generates a list of (base1, base2, a1, a2, b1, b2). There is |
|
315 | 315 | always a zero-length sync region at the end of all the files. |
|
316 | 316 | """ |
|
317 | 317 | |
|
318 | 318 | ia = ib = 0 |
|
319 | 319 | amatches = mdiff.get_matching_blocks(self.basetext, self.atext) |
|
320 | 320 | bmatches = mdiff.get_matching_blocks(self.basetext, self.btext) |
|
321 | 321 | len_a = len(amatches) |
|
322 | 322 | len_b = len(bmatches) |
|
323 | 323 | |
|
324 | 324 | sl = [] |
|
325 | 325 | |
|
326 | 326 | while ia < len_a and ib < len_b: |
|
327 | 327 | abase, amatch, alen = amatches[ia] |
|
328 | 328 | bbase, bmatch, blen = bmatches[ib] |
|
329 | 329 | |
|
330 | 330 | # there is an unconflicted block at i; how long does it |
|
331 | 331 | # extend? until whichever one ends earlier. |
|
332 | 332 | i = intersect((abase, abase + alen), (bbase, bbase + blen)) |
|
333 | 333 | if i: |
|
334 | 334 | intbase = i[0] |
|
335 | 335 | intend = i[1] |
|
336 | 336 | intlen = intend - intbase |
|
337 | 337 | |
|
338 | 338 | # found a match of base[i[0], i[1]]; this may be less than |
|
339 | 339 | # the region that matches in either one |
|
340 | 340 | assert intlen <= alen |
|
341 | 341 | assert intlen <= blen |
|
342 | 342 | assert abase <= intbase |
|
343 | 343 | assert bbase <= intbase |
|
344 | 344 | |
|
345 | 345 | asub = amatch + (intbase - abase) |
|
346 | 346 | bsub = bmatch + (intbase - bbase) |
|
347 | 347 | aend = asub + intlen |
|
348 | 348 | bend = bsub + intlen |
|
349 | 349 | |
|
350 | 350 | assert self.base[intbase:intend] == self.a[asub:aend], \ |
|
351 | 351 | (self.base[intbase:intend], self.a[asub:aend]) |
|
352 | 352 | |
|
353 | 353 | assert self.base[intbase:intend] == self.b[bsub:bend] |
|
354 | 354 | |
|
355 | 355 | sl.append((intbase, intend, |
|
356 | 356 | asub, aend, |
|
357 | 357 | bsub, bend)) |
|
358 | 358 | |
|
359 | 359 | # advance whichever one ends first in the base text |
|
360 | 360 | if (abase + alen) < (bbase + blen): |
|
361 | 361 | ia += 1 |
|
362 | 362 | else: |
|
363 | 363 | ib += 1 |
|
364 | 364 | |
|
365 | 365 | intbase = len(self.base) |
|
366 | 366 | abase = len(self.a) |
|
367 | 367 | bbase = len(self.b) |
|
368 | 368 | sl.append((intbase, intbase, abase, abase, bbase, bbase)) |
|
369 | 369 | |
|
370 | 370 | return sl |
|
371 | 371 | |
|
372 | 372 | def find_unconflicted(self): |
|
373 | 373 | """Return a list of ranges in base that are not conflicted.""" |
|
374 | 374 | am = mdiff.get_matching_blocks(self.basetext, self.atext) |
|
375 | 375 | bm = mdiff.get_matching_blocks(self.basetext, self.btext) |
|
376 | 376 | |
|
377 | 377 | unc = [] |
|
378 | 378 | |
|
379 | 379 | while am and bm: |
|
380 | 380 | # there is an unconflicted block at i; how long does it |
|
381 | 381 | # extend? until whichever one ends earlier. |
|
382 | 382 | a1 = am[0][0] |
|
383 | 383 | a2 = a1 + am[0][2] |
|
384 | 384 | b1 = bm[0][0] |
|
385 | 385 | b2 = b1 + bm[0][2] |
|
386 | 386 | i = intersect((a1, a2), (b1, b2)) |
|
387 | 387 | if i: |
|
388 | 388 | unc.append(i) |
|
389 | 389 | |
|
390 | 390 | if a2 < b2: |
|
391 | 391 | del am[0] |
|
392 | 392 | else: |
|
393 | 393 | del bm[0] |
|
394 | 394 | |
|
395 | 395 | return unc |
|
396 | 396 | |
|
397 | 397 | def _verifytext(text, path, ui, opts): |
|
398 | 398 | """verifies that text is non-binary (unless opts[text] is passed, |
|
399 | 399 | then we just warn)""" |
|
400 | 400 | if util.binary(text): |
|
401 | 401 | msg = _("%s looks like a binary file.") % path |
|
402 | 402 | if not opts.get('quiet'): |
|
403 | 403 | ui.warn(_('warning: %s\n') % msg) |
|
404 | 404 | if not opts.get('text'): |
|
405 | 405 | raise error.Abort(msg) |
|
406 | 406 | return text |
|
407 | 407 | |
|
408 | 408 | def _picklabels(defaults, overrides): |
|
409 | name_a, name_b, name_base = defaults | |
|
410 | ||
|
411 | if len(overrides) > 0: | |
|
412 | name_a = overrides[0] | |
|
413 | if len(overrides) > 1: | |
|
414 | name_b = overrides[1] | |
|
415 | if len(overrides) > 2: | |
|
416 | name_base = overrides[2] | |
|
417 | 409 | if len(overrides) > 3: |
|
418 | 410 | raise error.Abort(_("can only specify three labels.")) |
|
419 | ||
|
420 | return [name_a, name_b, name_base] | |
|
411 | result = defaults[:] | |
|
412 | for i, override in enumerate(overrides): | |
|
413 | result[i] = override | |
|
414 | return result | |
|
421 | 415 | |
|
422 | 416 | def simplemerge(ui, localctx, basectx, otherctx, repo=None, **opts): |
|
423 | 417 | """Performs the simplemerge algorithm. |
|
424 | 418 | |
|
425 | 419 | The merged result is written into `localctx`. |
|
426 | 420 | """ |
|
427 | 421 | def readctx(ctx): |
|
428 | 422 | # Merges were always run in the working copy before, which means |
|
429 | 423 | # they used decoded data, if the user defined any repository |
|
430 | 424 | # filters. |
|
431 | 425 | # |
|
432 | 426 | # Maintain that behavior today for BC, though perhaps in the future |
|
433 | 427 | # it'd be worth considering whether merging encoded data (what the |
|
434 | 428 | # repository usually sees) might be more useful. |
|
435 | 429 | return _verifytext(ctx.decodeddata(), ctx.path(), ui, opts) |
|
436 | 430 | |
|
437 | 431 | mode = opts.get('mode','merge') |
|
438 | 432 | name_a, name_b, name_base = None, None, None |
|
439 | 433 | if mode != 'union': |
|
440 | 434 | name_a, name_b, name_base = _picklabels([localctx.path(), |
|
441 | 435 | otherctx.path(), None], |
|
442 | 436 | opts.get('label', [])) |
|
443 | 437 | |
|
444 | 438 | try: |
|
445 | 439 | localtext = readctx(localctx) |
|
446 | 440 | basetext = readctx(basectx) |
|
447 | 441 | othertext = readctx(otherctx) |
|
448 | 442 | except error.Abort: |
|
449 | 443 | return 1 |
|
450 | 444 | |
|
451 | 445 | m3 = Merge3Text(basetext, localtext, othertext) |
|
452 | 446 | extrakwargs = { |
|
453 | 447 | "localorother": opts.get("localorother", None), |
|
454 | 448 | 'minimize': True, |
|
455 | 449 | } |
|
456 | 450 | if mode == 'union': |
|
457 | 451 | extrakwargs['start_marker'] = None |
|
458 | 452 | extrakwargs['mid_marker'] = None |
|
459 | 453 | extrakwargs['end_marker'] = None |
|
460 | 454 | elif name_base is not None: |
|
461 | 455 | extrakwargs['base_marker'] = '|||||||' |
|
462 | 456 | extrakwargs['name_base'] = name_base |
|
463 | 457 | extrakwargs['minimize'] = False |
|
464 | 458 | |
|
465 | 459 | mergedtext = "" |
|
466 | 460 | for line in m3.merge_lines(name_a=name_a, name_b=name_b, |
|
467 | 461 | **pycompat.strkwargs(extrakwargs)): |
|
468 | 462 | if opts.get('print'): |
|
469 | 463 | ui.fout.write(line) |
|
470 | 464 | else: |
|
471 | 465 | mergedtext += line |
|
472 | 466 | |
|
473 | 467 | if not opts.get('print'): |
|
474 | 468 | localctx.write(mergedtext, localctx.flags()) |
|
475 | 469 | |
|
476 | 470 | if m3.conflicts and not mode == 'union': |
|
477 | 471 | return 1 |
General Comments 0
You need to be logged in to leave comments.
Login now