Show More
| @@ -1,1112 +1,1130 b'' | |||
|
|
1 | 1 | # copies.py - copy detection for Mercurial |
|
|
2 | 2 | # |
|
|
3 | 3 | # Copyright 2008 Matt Mackall <mpm@selenic.com> |
|
|
4 | 4 | # |
|
|
5 | 5 | # This software may be used and distributed according to the terms of the |
|
|
6 | 6 | # GNU General Public License version 2 or any later version. |
|
|
7 | 7 | |
|
|
8 | 8 | from __future__ import absolute_import |
|
|
9 | 9 | |
|
|
10 | 10 | import collections |
|
|
11 | 11 | import os |
|
|
12 | 12 | |
|
|
13 | 13 | from .i18n import _ |
|
|
14 | 14 | |
|
|
15 | 15 | |
|
|
16 | 16 | from . import ( |
|
|
17 | 17 | match as matchmod, |
|
|
18 | 18 | node, |
|
|
19 | 19 | pathutil, |
|
|
20 | 20 | pycompat, |
|
|
21 | 21 | util, |
|
|
22 | 22 | ) |
|
|
23 | 23 | |
|
|
24 | 24 | |
|
|
25 | 25 | from .utils import stringutil |
|
|
26 | 26 | |
|
|
27 | 27 | from .revlogutils import flagutil |
|
|
28 | 28 | |
|
|
29 | 29 | |
|
|
30 | 30 | def _filter(src, dst, t): |
|
|
31 | 31 | """filters out invalid copies after chaining""" |
|
|
32 | 32 | |
|
|
33 | 33 | # When _chain()'ing copies in 'a' (from 'src' via some other commit 'mid') |
|
|
34 | 34 | # with copies in 'b' (from 'mid' to 'dst'), we can get the different cases |
|
|
35 | 35 | # in the following table (not including trivial cases). For example, case 2 |
|
|
36 | 36 | # is where a file existed in 'src' and remained under that name in 'mid' and |
|
|
37 | 37 | # then was renamed between 'mid' and 'dst'. |
|
|
38 | 38 | # |
|
|
39 | 39 | # case src mid dst result |
|
|
40 | 40 | # 1 x y - - |
|
|
41 | 41 | # 2 x y y x->y |
|
|
42 | 42 | # 3 x y x - |
|
|
43 | 43 | # 4 x y z x->z |
|
|
44 | 44 | # 5 - x y - |
|
|
45 | 45 | # 6 x x y x->y |
|
|
46 | 46 | # |
|
|
47 | 47 | # _chain() takes care of chaining the copies in 'a' and 'b', but it |
|
|
48 | 48 | # cannot tell the difference between cases 1 and 2, between 3 and 4, or |
|
|
49 | 49 | # between 5 and 6, so it includes all cases in its result. |
|
|
50 | 50 | # Cases 1, 3, and 5 are then removed by _filter(). |
|
|
51 | 51 | |
|
|
52 | 52 | for k, v in list(t.items()): |
|
|
53 | 53 | # remove copies from files that didn't exist |
|
|
54 | 54 | if v not in src: |
|
|
55 | 55 | del t[k] |
|
|
56 | 56 | # remove criss-crossed copies |
|
|
57 | 57 | elif k in src and v in dst: |
|
|
58 | 58 | del t[k] |
|
|
59 | 59 | # remove copies to files that were then removed |
|
|
60 | 60 | elif k not in dst: |
|
|
61 | 61 | del t[k] |
|
|
62 | 62 | |
|
|
63 | 63 | |
|
|
64 | 64 | def _chain(prefix, suffix): |
|
|
65 | 65 | """chain two sets of copies 'prefix' and 'suffix'""" |
|
|
66 | 66 | result = prefix.copy() |
|
|
67 | 67 | for key, value in pycompat.iteritems(suffix): |
|
|
68 | 68 | result[key] = prefix.get(value, value) |
|
|
69 | 69 | return result |
|
|
70 | 70 | |
|
|
71 | 71 | |
|
|
72 | 72 | def _tracefile(fctx, am, basemf): |
|
|
73 | 73 | """return file context that is the ancestor of fctx present in ancestor |
|
|
74 | 74 | manifest am |
|
|
75 | 75 | |
|
|
76 | 76 | Note: we used to try and stop after a given limit, however checking if that |
|
|
77 | 77 | limit is reached turned out to be very expensive. we are better off |
|
|
78 | 78 | disabling that feature.""" |
|
|
79 | 79 | |
|
|
80 | 80 | for f in fctx.ancestors(): |
|
|
81 | 81 | path = f.path() |
|
|
82 | 82 | if am.get(path, None) == f.filenode(): |
|
|
83 | 83 | return path |
|
|
84 | 84 | if basemf and basemf.get(path, None) == f.filenode(): |
|
|
85 | 85 | return path |
|
|
86 | 86 | |
|
|
87 | 87 | |
|
|
88 | 88 | def _dirstatecopies(repo, match=None): |
|
|
89 | 89 | ds = repo.dirstate |
|
|
90 | 90 | c = ds.copies().copy() |
|
|
91 | 91 | for k in list(c): |
|
|
92 | 92 | if ds[k] not in b'anm' or (match and not match(k)): |
|
|
93 | 93 | del c[k] |
|
|
94 | 94 | return c |
|
|
95 | 95 | |
|
|
96 | 96 | |
|
|
97 | 97 | def _computeforwardmissing(a, b, match=None): |
|
|
98 | 98 | """Computes which files are in b but not a. |
|
|
99 | 99 | This is its own function so extensions can easily wrap this call to see what |
|
|
100 | 100 | files _forwardcopies is about to process. |
|
|
101 | 101 | """ |
|
|
102 | 102 | ma = a.manifest() |
|
|
103 | 103 | mb = b.manifest() |
|
|
104 | 104 | return mb.filesnotin(ma, match=match) |
|
|
105 | 105 | |
|
|
106 | 106 | |
|
|
107 | 107 | def usechangesetcentricalgo(repo): |
|
|
108 | 108 | """Checks if we should use changeset-centric copy algorithms""" |
|
|
109 | 109 | if repo.filecopiesmode == b'changeset-sidedata': |
|
|
110 | 110 | return True |
|
|
111 | 111 | readfrom = repo.ui.config(b'experimental', b'copies.read-from') |
|
|
112 | 112 | changesetsource = (b'changeset-only', b'compatibility') |
|
|
113 | 113 | return readfrom in changesetsource |
|
|
114 | 114 | |
|
|
115 | 115 | |
|
|
116 | 116 | def _committedforwardcopies(a, b, base, match): |
|
|
117 | 117 | """Like _forwardcopies(), but b.rev() cannot be None (working copy)""" |
|
|
118 | 118 | # files might have to be traced back to the fctx parent of the last |
|
|
119 | 119 | # one-side-only changeset, but not further back than that |
|
|
120 | 120 | repo = a._repo |
|
|
121 | 121 | |
|
|
122 | 122 | if usechangesetcentricalgo(repo): |
|
|
123 | 123 | return _changesetforwardcopies(a, b, match) |
|
|
124 | 124 | |
|
|
125 | 125 | debug = repo.ui.debugflag and repo.ui.configbool(b'devel', b'debug.copies') |
|
|
126 | 126 | dbg = repo.ui.debug |
|
|
127 | 127 | if debug: |
|
|
128 | 128 | dbg(b'debug.copies: looking into rename from %s to %s\n' % (a, b)) |
|
|
129 | 129 | am = a.manifest() |
|
|
130 | 130 | basemf = None if base is None else base.manifest() |
|
|
131 | 131 | |
|
|
132 | 132 | # find where new files came from |
|
|
133 | 133 | # we currently don't try to find where old files went, too expensive |
|
|
134 | 134 | # this means we can miss a case like 'hg rm b; hg cp a b' |
|
|
135 | 135 | cm = {} |
|
|
136 | 136 | |
|
|
137 | 137 | # Computing the forward missing is quite expensive on large manifests, since |
|
|
138 | 138 | # it compares the entire manifests. We can optimize it in the common use |
|
|
139 | 139 | # case of computing what copies are in a commit versus its parent (like |
|
|
140 | 140 | # during a rebase or histedit). Note, we exclude merge commits from this |
|
|
141 | 141 | # optimization, since the ctx.files() for a merge commit is not correct for |
|
|
142 | 142 | # this comparison. |
|
|
143 | 143 | forwardmissingmatch = match |
|
|
144 | 144 | if b.p1() == a and b.p2().node() == node.nullid: |
|
|
145 | 145 | filesmatcher = matchmod.exact(b.files()) |
|
|
146 | 146 | forwardmissingmatch = matchmod.intersectmatchers(match, filesmatcher) |
|
|
147 | 147 | missing = _computeforwardmissing(a, b, match=forwardmissingmatch) |
|
|
148 | 148 | |
|
|
149 | 149 | ancestrycontext = a._repo.changelog.ancestors([b.rev()], inclusive=True) |
|
|
150 | 150 | |
|
|
151 | 151 | if debug: |
|
|
152 | 152 | dbg(b'debug.copies: missing files to search: %d\n' % len(missing)) |
|
|
153 | 153 | |
|
|
154 | 154 | for f in sorted(missing): |
|
|
155 | 155 | if debug: |
|
|
156 | 156 | dbg(b'debug.copies: tracing file: %s\n' % f) |
|
|
157 | 157 | fctx = b[f] |
|
|
158 | 158 | fctx._ancestrycontext = ancestrycontext |
|
|
159 | 159 | |
|
|
160 | 160 | if debug: |
|
|
161 | 161 | start = util.timer() |
|
|
162 | 162 | opath = _tracefile(fctx, am, basemf) |
|
|
163 | 163 | if opath: |
|
|
164 | 164 | if debug: |
|
|
165 | 165 | dbg(b'debug.copies: rename of: %s\n' % opath) |
|
|
166 | 166 | cm[f] = opath |
|
|
167 | 167 | if debug: |
|
|
168 | 168 | dbg( |
|
|
169 | 169 | b'debug.copies: time: %f seconds\n' |
|
|
170 | 170 | % (util.timer() - start) |
|
|
171 | 171 | ) |
|
|
172 | 172 | return cm |
|
|
173 | 173 | |
|
|
174 | 174 | |
|
|
175 | 175 | def _revinfo_getter(repo): |
|
|
176 | 176 | """returns a function that returns the following data given a <rev>" |
|
|
177 | 177 | |
|
|
178 | 178 | * p1: revision number of first parent |
|
|
179 | 179 | * p2: revision number of first parent |
|
|
180 | 180 | * changes: a ChangingFiles object |
|
|
181 | 181 | """ |
|
|
182 | 182 | cl = repo.changelog |
|
|
183 | 183 | parents = cl.parentrevs |
|
|
184 | 184 | flags = cl.flags |
|
|
185 | 185 | |
|
|
186 | 186 | HASCOPIESINFO = flagutil.REVIDX_HASCOPIESINFO |
|
|
187 | 187 | |
|
|
188 | 188 | changelogrevision = cl.changelogrevision |
|
|
189 | 189 | |
|
|
190 | 190 | # A small cache to avoid doing the work twice for merges |
|
|
191 | 191 | # |
|
|
192 | 192 | # In the vast majority of cases, if we ask information for a revision |
|
|
193 | 193 | # about 1 parent, we'll later ask it for the other. So it make sense to |
|
|
194 | 194 | # keep the information around when reaching the first parent of a merge |
|
|
195 | 195 | # and dropping it after it was provided for the second parents. |
|
|
196 | 196 | # |
|
|
197 | 197 | # It exists cases were only one parent of the merge will be walked. It |
|
|
198 | 198 | # happens when the "destination" the copy tracing is descendant from a |
|
|
199 | 199 | # new root, not common with the "source". In that case, we will only walk |
|
|
200 | 200 | # through merge parents that are descendant of changesets common |
|
|
201 | 201 | # between "source" and "destination". |
|
|
202 | 202 | # |
|
|
203 | 203 | # With the current case implementation if such changesets have a copy |
|
|
204 | 204 | # information, we'll keep them in memory until the end of |
|
|
205 | 205 | # _changesetforwardcopies. We don't expect the case to be frequent |
|
|
206 | 206 | # enough to matters. |
|
|
207 | 207 | # |
|
|
208 | 208 | # In addition, it would be possible to reach pathological case, were |
|
|
209 | 209 | # many first parent are met before any second parent is reached. In |
|
|
210 | 210 | # that case the cache could grow. If this even become an issue one can |
|
|
211 | 211 | # safely introduce a maximum cache size. This would trade extra CPU/IO |
|
|
212 | 212 | # time to save memory. |
|
|
213 | 213 | merge_caches = {} |
|
|
214 | 214 | |
|
|
215 | 215 | def revinfo(rev): |
|
|
216 | 216 | p1, p2 = parents(rev) |
|
|
217 | 217 | value = None |
|
|
218 | 218 | e = merge_caches.pop(rev, None) |
|
|
219 | 219 | if e is not None: |
|
|
220 | 220 | return e |
|
|
221 | 221 | changes = None |
|
|
222 | 222 | if flags(rev) & HASCOPIESINFO: |
|
|
223 | 223 | changes = changelogrevision(rev).changes |
|
|
224 | 224 | value = (p1, p2, changes) |
|
|
225 | 225 | if p1 != node.nullrev and p2 != node.nullrev: |
|
|
226 | 226 | # XXX some case we over cache, IGNORE |
|
|
227 | 227 | merge_caches[rev] = value |
|
|
228 | 228 | return value |
|
|
229 | 229 | |
|
|
230 | 230 | return revinfo |
|
|
231 | 231 | |
|
|
232 | 232 | |
|
|
233 | def cached_is_ancestor(is_ancestor): | |
|
|
234 | """return a cached version of is_ancestor""" | |
|
|
235 | cache = {} | |
|
|
236 | ||
|
|
237 | def _is_ancestor(anc, desc): | |
|
|
238 | if anc > desc: | |
|
|
239 | return False | |
|
|
240 | elif anc == desc: | |
|
|
241 | return True | |
|
|
242 | key = (anc, desc) | |
|
|
243 | ret = cache.get(key) | |
|
|
244 | if ret is None: | |
|
|
245 | ret = cache[key] = is_ancestor(anc, desc) | |
|
|
246 | return ret | |
|
|
247 | ||
|
|
248 | return _is_ancestor | |
|
|
249 | ||
|
|
250 | ||
|
|
233 | 251 | def _changesetforwardcopies(a, b, match): |
|
|
234 | 252 | if a.rev() in (node.nullrev, b.rev()): |
|
|
235 | 253 | return {} |
|
|
236 | 254 | |
|
|
237 | 255 | repo = a.repo().unfiltered() |
|
|
238 | 256 | children = {} |
|
|
239 | 257 | |
|
|
240 | 258 | cl = repo.changelog |
|
|
241 | isancestor = cl.isancestorrev # XXX we should had chaching to this. | |
|
|
259 | isancestor = cached_is_ancestor(cl.isancestorrev) | |
|
|
242 | 260 | missingrevs = cl.findmissingrevs(common=[a.rev()], heads=[b.rev()]) |
|
|
243 | 261 | mrset = set(missingrevs) |
|
|
244 | 262 | roots = set() |
|
|
245 | 263 | for r in missingrevs: |
|
|
246 | 264 | for p in cl.parentrevs(r): |
|
|
247 | 265 | if p == node.nullrev: |
|
|
248 | 266 | continue |
|
|
249 | 267 | if p not in children: |
|
|
250 | 268 | children[p] = [r] |
|
|
251 | 269 | else: |
|
|
252 | 270 | children[p].append(r) |
|
|
253 | 271 | if p not in mrset: |
|
|
254 | 272 | roots.add(p) |
|
|
255 | 273 | if not roots: |
|
|
256 | 274 | # no common revision to track copies from |
|
|
257 | 275 | return {} |
|
|
258 | 276 | min_root = min(roots) |
|
|
259 | 277 | |
|
|
260 | 278 | from_head = set( |
|
|
261 | 279 | cl.reachableroots(min_root, [b.rev()], list(roots), includepath=True) |
|
|
262 | 280 | ) |
|
|
263 | 281 | |
|
|
264 | 282 | iterrevs = set(from_head) |
|
|
265 | 283 | iterrevs &= mrset |
|
|
266 | 284 | iterrevs.update(roots) |
|
|
267 | 285 | iterrevs.remove(b.rev()) |
|
|
268 | 286 | revs = sorted(iterrevs) |
|
|
269 | 287 | |
|
|
270 | 288 | if repo.filecopiesmode == b'changeset-sidedata': |
|
|
271 | 289 | revinfo = _revinfo_getter(repo) |
|
|
272 | 290 | return _combine_changeset_copies( |
|
|
273 | 291 | revs, children, b.rev(), revinfo, match, isancestor |
|
|
274 | 292 | ) |
|
|
275 | 293 | else: |
|
|
276 | 294 | revinfo = _revinfo_getter_extra(repo) |
|
|
277 | 295 | return _combine_changeset_copies_extra( |
|
|
278 | 296 | revs, children, b.rev(), revinfo, match, isancestor |
|
|
279 | 297 | ) |
|
|
280 | 298 | |
|
|
281 | 299 | |
|
|
282 | 300 | def _combine_changeset_copies( |
|
|
283 | 301 | revs, children, targetrev, revinfo, match, isancestor |
|
|
284 | 302 | ): |
|
|
285 | 303 | """combine the copies information for each item of iterrevs |
|
|
286 | 304 | |
|
|
287 | 305 | revs: sorted iterable of revision to visit |
|
|
288 | 306 | children: a {parent: [children]} mapping. |
|
|
289 | 307 | targetrev: the final copies destination revision (not in iterrevs) |
|
|
290 | 308 | revinfo(rev): a function that return (p1, p2, p1copies, p2copies, removed) |
|
|
291 | 309 | match: a matcher |
|
|
292 | 310 | |
|
|
293 | 311 | It returns the aggregated copies information for `targetrev`. |
|
|
294 | 312 | """ |
|
|
295 | 313 | all_copies = {} |
|
|
296 | 314 | alwaysmatch = match.always() |
|
|
297 | 315 | for r in revs: |
|
|
298 | 316 | copies = all_copies.pop(r, None) |
|
|
299 | 317 | if copies is None: |
|
|
300 | 318 | # this is a root |
|
|
301 | 319 | copies = {} |
|
|
302 | 320 | for i, c in enumerate(children[r]): |
|
|
303 | 321 | p1, p2, changes = revinfo(c) |
|
|
304 | 322 | childcopies = {} |
|
|
305 | 323 | if r == p1: |
|
|
306 | 324 | parent = 1 |
|
|
307 | 325 | if changes is not None: |
|
|
308 | 326 | childcopies = changes.copied_from_p1 |
|
|
309 | 327 | else: |
|
|
310 | 328 | assert r == p2 |
|
|
311 | 329 | parent = 2 |
|
|
312 | 330 | if changes is not None: |
|
|
313 | 331 | childcopies = changes.copied_from_p2 |
|
|
314 | 332 | if not alwaysmatch: |
|
|
315 | 333 | childcopies = { |
|
|
316 | 334 | dst: src for dst, src in childcopies.items() if match(dst) |
|
|
317 | 335 | } |
|
|
318 | 336 | newcopies = copies |
|
|
319 | 337 | if childcopies: |
|
|
320 | 338 | newcopies = copies.copy() |
|
|
321 | 339 | for dest, source in pycompat.iteritems(childcopies): |
|
|
322 | 340 | prev = copies.get(source) |
|
|
323 | 341 | if prev is not None and prev[1] is not None: |
|
|
324 | 342 | source = prev[1] |
|
|
325 | 343 | newcopies[dest] = (c, source) |
|
|
326 | 344 | assert newcopies is not copies |
|
|
327 | 345 | if changes is not None: |
|
|
328 | 346 | for f in changes.removed: |
|
|
329 | 347 | if f in newcopies: |
|
|
330 | 348 | if newcopies is copies: |
|
|
331 | 349 | # copy on write to avoid affecting potential other |
|
|
332 | 350 | # branches. when there are no other branches, this |
|
|
333 | 351 | # could be avoided. |
|
|
334 | 352 | newcopies = copies.copy() |
|
|
335 | 353 | newcopies[f] = (c, None) |
|
|
336 | 354 | othercopies = all_copies.get(c) |
|
|
337 | 355 | if othercopies is None: |
|
|
338 | 356 | all_copies[c] = newcopies |
|
|
339 | 357 | else: |
|
|
340 | 358 | # we are the second parent to work on c, we need to merge our |
|
|
341 | 359 | # work with the other. |
|
|
342 | 360 | # |
|
|
343 | 361 | # In case of conflict, parent 1 take precedence over parent 2. |
|
|
344 | 362 | # This is an arbitrary choice made anew when implementing |
|
|
345 | 363 | # changeset based copies. It was made without regards with |
|
|
346 | 364 | # potential filelog related behavior. |
|
|
347 | 365 | if parent == 1: |
|
|
348 | 366 | _merge_copies_dict( |
|
|
349 | 367 | othercopies, newcopies, isancestor, changes |
|
|
350 | 368 | ) |
|
|
351 | 369 | else: |
|
|
352 | 370 | _merge_copies_dict( |
|
|
353 | 371 | newcopies, othercopies, isancestor, changes |
|
|
354 | 372 | ) |
|
|
355 | 373 | all_copies[c] = newcopies |
|
|
356 | 374 | |
|
|
357 | 375 | final_copies = {} |
|
|
358 | 376 | for dest, (tt, source) in all_copies[targetrev].items(): |
|
|
359 | 377 | if source is not None: |
|
|
360 | 378 | final_copies[dest] = source |
|
|
361 | 379 | return final_copies |
|
|
362 | 380 | |
|
|
363 | 381 | |
|
|
364 | 382 | def _merge_copies_dict(minor, major, isancestor, changes): |
|
|
365 | 383 | """merge two copies-mapping together, minor and major |
|
|
366 | 384 | |
|
|
367 | 385 | In case of conflict, value from "major" will be picked. |
|
|
368 | 386 | |
|
|
369 | 387 | - `isancestors(low_rev, high_rev)`: callable return True if `low_rev` is an |
|
|
370 | 388 | ancestors of `high_rev`, |
|
|
371 | 389 | |
|
|
372 | 390 | - `ismerged(path)`: callable return True if `path` have been merged in the |
|
|
373 | 391 | current revision, |
|
|
374 | 392 | """ |
|
|
375 | 393 | for dest, value in major.items(): |
|
|
376 | 394 | other = minor.get(dest) |
|
|
377 | 395 | if other is None: |
|
|
378 | 396 | minor[dest] = value |
|
|
379 | 397 | else: |
|
|
380 | 398 | new_tt = value[0] |
|
|
381 | 399 | other_tt = other[0] |
|
|
382 | 400 | if value[1] == other[1]: |
|
|
383 | 401 | continue |
|
|
384 | 402 | # content from "major" wins, unless it is older |
|
|
385 | 403 | # than the branch point or there is a merge |
|
|
386 | 404 | if new_tt == other_tt: |
|
|
387 | 405 | minor[dest] = value |
|
|
388 | 406 | elif ( |
|
|
389 | 407 | changes is not None |
|
|
390 | 408 | and value[1] is None |
|
|
391 | 409 | and dest in changes.salvaged |
|
|
392 | 410 | ): |
|
|
393 | 411 | pass |
|
|
394 | 412 | elif ( |
|
|
395 | 413 | changes is not None |
|
|
396 | 414 | and other[1] is None |
|
|
397 | 415 | and dest in changes.salvaged |
|
|
398 | 416 | ): |
|
|
399 | 417 | minor[dest] = value |
|
|
400 | 418 | elif changes is not None and dest in changes.merged: |
|
|
401 | 419 | minor[dest] = value |
|
|
402 | 420 | elif not isancestor(new_tt, other_tt): |
|
|
403 | 421 | if value[1] is not None: |
|
|
404 | 422 | minor[dest] = value |
|
|
405 | 423 | elif isancestor(other_tt, new_tt): |
|
|
406 | 424 | minor[dest] = value |
|
|
407 | 425 | |
|
|
408 | 426 | |
|
|
409 | 427 | def _revinfo_getter_extra(repo): |
|
|
410 | 428 | """return a function that return multiple data given a <rev>"i |
|
|
411 | 429 | |
|
|
412 | 430 | * p1: revision number of first parent |
|
|
413 | 431 | * p2: revision number of first parent |
|
|
414 | 432 | * p1copies: mapping of copies from p1 |
|
|
415 | 433 | * p2copies: mapping of copies from p2 |
|
|
416 | 434 | * removed: a list of removed files |
|
|
417 | 435 | * ismerged: a callback to know if file was merged in that revision |
|
|
418 | 436 | """ |
|
|
419 | 437 | cl = repo.changelog |
|
|
420 | 438 | parents = cl.parentrevs |
|
|
421 | 439 | |
|
|
422 | 440 | def get_ismerged(rev): |
|
|
423 | 441 | ctx = repo[rev] |
|
|
424 | 442 | |
|
|
425 | 443 | def ismerged(path): |
|
|
426 | 444 | if path not in ctx.files(): |
|
|
427 | 445 | return False |
|
|
428 | 446 | fctx = ctx[path] |
|
|
429 | 447 | parents = fctx._filelog.parents(fctx._filenode) |
|
|
430 | 448 | nb_parents = 0 |
|
|
431 | 449 | for n in parents: |
|
|
432 | 450 | if n != node.nullid: |
|
|
433 | 451 | nb_parents += 1 |
|
|
434 | 452 | return nb_parents >= 2 |
|
|
435 | 453 | |
|
|
436 | 454 | return ismerged |
|
|
437 | 455 | |
|
|
438 | 456 | def revinfo(rev): |
|
|
439 | 457 | p1, p2 = parents(rev) |
|
|
440 | 458 | ctx = repo[rev] |
|
|
441 | 459 | p1copies, p2copies = ctx._copies |
|
|
442 | 460 | removed = ctx.filesremoved() |
|
|
443 | 461 | return p1, p2, p1copies, p2copies, removed, get_ismerged(rev) |
|
|
444 | 462 | |
|
|
445 | 463 | return revinfo |
|
|
446 | 464 | |
|
|
447 | 465 | |
|
|
448 | 466 | def _combine_changeset_copies_extra( |
|
|
449 | 467 | revs, children, targetrev, revinfo, match, isancestor |
|
|
450 | 468 | ): |
|
|
451 | 469 | """version of `_combine_changeset_copies` that works with the Google |
|
|
452 | 470 | specific "extra" based storage for copy information""" |
|
|
453 | 471 | all_copies = {} |
|
|
454 | 472 | alwaysmatch = match.always() |
|
|
455 | 473 | for r in revs: |
|
|
456 | 474 | copies = all_copies.pop(r, None) |
|
|
457 | 475 | if copies is None: |
|
|
458 | 476 | # this is a root |
|
|
459 | 477 | copies = {} |
|
|
460 | 478 | for i, c in enumerate(children[r]): |
|
|
461 | 479 | p1, p2, p1copies, p2copies, removed, ismerged = revinfo(c) |
|
|
462 | 480 | if r == p1: |
|
|
463 | 481 | parent = 1 |
|
|
464 | 482 | childcopies = p1copies |
|
|
465 | 483 | else: |
|
|
466 | 484 | assert r == p2 |
|
|
467 | 485 | parent = 2 |
|
|
468 | 486 | childcopies = p2copies |
|
|
469 | 487 | if not alwaysmatch: |
|
|
470 | 488 | childcopies = { |
|
|
471 | 489 | dst: src for dst, src in childcopies.items() if match(dst) |
|
|
472 | 490 | } |
|
|
473 | 491 | newcopies = copies |
|
|
474 | 492 | if childcopies: |
|
|
475 | 493 | newcopies = copies.copy() |
|
|
476 | 494 | for dest, source in pycompat.iteritems(childcopies): |
|
|
477 | 495 | prev = copies.get(source) |
|
|
478 | 496 | if prev is not None and prev[1] is not None: |
|
|
479 | 497 | source = prev[1] |
|
|
480 | 498 | newcopies[dest] = (c, source) |
|
|
481 | 499 | assert newcopies is not copies |
|
|
482 | 500 | for f in removed: |
|
|
483 | 501 | if f in newcopies: |
|
|
484 | 502 | if newcopies is copies: |
|
|
485 | 503 | # copy on write to avoid affecting potential other |
|
|
486 | 504 | # branches. when there are no other branches, this |
|
|
487 | 505 | # could be avoided. |
|
|
488 | 506 | newcopies = copies.copy() |
|
|
489 | 507 | newcopies[f] = (c, None) |
|
|
490 | 508 | othercopies = all_copies.get(c) |
|
|
491 | 509 | if othercopies is None: |
|
|
492 | 510 | all_copies[c] = newcopies |
|
|
493 | 511 | else: |
|
|
494 | 512 | # we are the second parent to work on c, we need to merge our |
|
|
495 | 513 | # work with the other. |
|
|
496 | 514 | # |
|
|
497 | 515 | # In case of conflict, parent 1 take precedence over parent 2. |
|
|
498 | 516 | # This is an arbitrary choice made anew when implementing |
|
|
499 | 517 | # changeset based copies. It was made without regards with |
|
|
500 | 518 | # potential filelog related behavior. |
|
|
501 | 519 | if parent == 1: |
|
|
502 | 520 | _merge_copies_dict_extra( |
|
|
503 | 521 | othercopies, newcopies, isancestor, ismerged |
|
|
504 | 522 | ) |
|
|
505 | 523 | else: |
|
|
506 | 524 | _merge_copies_dict_extra( |
|
|
507 | 525 | newcopies, othercopies, isancestor, ismerged |
|
|
508 | 526 | ) |
|
|
509 | 527 | all_copies[c] = newcopies |
|
|
510 | 528 | |
|
|
511 | 529 | final_copies = {} |
|
|
512 | 530 | for dest, (tt, source) in all_copies[targetrev].items(): |
|
|
513 | 531 | if source is not None: |
|
|
514 | 532 | final_copies[dest] = source |
|
|
515 | 533 | return final_copies |
|
|
516 | 534 | |
|
|
517 | 535 | |
|
|
518 | 536 | def _merge_copies_dict_extra(minor, major, isancestor, ismerged): |
|
|
519 | 537 | """version of `_merge_copies_dict` that works with the Google |
|
|
520 | 538 | specific "extra" based storage for copy information""" |
|
|
521 | 539 | for dest, value in major.items(): |
|
|
522 | 540 | other = minor.get(dest) |
|
|
523 | 541 | if other is None: |
|
|
524 | 542 | minor[dest] = value |
|
|
525 | 543 | else: |
|
|
526 | 544 | new_tt = value[0] |
|
|
527 | 545 | other_tt = other[0] |
|
|
528 | 546 | if value[1] == other[1]: |
|
|
529 | 547 | continue |
|
|
530 | 548 | # content from "major" wins, unless it is older |
|
|
531 | 549 | # than the branch point or there is a merge |
|
|
532 | 550 | if ( |
|
|
533 | 551 | new_tt == other_tt |
|
|
534 | 552 | or not isancestor(new_tt, other_tt) |
|
|
535 | 553 | or ismerged(dest) |
|
|
536 | 554 | ): |
|
|
537 | 555 | minor[dest] = value |
|
|
538 | 556 | |
|
|
539 | 557 | |
|
|
540 | 558 | def _forwardcopies(a, b, base=None, match=None): |
|
|
541 | 559 | """find {dst@b: src@a} copy mapping where a is an ancestor of b""" |
|
|
542 | 560 | |
|
|
543 | 561 | if base is None: |
|
|
544 | 562 | base = a |
|
|
545 | 563 | match = a.repo().narrowmatch(match) |
|
|
546 | 564 | # check for working copy |
|
|
547 | 565 | if b.rev() is None: |
|
|
548 | 566 | cm = _committedforwardcopies(a, b.p1(), base, match) |
|
|
549 | 567 | # combine copies from dirstate if necessary |
|
|
550 | 568 | copies = _chain(cm, _dirstatecopies(b._repo, match)) |
|
|
551 | 569 | else: |
|
|
552 | 570 | copies = _committedforwardcopies(a, b, base, match) |
|
|
553 | 571 | return copies |
|
|
554 | 572 | |
|
|
555 | 573 | |
|
|
556 | 574 | def _backwardrenames(a, b, match): |
|
|
557 | 575 | if a._repo.ui.config(b'experimental', b'copytrace') == b'off': |
|
|
558 | 576 | return {} |
|
|
559 | 577 | |
|
|
560 | 578 | # Even though we're not taking copies into account, 1:n rename situations |
|
|
561 | 579 | # can still exist (e.g. hg cp a b; hg mv a c). In those cases we |
|
|
562 | 580 | # arbitrarily pick one of the renames. |
|
|
563 | 581 | # We don't want to pass in "match" here, since that would filter |
|
|
564 | 582 | # the destination by it. Since we're reversing the copies, we want |
|
|
565 | 583 | # to filter the source instead. |
|
|
566 | 584 | f = _forwardcopies(b, a) |
|
|
567 | 585 | r = {} |
|
|
568 | 586 | for k, v in sorted(pycompat.iteritems(f)): |
|
|
569 | 587 | if match and not match(v): |
|
|
570 | 588 | continue |
|
|
571 | 589 | # remove copies |
|
|
572 | 590 | if v in a: |
|
|
573 | 591 | continue |
|
|
574 | 592 | r[v] = k |
|
|
575 | 593 | return r |
|
|
576 | 594 | |
|
|
577 | 595 | |
|
|
578 | 596 | def pathcopies(x, y, match=None): |
|
|
579 | 597 | """find {dst@y: src@x} copy mapping for directed compare""" |
|
|
580 | 598 | repo = x._repo |
|
|
581 | 599 | debug = repo.ui.debugflag and repo.ui.configbool(b'devel', b'debug.copies') |
|
|
582 | 600 | if debug: |
|
|
583 | 601 | repo.ui.debug( |
|
|
584 | 602 | b'debug.copies: searching copies from %s to %s\n' % (x, y) |
|
|
585 | 603 | ) |
|
|
586 | 604 | if x == y or not x or not y: |
|
|
587 | 605 | return {} |
|
|
588 | 606 | if y.rev() is None and x == y.p1(): |
|
|
589 | 607 | if debug: |
|
|
590 | 608 | repo.ui.debug(b'debug.copies: search mode: dirstate\n') |
|
|
591 | 609 | # short-circuit to avoid issues with merge states |
|
|
592 | 610 | return _dirstatecopies(repo, match) |
|
|
593 | 611 | a = y.ancestor(x) |
|
|
594 | 612 | if a == x: |
|
|
595 | 613 | if debug: |
|
|
596 | 614 | repo.ui.debug(b'debug.copies: search mode: forward\n') |
|
|
597 | 615 | copies = _forwardcopies(x, y, match=match) |
|
|
598 | 616 | elif a == y: |
|
|
599 | 617 | if debug: |
|
|
600 | 618 | repo.ui.debug(b'debug.copies: search mode: backward\n') |
|
|
601 | 619 | copies = _backwardrenames(x, y, match=match) |
|
|
602 | 620 | else: |
|
|
603 | 621 | if debug: |
|
|
604 | 622 | repo.ui.debug(b'debug.copies: search mode: combined\n') |
|
|
605 | 623 | base = None |
|
|
606 | 624 | if a.rev() != node.nullrev: |
|
|
607 | 625 | base = x |
|
|
608 | 626 | copies = _chain( |
|
|
609 | 627 | _backwardrenames(x, a, match=match), |
|
|
610 | 628 | _forwardcopies(a, y, base, match=match), |
|
|
611 | 629 | ) |
|
|
612 | 630 | _filter(x, y, copies) |
|
|
613 | 631 | return copies |
|
|
614 | 632 | |
|
|
615 | 633 | |
|
|
616 | 634 | def mergecopies(repo, c1, c2, base): |
|
|
617 | 635 | """ |
|
|
618 | 636 | Finds moves and copies between context c1 and c2 that are relevant for |
|
|
619 | 637 | merging. 'base' will be used as the merge base. |
|
|
620 | 638 | |
|
|
621 | 639 | Copytracing is used in commands like rebase, merge, unshelve, etc to merge |
|
|
622 | 640 | files that were moved/ copied in one merge parent and modified in another. |
|
|
623 | 641 | For example: |
|
|
624 | 642 | |
|
|
625 | 643 | o ---> 4 another commit |
|
|
626 | 644 | | |
|
|
627 | 645 | | o ---> 3 commit that modifies a.txt |
|
|
628 | 646 | | / |
|
|
629 | 647 | o / ---> 2 commit that moves a.txt to b.txt |
|
|
630 | 648 | |/ |
|
|
631 | 649 | o ---> 1 merge base |
|
|
632 | 650 | |
|
|
633 | 651 | If we try to rebase revision 3 on revision 4, since there is no a.txt in |
|
|
634 | 652 | revision 4, and if user have copytrace disabled, we prints the following |
|
|
635 | 653 | message: |
|
|
636 | 654 | |
|
|
637 | 655 | ```other changed <file> which local deleted``` |
|
|
638 | 656 | |
|
|
639 | 657 | Returns a tuple where: |
|
|
640 | 658 | |
|
|
641 | 659 | "branch_copies" an instance of branch_copies. |
|
|
642 | 660 | |
|
|
643 | 661 | "diverge" is a mapping of source name -> list of destination names |
|
|
644 | 662 | for divergent renames. |
|
|
645 | 663 | |
|
|
646 | 664 | This function calls different copytracing algorithms based on config. |
|
|
647 | 665 | """ |
|
|
648 | 666 | # avoid silly behavior for update from empty dir |
|
|
649 | 667 | if not c1 or not c2 or c1 == c2: |
|
|
650 | 668 | return branch_copies(), branch_copies(), {} |
|
|
651 | 669 | |
|
|
652 | 670 | narrowmatch = c1.repo().narrowmatch() |
|
|
653 | 671 | |
|
|
654 | 672 | # avoid silly behavior for parent -> working dir |
|
|
655 | 673 | if c2.node() is None and c1.node() == repo.dirstate.p1(): |
|
|
656 | 674 | return ( |
|
|
657 | 675 | branch_copies(_dirstatecopies(repo, narrowmatch)), |
|
|
658 | 676 | branch_copies(), |
|
|
659 | 677 | {}, |
|
|
660 | 678 | ) |
|
|
661 | 679 | |
|
|
662 | 680 | copytracing = repo.ui.config(b'experimental', b'copytrace') |
|
|
663 | 681 | if stringutil.parsebool(copytracing) is False: |
|
|
664 | 682 | # stringutil.parsebool() returns None when it is unable to parse the |
|
|
665 | 683 | # value, so we should rely on making sure copytracing is on such cases |
|
|
666 | 684 | return branch_copies(), branch_copies(), {} |
|
|
667 | 685 | |
|
|
668 | 686 | if usechangesetcentricalgo(repo): |
|
|
669 | 687 | # The heuristics don't make sense when we need changeset-centric algos |
|
|
670 | 688 | return _fullcopytracing(repo, c1, c2, base) |
|
|
671 | 689 | |
|
|
672 | 690 | # Copy trace disabling is explicitly below the node == p1 logic above |
|
|
673 | 691 | # because the logic above is required for a simple copy to be kept across a |
|
|
674 | 692 | # rebase. |
|
|
675 | 693 | if copytracing == b'heuristics': |
|
|
676 | 694 | # Do full copytracing if only non-public revisions are involved as |
|
|
677 | 695 | # that will be fast enough and will also cover the copies which could |
|
|
678 | 696 | # be missed by heuristics |
|
|
679 | 697 | if _isfullcopytraceable(repo, c1, base): |
|
|
680 | 698 | return _fullcopytracing(repo, c1, c2, base) |
|
|
681 | 699 | return _heuristicscopytracing(repo, c1, c2, base) |
|
|
682 | 700 | else: |
|
|
683 | 701 | return _fullcopytracing(repo, c1, c2, base) |
|
|
684 | 702 | |
|
|
685 | 703 | |
|
|
686 | 704 | def _isfullcopytraceable(repo, c1, base): |
|
|
687 | 705 | """ Checks that if base, source and destination are all no-public branches, |
|
|
688 | 706 | if yes let's use the full copytrace algorithm for increased capabilities |
|
|
689 | 707 | since it will be fast enough. |
|
|
690 | 708 | |
|
|
691 | 709 | `experimental.copytrace.sourcecommitlimit` can be used to set a limit for |
|
|
692 | 710 | number of changesets from c1 to base such that if number of changesets are |
|
|
693 | 711 | more than the limit, full copytracing algorithm won't be used. |
|
|
694 | 712 | """ |
|
|
695 | 713 | if c1.rev() is None: |
|
|
696 | 714 | c1 = c1.p1() |
|
|
697 | 715 | if c1.mutable() and base.mutable(): |
|
|
698 | 716 | sourcecommitlimit = repo.ui.configint( |
|
|
699 | 717 | b'experimental', b'copytrace.sourcecommitlimit' |
|
|
700 | 718 | ) |
|
|
701 | 719 | commits = len(repo.revs(b'%d::%d', base.rev(), c1.rev())) |
|
|
702 | 720 | return commits < sourcecommitlimit |
|
|
703 | 721 | return False |
|
|
704 | 722 | |
|
|
705 | 723 | |
|
|
706 | 724 | def _checksinglesidecopies( |
|
|
707 | 725 | src, dsts1, m1, m2, mb, c2, base, copy, renamedelete |
|
|
708 | 726 | ): |
|
|
709 | 727 | if src not in m2: |
|
|
710 | 728 | # deleted on side 2 |
|
|
711 | 729 | if src not in m1: |
|
|
712 | 730 | # renamed on side 1, deleted on side 2 |
|
|
713 | 731 | renamedelete[src] = dsts1 |
|
|
714 | 732 | elif src not in mb: |
|
|
715 | 733 | # Work around the "short-circuit to avoid issues with merge states" |
|
|
716 | 734 | # thing in pathcopies(): pathcopies(x, y) can return a copy where the |
|
|
717 | 735 | # destination doesn't exist in y. |
|
|
718 | 736 | pass |
|
|
719 | 737 | elif mb[src] != m2[src] and not _related(c2[src], base[src]): |
|
|
720 | 738 | return |
|
|
721 | 739 | elif mb[src] != m2[src] or mb.flags(src) != m2.flags(src): |
|
|
722 | 740 | # modified on side 2 |
|
|
723 | 741 | for dst in dsts1: |
|
|
724 | 742 | copy[dst] = src |
|
|
725 | 743 | |
|
|
726 | 744 | |
|
|
727 | 745 | class branch_copies(object): |
|
|
728 | 746 | """Information about copies made on one side of a merge/graft. |
|
|
729 | 747 | |
|
|
730 | 748 | "copy" is a mapping from destination name -> source name, |
|
|
731 | 749 | where source is in c1 and destination is in c2 or vice-versa. |
|
|
732 | 750 | |
|
|
733 | 751 | "movewithdir" is a mapping from source name -> destination name, |
|
|
734 | 752 | where the file at source present in one context but not the other |
|
|
735 | 753 | needs to be moved to destination by the merge process, because the |
|
|
736 | 754 | other context moved the directory it is in. |
|
|
737 | 755 | |
|
|
738 | 756 | "renamedelete" is a mapping of source name -> list of destination |
|
|
739 | 757 | names for files deleted in c1 that were renamed in c2 or vice-versa. |
|
|
740 | 758 | |
|
|
741 | 759 | "dirmove" is a mapping of detected source dir -> destination dir renames. |
|
|
742 | 760 | This is needed for handling changes to new files previously grafted into |
|
|
743 | 761 | renamed directories. |
|
|
744 | 762 | """ |
|
|
745 | 763 | |
|
|
746 | 764 | def __init__( |
|
|
747 | 765 | self, copy=None, renamedelete=None, dirmove=None, movewithdir=None |
|
|
748 | 766 | ): |
|
|
749 | 767 | self.copy = {} if copy is None else copy |
|
|
750 | 768 | self.renamedelete = {} if renamedelete is None else renamedelete |
|
|
751 | 769 | self.dirmove = {} if dirmove is None else dirmove |
|
|
752 | 770 | self.movewithdir = {} if movewithdir is None else movewithdir |
|
|
753 | 771 | |
|
|
754 | 772 | def __repr__(self): |
|
|
755 | 773 | return ( |
|
|
756 | 774 | '<branch_copies\n copy=%r\n renamedelete=%r\n dirmove=%r\n movewithdir=%r\n>' |
|
|
757 | 775 | % (self.copy, self.renamedelete, self.dirmove, self.movewithdir,) |
|
|
758 | 776 | ) |
|
|
759 | 777 | |
|
|
760 | 778 | |
|
|
761 | 779 | def _fullcopytracing(repo, c1, c2, base): |
|
|
762 | 780 | """ The full copytracing algorithm which finds all the new files that were |
|
|
763 | 781 | added from merge base up to the top commit and for each file it checks if |
|
|
764 | 782 | this file was copied from another file. |
|
|
765 | 783 | |
|
|
766 | 784 | This is pretty slow when a lot of changesets are involved but will track all |
|
|
767 | 785 | the copies. |
|
|
768 | 786 | """ |
|
|
769 | 787 | m1 = c1.manifest() |
|
|
770 | 788 | m2 = c2.manifest() |
|
|
771 | 789 | mb = base.manifest() |
|
|
772 | 790 | |
|
|
773 | 791 | copies1 = pathcopies(base, c1) |
|
|
774 | 792 | copies2 = pathcopies(base, c2) |
|
|
775 | 793 | |
|
|
776 | 794 | if not (copies1 or copies2): |
|
|
777 | 795 | return branch_copies(), branch_copies(), {} |
|
|
778 | 796 | |
|
|
779 | 797 | inversecopies1 = {} |
|
|
780 | 798 | inversecopies2 = {} |
|
|
781 | 799 | for dst, src in copies1.items(): |
|
|
782 | 800 | inversecopies1.setdefault(src, []).append(dst) |
|
|
783 | 801 | for dst, src in copies2.items(): |
|
|
784 | 802 | inversecopies2.setdefault(src, []).append(dst) |
|
|
785 | 803 | |
|
|
786 | 804 | copy1 = {} |
|
|
787 | 805 | copy2 = {} |
|
|
788 | 806 | diverge = {} |
|
|
789 | 807 | renamedelete1 = {} |
|
|
790 | 808 | renamedelete2 = {} |
|
|
791 | 809 | allsources = set(inversecopies1) | set(inversecopies2) |
|
|
792 | 810 | for src in allsources: |
|
|
793 | 811 | dsts1 = inversecopies1.get(src) |
|
|
794 | 812 | dsts2 = inversecopies2.get(src) |
|
|
795 | 813 | if dsts1 and dsts2: |
|
|
796 | 814 | # copied/renamed on both sides |
|
|
797 | 815 | if src not in m1 and src not in m2: |
|
|
798 | 816 | # renamed on both sides |
|
|
799 | 817 | dsts1 = set(dsts1) |
|
|
800 | 818 | dsts2 = set(dsts2) |
|
|
801 | 819 | # If there's some overlap in the rename destinations, we |
|
|
802 | 820 | # consider it not divergent. For example, if side 1 copies 'a' |
|
|
803 | 821 | # to 'b' and 'c' and deletes 'a', and side 2 copies 'a' to 'c' |
|
|
804 | 822 | # and 'd' and deletes 'a'. |
|
|
805 | 823 | if dsts1 & dsts2: |
|
|
806 | 824 | for dst in dsts1 & dsts2: |
|
|
807 | 825 | copy1[dst] = src |
|
|
808 | 826 | copy2[dst] = src |
|
|
809 | 827 | else: |
|
|
810 | 828 | diverge[src] = sorted(dsts1 | dsts2) |
|
|
811 | 829 | elif src in m1 and src in m2: |
|
|
812 | 830 | # copied on both sides |
|
|
813 | 831 | dsts1 = set(dsts1) |
|
|
814 | 832 | dsts2 = set(dsts2) |
|
|
815 | 833 | for dst in dsts1 & dsts2: |
|
|
816 | 834 | copy1[dst] = src |
|
|
817 | 835 | copy2[dst] = src |
|
|
818 | 836 | # TODO: Handle cases where it was renamed on one side and copied |
|
|
819 | 837 | # on the other side |
|
|
820 | 838 | elif dsts1: |
|
|
821 | 839 | # copied/renamed only on side 1 |
|
|
822 | 840 | _checksinglesidecopies( |
|
|
823 | 841 | src, dsts1, m1, m2, mb, c2, base, copy1, renamedelete1 |
|
|
824 | 842 | ) |
|
|
825 | 843 | elif dsts2: |
|
|
826 | 844 | # copied/renamed only on side 2 |
|
|
827 | 845 | _checksinglesidecopies( |
|
|
828 | 846 | src, dsts2, m2, m1, mb, c1, base, copy2, renamedelete2 |
|
|
829 | 847 | ) |
|
|
830 | 848 | |
|
|
831 | 849 | # find interesting file sets from manifests |
|
|
832 | 850 | addedinm1 = m1.filesnotin(mb, repo.narrowmatch()) |
|
|
833 | 851 | addedinm2 = m2.filesnotin(mb, repo.narrowmatch()) |
|
|
834 | 852 | u1 = sorted(addedinm1 - addedinm2) |
|
|
835 | 853 | u2 = sorted(addedinm2 - addedinm1) |
|
|
836 | 854 | |
|
|
837 | 855 | header = b" unmatched files in %s" |
|
|
838 | 856 | if u1: |
|
|
839 | 857 | repo.ui.debug(b"%s:\n %s\n" % (header % b'local', b"\n ".join(u1))) |
|
|
840 | 858 | if u2: |
|
|
841 | 859 | repo.ui.debug(b"%s:\n %s\n" % (header % b'other', b"\n ".join(u2))) |
|
|
842 | 860 | |
|
|
843 | 861 | if repo.ui.debugflag: |
|
|
844 | 862 | renamedeleteset = set() |
|
|
845 | 863 | divergeset = set() |
|
|
846 | 864 | for dsts in diverge.values(): |
|
|
847 | 865 | divergeset.update(dsts) |
|
|
848 | 866 | for dsts in renamedelete1.values(): |
|
|
849 | 867 | renamedeleteset.update(dsts) |
|
|
850 | 868 | for dsts in renamedelete2.values(): |
|
|
851 | 869 | renamedeleteset.update(dsts) |
|
|
852 | 870 | |
|
|
853 | 871 | repo.ui.debug( |
|
|
854 | 872 | b" all copies found (* = to merge, ! = divergent, " |
|
|
855 | 873 | b"% = renamed and deleted):\n" |
|
|
856 | 874 | ) |
|
|
857 | 875 | for side, copies in ((b"local", copies1), (b"remote", copies2)): |
|
|
858 | 876 | if not copies: |
|
|
859 | 877 | continue |
|
|
860 | 878 | repo.ui.debug(b" on %s side:\n" % side) |
|
|
861 | 879 | for f in sorted(copies): |
|
|
862 | 880 | note = b"" |
|
|
863 | 881 | if f in copy1 or f in copy2: |
|
|
864 | 882 | note += b"*" |
|
|
865 | 883 | if f in divergeset: |
|
|
866 | 884 | note += b"!" |
|
|
867 | 885 | if f in renamedeleteset: |
|
|
868 | 886 | note += b"%" |
|
|
869 | 887 | repo.ui.debug( |
|
|
870 | 888 | b" src: '%s' -> dst: '%s' %s\n" % (copies[f], f, note) |
|
|
871 | 889 | ) |
|
|
872 | 890 | del renamedeleteset |
|
|
873 | 891 | del divergeset |
|
|
874 | 892 | |
|
|
875 | 893 | repo.ui.debug(b" checking for directory renames\n") |
|
|
876 | 894 | |
|
|
877 | 895 | dirmove1, movewithdir2 = _dir_renames(repo, c1, copy1, copies1, u2) |
|
|
878 | 896 | dirmove2, movewithdir1 = _dir_renames(repo, c2, copy2, copies2, u1) |
|
|
879 | 897 | |
|
|
880 | 898 | branch_copies1 = branch_copies(copy1, renamedelete1, dirmove1, movewithdir1) |
|
|
881 | 899 | branch_copies2 = branch_copies(copy2, renamedelete2, dirmove2, movewithdir2) |
|
|
882 | 900 | |
|
|
883 | 901 | return branch_copies1, branch_copies2, diverge |
|
|
884 | 902 | |
|
|
885 | 903 | |
|
|
886 | 904 | def _dir_renames(repo, ctx, copy, fullcopy, addedfiles): |
|
|
887 | 905 | """Finds moved directories and files that should move with them. |
|
|
888 | 906 | |
|
|
889 | 907 | ctx: the context for one of the sides |
|
|
890 | 908 | copy: files copied on the same side (as ctx) |
|
|
891 | 909 | fullcopy: files copied on the same side (as ctx), including those that |
|
|
892 | 910 | merge.manifestmerge() won't care about |
|
|
893 | 911 | addedfiles: added files on the other side (compared to ctx) |
|
|
894 | 912 | """ |
|
|
895 | 913 | # generate a directory move map |
|
|
896 | 914 | d = ctx.dirs() |
|
|
897 | 915 | invalid = set() |
|
|
898 | 916 | dirmove = {} |
|
|
899 | 917 | |
|
|
900 | 918 | # examine each file copy for a potential directory move, which is |
|
|
901 | 919 | # when all the files in a directory are moved to a new directory |
|
|
902 | 920 | for dst, src in pycompat.iteritems(fullcopy): |
|
|
903 | 921 | dsrc, ddst = pathutil.dirname(src), pathutil.dirname(dst) |
|
|
904 | 922 | if dsrc in invalid: |
|
|
905 | 923 | # already seen to be uninteresting |
|
|
906 | 924 | continue |
|
|
907 | 925 | elif dsrc in d and ddst in d: |
|
|
908 | 926 | # directory wasn't entirely moved locally |
|
|
909 | 927 | invalid.add(dsrc) |
|
|
910 | 928 | elif dsrc in dirmove and dirmove[dsrc] != ddst: |
|
|
911 | 929 | # files from the same directory moved to two different places |
|
|
912 | 930 | invalid.add(dsrc) |
|
|
913 | 931 | else: |
|
|
914 | 932 | # looks good so far |
|
|
915 | 933 | dirmove[dsrc] = ddst |
|
|
916 | 934 | |
|
|
917 | 935 | for i in invalid: |
|
|
918 | 936 | if i in dirmove: |
|
|
919 | 937 | del dirmove[i] |
|
|
920 | 938 | del d, invalid |
|
|
921 | 939 | |
|
|
922 | 940 | if not dirmove: |
|
|
923 | 941 | return {}, {} |
|
|
924 | 942 | |
|
|
925 | 943 | dirmove = {k + b"/": v + b"/" for k, v in pycompat.iteritems(dirmove)} |
|
|
926 | 944 | |
|
|
927 | 945 | for d in dirmove: |
|
|
928 | 946 | repo.ui.debug( |
|
|
929 | 947 | b" discovered dir src: '%s' -> dst: '%s'\n" % (d, dirmove[d]) |
|
|
930 | 948 | ) |
|
|
931 | 949 | |
|
|
932 | 950 | movewithdir = {} |
|
|
933 | 951 | # check unaccounted nonoverlapping files against directory moves |
|
|
934 | 952 | for f in addedfiles: |
|
|
935 | 953 | if f not in fullcopy: |
|
|
936 | 954 | for d in dirmove: |
|
|
937 | 955 | if f.startswith(d): |
|
|
938 | 956 | # new file added in a directory that was moved, move it |
|
|
939 | 957 | df = dirmove[d] + f[len(d) :] |
|
|
940 | 958 | if df not in copy: |
|
|
941 | 959 | movewithdir[f] = df |
|
|
942 | 960 | repo.ui.debug( |
|
|
943 | 961 | b" pending file src: '%s' -> dst: '%s'\n" |
|
|
944 | 962 | % (f, df) |
|
|
945 | 963 | ) |
|
|
946 | 964 | break |
|
|
947 | 965 | |
|
|
948 | 966 | return dirmove, movewithdir |
|
|
949 | 967 | |
|
|
950 | 968 | |
|
|
951 | 969 | def _heuristicscopytracing(repo, c1, c2, base): |
|
|
952 | 970 | """ Fast copytracing using filename heuristics |
|
|
953 | 971 | |
|
|
954 | 972 | Assumes that moves or renames are of following two types: |
|
|
955 | 973 | |
|
|
956 | 974 | 1) Inside a directory only (same directory name but different filenames) |
|
|
957 | 975 | 2) Move from one directory to another |
|
|
958 | 976 | (same filenames but different directory names) |
|
|
959 | 977 | |
|
|
960 | 978 | Works only when there are no merge commits in the "source branch". |
|
|
961 | 979 | Source branch is commits from base up to c2 not including base. |
|
|
962 | 980 | |
|
|
963 | 981 | If merge is involved it fallbacks to _fullcopytracing(). |
|
|
964 | 982 | |
|
|
965 | 983 | Can be used by setting the following config: |
|
|
966 | 984 | |
|
|
967 | 985 | [experimental] |
|
|
968 | 986 | copytrace = heuristics |
|
|
969 | 987 | |
|
|
970 | 988 | In some cases the copy/move candidates found by heuristics can be very large |
|
|
971 | 989 | in number and that will make the algorithm slow. The number of possible |
|
|
972 | 990 | candidates to check can be limited by using the config |
|
|
973 | 991 | `experimental.copytrace.movecandidateslimit` which defaults to 100. |
|
|
974 | 992 | """ |
|
|
975 | 993 | |
|
|
976 | 994 | if c1.rev() is None: |
|
|
977 | 995 | c1 = c1.p1() |
|
|
978 | 996 | if c2.rev() is None: |
|
|
979 | 997 | c2 = c2.p1() |
|
|
980 | 998 | |
|
|
981 | 999 | changedfiles = set() |
|
|
982 | 1000 | m1 = c1.manifest() |
|
|
983 | 1001 | if not repo.revs(b'%d::%d', base.rev(), c2.rev()): |
|
|
984 | 1002 | # If base is not in c2 branch, we switch to fullcopytracing |
|
|
985 | 1003 | repo.ui.debug( |
|
|
986 | 1004 | b"switching to full copytracing as base is not " |
|
|
987 | 1005 | b"an ancestor of c2\n" |
|
|
988 | 1006 | ) |
|
|
989 | 1007 | return _fullcopytracing(repo, c1, c2, base) |
|
|
990 | 1008 | |
|
|
991 | 1009 | ctx = c2 |
|
|
992 | 1010 | while ctx != base: |
|
|
993 | 1011 | if len(ctx.parents()) == 2: |
|
|
994 | 1012 | # To keep things simple let's not handle merges |
|
|
995 | 1013 | repo.ui.debug(b"switching to full copytracing because of merges\n") |
|
|
996 | 1014 | return _fullcopytracing(repo, c1, c2, base) |
|
|
997 | 1015 | changedfiles.update(ctx.files()) |
|
|
998 | 1016 | ctx = ctx.p1() |
|
|
999 | 1017 | |
|
|
1000 | 1018 | copies2 = {} |
|
|
1001 | 1019 | cp = _forwardcopies(base, c2) |
|
|
1002 | 1020 | for dst, src in pycompat.iteritems(cp): |
|
|
1003 | 1021 | if src in m1: |
|
|
1004 | 1022 | copies2[dst] = src |
|
|
1005 | 1023 | |
|
|
1006 | 1024 | # file is missing if it isn't present in the destination, but is present in |
|
|
1007 | 1025 | # the base and present in the source. |
|
|
1008 | 1026 | # Presence in the base is important to exclude added files, presence in the |
|
|
1009 | 1027 | # source is important to exclude removed files. |
|
|
1010 | 1028 | filt = lambda f: f not in m1 and f in base and f in c2 |
|
|
1011 | 1029 | missingfiles = [f for f in changedfiles if filt(f)] |
|
|
1012 | 1030 | |
|
|
1013 | 1031 | copies1 = {} |
|
|
1014 | 1032 | if missingfiles: |
|
|
1015 | 1033 | basenametofilename = collections.defaultdict(list) |
|
|
1016 | 1034 | dirnametofilename = collections.defaultdict(list) |
|
|
1017 | 1035 | |
|
|
1018 | 1036 | for f in m1.filesnotin(base.manifest()): |
|
|
1019 | 1037 | basename = os.path.basename(f) |
|
|
1020 | 1038 | dirname = os.path.dirname(f) |
|
|
1021 | 1039 | basenametofilename[basename].append(f) |
|
|
1022 | 1040 | dirnametofilename[dirname].append(f) |
|
|
1023 | 1041 | |
|
|
1024 | 1042 | for f in missingfiles: |
|
|
1025 | 1043 | basename = os.path.basename(f) |
|
|
1026 | 1044 | dirname = os.path.dirname(f) |
|
|
1027 | 1045 | samebasename = basenametofilename[basename] |
|
|
1028 | 1046 | samedirname = dirnametofilename[dirname] |
|
|
1029 | 1047 | movecandidates = samebasename + samedirname |
|
|
1030 | 1048 | # f is guaranteed to be present in c2, that's why |
|
|
1031 | 1049 | # c2.filectx(f) won't fail |
|
|
1032 | 1050 | f2 = c2.filectx(f) |
|
|
1033 | 1051 | # we can have a lot of candidates which can slow down the heuristics |
|
|
1034 | 1052 | # config value to limit the number of candidates moves to check |
|
|
1035 | 1053 | maxcandidates = repo.ui.configint( |
|
|
1036 | 1054 | b'experimental', b'copytrace.movecandidateslimit' |
|
|
1037 | 1055 | ) |
|
|
1038 | 1056 | |
|
|
1039 | 1057 | if len(movecandidates) > maxcandidates: |
|
|
1040 | 1058 | repo.ui.status( |
|
|
1041 | 1059 | _( |
|
|
1042 | 1060 | b"skipping copytracing for '%s', more " |
|
|
1043 | 1061 | b"candidates than the limit: %d\n" |
|
|
1044 | 1062 | ) |
|
|
1045 | 1063 | % (f, len(movecandidates)) |
|
|
1046 | 1064 | ) |
|
|
1047 | 1065 | continue |
|
|
1048 | 1066 | |
|
|
1049 | 1067 | for candidate in movecandidates: |
|
|
1050 | 1068 | f1 = c1.filectx(candidate) |
|
|
1051 | 1069 | if _related(f1, f2): |
|
|
1052 | 1070 | # if there are a few related copies then we'll merge |
|
|
1053 | 1071 | # changes into all of them. This matches the behaviour |
|
|
1054 | 1072 | # of upstream copytracing |
|
|
1055 | 1073 | copies1[candidate] = f |
|
|
1056 | 1074 | |
|
|
1057 | 1075 | return branch_copies(copies1), branch_copies(copies2), {} |
|
|
1058 | 1076 | |
|
|
1059 | 1077 | |
|
|
1060 | 1078 | def _related(f1, f2): |
|
|
1061 | 1079 | """return True if f1 and f2 filectx have a common ancestor |
|
|
1062 | 1080 | |
|
|
1063 | 1081 | Walk back to common ancestor to see if the two files originate |
|
|
1064 | 1082 | from the same file. Since workingfilectx's rev() is None it messes |
|
|
1065 | 1083 | up the integer comparison logic, hence the pre-step check for |
|
|
1066 | 1084 | None (f1 and f2 can only be workingfilectx's initially). |
|
|
1067 | 1085 | """ |
|
|
1068 | 1086 | |
|
|
1069 | 1087 | if f1 == f2: |
|
|
1070 | 1088 | return True # a match |
|
|
1071 | 1089 | |
|
|
1072 | 1090 | g1, g2 = f1.ancestors(), f2.ancestors() |
|
|
1073 | 1091 | try: |
|
|
1074 | 1092 | f1r, f2r = f1.linkrev(), f2.linkrev() |
|
|
1075 | 1093 | |
|
|
1076 | 1094 | if f1r is None: |
|
|
1077 | 1095 | f1 = next(g1) |
|
|
1078 | 1096 | if f2r is None: |
|
|
1079 | 1097 | f2 = next(g2) |
|
|
1080 | 1098 | |
|
|
1081 | 1099 | while True: |
|
|
1082 | 1100 | f1r, f2r = f1.linkrev(), f2.linkrev() |
|
|
1083 | 1101 | if f1r > f2r: |
|
|
1084 | 1102 | f1 = next(g1) |
|
|
1085 | 1103 | elif f2r > f1r: |
|
|
1086 | 1104 | f2 = next(g2) |
|
|
1087 | 1105 | else: # f1 and f2 point to files in the same linkrev |
|
|
1088 | 1106 | return f1 == f2 # true if they point to the same file |
|
|
1089 | 1107 | except StopIteration: |
|
|
1090 | 1108 | return False |
|
|
1091 | 1109 | |
|
|
1092 | 1110 | |
|
|
1093 | 1111 | def graftcopies(wctx, ctx, base): |
|
|
1094 | 1112 | """reproduce copies between base and ctx in the wctx |
|
|
1095 | 1113 | |
|
|
1096 | 1114 | Unlike mergecopies(), this function will only consider copies between base |
|
|
1097 | 1115 | and ctx; it will ignore copies between base and wctx. Also unlike |
|
|
1098 | 1116 | mergecopies(), this function will apply copies to the working copy (instead |
|
|
1099 | 1117 | of just returning information about the copies). That makes it cheaper |
|
|
1100 | 1118 | (especially in the common case of base==ctx.p1()) and useful also when |
|
|
1101 | 1119 | experimental.copytrace=off. |
|
|
1102 | 1120 | |
|
|
1103 | 1121 | merge.update() will have already marked most copies, but it will only |
|
|
1104 | 1122 | mark copies if it thinks the source files are related (see |
|
|
1105 | 1123 | merge._related()). It will also not mark copies if the file wasn't modified |
|
|
1106 | 1124 | on the local side. This function adds the copies that were "missed" |
|
|
1107 | 1125 | by merge.update(). |
|
|
1108 | 1126 | """ |
|
|
1109 | 1127 | new_copies = pathcopies(base, ctx) |
|
|
1110 | 1128 | _filter(wctx.p1(), wctx, new_copies) |
|
|
1111 | 1129 | for dst, src in pycompat.iteritems(new_copies): |
|
|
1112 | 1130 | wctx[dst].markcopied(src) |
General Comments 0
You need to be logged in to leave comments.
Login now