##// END OF EJS Templates
copies: move short-circuiting of dirstate copies out of _forwardcopies()...
Martin von Zweigbergk -
r42794:bcb4b5c5 default
parent child Browse files
Show More
@@ -1,806 +1,805 b''
1 # copies.py - copy detection for Mercurial
1 # copies.py - copy detection for Mercurial
2 #
2 #
3 # Copyright 2008 Matt Mackall <mpm@selenic.com>
3 # Copyright 2008 Matt Mackall <mpm@selenic.com>
4 #
4 #
5 # This software may be used and distributed according to the terms of the
5 # This software may be used and distributed according to the terms of the
6 # GNU General Public License version 2 or any later version.
6 # GNU General Public License version 2 or any later version.
7
7
8 from __future__ import absolute_import
8 from __future__ import absolute_import
9
9
10 import collections
10 import collections
11 import heapq
11 import heapq
12 import os
12 import os
13
13
14 from .i18n import _
14 from .i18n import _
15
15
16 from . import (
16 from . import (
17 match as matchmod,
17 match as matchmod,
18 node,
18 node,
19 pathutil,
19 pathutil,
20 util,
20 util,
21 )
21 )
22 from .utils import (
22 from .utils import (
23 stringutil,
23 stringutil,
24 )
24 )
25
25
26 def _findlimit(repo, ctxa, ctxb):
26 def _findlimit(repo, ctxa, ctxb):
27 """
27 """
28 Find the last revision that needs to be checked to ensure that a full
28 Find the last revision that needs to be checked to ensure that a full
29 transitive closure for file copies can be properly calculated.
29 transitive closure for file copies can be properly calculated.
30 Generally, this means finding the earliest revision number that's an
30 Generally, this means finding the earliest revision number that's an
31 ancestor of a or b but not both, except when a or b is a direct descendent
31 ancestor of a or b but not both, except when a or b is a direct descendent
32 of the other, in which case we can return the minimum revnum of a and b.
32 of the other, in which case we can return the minimum revnum of a and b.
33 """
33 """
34
34
35 # basic idea:
35 # basic idea:
36 # - mark a and b with different sides
36 # - mark a and b with different sides
37 # - if a parent's children are all on the same side, the parent is
37 # - if a parent's children are all on the same side, the parent is
38 # on that side, otherwise it is on no side
38 # on that side, otherwise it is on no side
39 # - walk the graph in topological order with the help of a heap;
39 # - walk the graph in topological order with the help of a heap;
40 # - add unseen parents to side map
40 # - add unseen parents to side map
41 # - clear side of any parent that has children on different sides
41 # - clear side of any parent that has children on different sides
42 # - track number of interesting revs that might still be on a side
42 # - track number of interesting revs that might still be on a side
43 # - track the lowest interesting rev seen
43 # - track the lowest interesting rev seen
44 # - quit when interesting revs is zero
44 # - quit when interesting revs is zero
45
45
46 cl = repo.changelog
46 cl = repo.changelog
47 wdirparents = None
47 wdirparents = None
48 a = ctxa.rev()
48 a = ctxa.rev()
49 b = ctxb.rev()
49 b = ctxb.rev()
50 if a is None:
50 if a is None:
51 wdirparents = (ctxa.p1(), ctxa.p2())
51 wdirparents = (ctxa.p1(), ctxa.p2())
52 a = node.wdirrev
52 a = node.wdirrev
53 if b is None:
53 if b is None:
54 assert not wdirparents
54 assert not wdirparents
55 wdirparents = (ctxb.p1(), ctxb.p2())
55 wdirparents = (ctxb.p1(), ctxb.p2())
56 b = node.wdirrev
56 b = node.wdirrev
57
57
58 side = {a: -1, b: 1}
58 side = {a: -1, b: 1}
59 visit = [-a, -b]
59 visit = [-a, -b]
60 heapq.heapify(visit)
60 heapq.heapify(visit)
61 interesting = len(visit)
61 interesting = len(visit)
62 limit = node.wdirrev
62 limit = node.wdirrev
63
63
64 while interesting:
64 while interesting:
65 r = -heapq.heappop(visit)
65 r = -heapq.heappop(visit)
66 if r == node.wdirrev:
66 if r == node.wdirrev:
67 parents = [pctx.rev() for pctx in wdirparents]
67 parents = [pctx.rev() for pctx in wdirparents]
68 else:
68 else:
69 parents = cl.parentrevs(r)
69 parents = cl.parentrevs(r)
70 if parents[1] == node.nullrev:
70 if parents[1] == node.nullrev:
71 parents = parents[:1]
71 parents = parents[:1]
72 for p in parents:
72 for p in parents:
73 if p not in side:
73 if p not in side:
74 # first time we see p; add it to visit
74 # first time we see p; add it to visit
75 side[p] = side[r]
75 side[p] = side[r]
76 if side[p]:
76 if side[p]:
77 interesting += 1
77 interesting += 1
78 heapq.heappush(visit, -p)
78 heapq.heappush(visit, -p)
79 elif side[p] and side[p] != side[r]:
79 elif side[p] and side[p] != side[r]:
80 # p was interesting but now we know better
80 # p was interesting but now we know better
81 side[p] = 0
81 side[p] = 0
82 interesting -= 1
82 interesting -= 1
83 if side[r]:
83 if side[r]:
84 limit = r # lowest rev visited
84 limit = r # lowest rev visited
85 interesting -= 1
85 interesting -= 1
86
86
87 # Consider the following flow (see test-commit-amend.t under issue4405):
87 # Consider the following flow (see test-commit-amend.t under issue4405):
88 # 1/ File 'a0' committed
88 # 1/ File 'a0' committed
89 # 2/ File renamed from 'a0' to 'a1' in a new commit (call it 'a1')
89 # 2/ File renamed from 'a0' to 'a1' in a new commit (call it 'a1')
90 # 3/ Move back to first commit
90 # 3/ Move back to first commit
91 # 4/ Create a new commit via revert to contents of 'a1' (call it 'a1-amend')
91 # 4/ Create a new commit via revert to contents of 'a1' (call it 'a1-amend')
92 # 5/ Rename file from 'a1' to 'a2' and commit --amend 'a1-msg'
92 # 5/ Rename file from 'a1' to 'a2' and commit --amend 'a1-msg'
93 #
93 #
94 # During the amend in step five, we will be in this state:
94 # During the amend in step five, we will be in this state:
95 #
95 #
96 # @ 3 temporary amend commit for a1-amend
96 # @ 3 temporary amend commit for a1-amend
97 # |
97 # |
98 # o 2 a1-amend
98 # o 2 a1-amend
99 # |
99 # |
100 # | o 1 a1
100 # | o 1 a1
101 # |/
101 # |/
102 # o 0 a0
102 # o 0 a0
103 #
103 #
104 # When _findlimit is called, a and b are revs 3 and 0, so limit will be 2,
104 # When _findlimit is called, a and b are revs 3 and 0, so limit will be 2,
105 # yet the filelog has the copy information in rev 1 and we will not look
105 # yet the filelog has the copy information in rev 1 and we will not look
106 # back far enough unless we also look at the a and b as candidates.
106 # back far enough unless we also look at the a and b as candidates.
107 # This only occurs when a is a descendent of b or visa-versa.
107 # This only occurs when a is a descendent of b or visa-versa.
108 return min(limit, a, b)
108 return min(limit, a, b)
109
109
110 def _chainandfilter(src, dst, a, b):
110 def _chainandfilter(src, dst, a, b):
111 """chain two sets of copies 'a' and 'b' and filter result"""
111 """chain two sets of copies 'a' and 'b' and filter result"""
112
112
113 # When chaining copies in 'a' (from 'src' via some other commit 'mid') with
113 # When chaining copies in 'a' (from 'src' via some other commit 'mid') with
114 # copies in 'b' (from 'mid' to 'dst'), we can get the different cases in the
114 # copies in 'b' (from 'mid' to 'dst'), we can get the different cases in the
115 # following table (not including trivial cases). For example, case 2 is
115 # following table (not including trivial cases). For example, case 2 is
116 # where a file existed in 'src' and remained under that name in 'mid' and
116 # where a file existed in 'src' and remained under that name in 'mid' and
117 # then was renamed between 'mid' and 'dst'.
117 # then was renamed between 'mid' and 'dst'.
118 #
118 #
119 # case src mid dst result
119 # case src mid dst result
120 # 1 x y - -
120 # 1 x y - -
121 # 2 x y y x->y
121 # 2 x y y x->y
122 # 3 x y x -
122 # 3 x y x -
123 # 4 x y z x->z
123 # 4 x y z x->z
124 # 5 - x y -
124 # 5 - x y -
125 # 6 x x y x->y
125 # 6 x x y x->y
126 #
126 #
127 # _chain() takes care of chaining the copies in 'a' and 'b', but it
127 # _chain() takes care of chaining the copies in 'a' and 'b', but it
128 # cannot tell the difference between cases 1 and 2, between 3 and 4, or
128 # cannot tell the difference between cases 1 and 2, between 3 and 4, or
129 # between 5 and 6, so it includes all cases in its result.
129 # between 5 and 6, so it includes all cases in its result.
130 # Cases 1, 3, and 5 are then removed by _filter().
130 # Cases 1, 3, and 5 are then removed by _filter().
131
131
132 t = _chain(a, b)
132 t = _chain(a, b)
133 _filter(src, dst, t)
133 _filter(src, dst, t)
134 return t
134 return t
135
135
136 def _filter(src, dst, t):
136 def _filter(src, dst, t):
137 """filters out invalid copies after chaining"""
137 """filters out invalid copies after chaining"""
138 for k, v in list(t.items()):
138 for k, v in list(t.items()):
139 # remove copies from files that didn't exist
139 # remove copies from files that didn't exist
140 if v not in src:
140 if v not in src:
141 del t[k]
141 del t[k]
142 # remove criss-crossed copies
142 # remove criss-crossed copies
143 elif k in src and v in dst:
143 elif k in src and v in dst:
144 del t[k]
144 del t[k]
145 # remove copies to files that were then removed
145 # remove copies to files that were then removed
146 elif k not in dst:
146 elif k not in dst:
147 del t[k]
147 del t[k]
148
148
149 def _chain(a, b):
149 def _chain(a, b):
150 """chain two sets of copies 'a' and 'b'"""
150 """chain two sets of copies 'a' and 'b'"""
151 t = a.copy()
151 t = a.copy()
152 for k, v in b.iteritems():
152 for k, v in b.iteritems():
153 if v in t:
153 if v in t:
154 t[k] = t[v]
154 t[k] = t[v]
155 else:
155 else:
156 t[k] = v
156 t[k] = v
157 return t
157 return t
158
158
159 def _tracefile(fctx, am, limit):
159 def _tracefile(fctx, am, limit):
160 """return file context that is the ancestor of fctx present in ancestor
160 """return file context that is the ancestor of fctx present in ancestor
161 manifest am, stopping after the first ancestor lower than limit"""
161 manifest am, stopping after the first ancestor lower than limit"""
162
162
163 for f in fctx.ancestors():
163 for f in fctx.ancestors():
164 path = f.path()
164 path = f.path()
165 if am.get(path, None) == f.filenode():
165 if am.get(path, None) == f.filenode():
166 return path
166 return path
167 if not f.isintroducedafter(limit):
167 if not f.isintroducedafter(limit):
168 return None
168 return None
169
169
170 def _dirstatecopies(repo, match=None):
170 def _dirstatecopies(repo, match=None):
171 ds = repo.dirstate
171 ds = repo.dirstate
172 c = ds.copies().copy()
172 c = ds.copies().copy()
173 for k in list(c):
173 for k in list(c):
174 if ds[k] not in 'anm' or (match and not match(k)):
174 if ds[k] not in 'anm' or (match and not match(k)):
175 del c[k]
175 del c[k]
176 return c
176 return c
177
177
178 def _computeforwardmissing(a, b, match=None):
178 def _computeforwardmissing(a, b, match=None):
179 """Computes which files are in b but not a.
179 """Computes which files are in b but not a.
180 This is its own function so extensions can easily wrap this call to see what
180 This is its own function so extensions can easily wrap this call to see what
181 files _forwardcopies is about to process.
181 files _forwardcopies is about to process.
182 """
182 """
183 ma = a.manifest()
183 ma = a.manifest()
184 mb = b.manifest()
184 mb = b.manifest()
185 return mb.filesnotin(ma, match=match)
185 return mb.filesnotin(ma, match=match)
186
186
187 def usechangesetcentricalgo(repo):
187 def usechangesetcentricalgo(repo):
188 """Checks if we should use changeset-centric copy algorithms"""
188 """Checks if we should use changeset-centric copy algorithms"""
189 return (repo.ui.config('experimental', 'copies.read-from') in
189 return (repo.ui.config('experimental', 'copies.read-from') in
190 ('changeset-only', 'compatibility'))
190 ('changeset-only', 'compatibility'))
191
191
192 def _committedforwardcopies(a, b, match):
192 def _committedforwardcopies(a, b, match):
193 """Like _forwardcopies(), but b.rev() cannot be None (working copy)"""
193 """Like _forwardcopies(), but b.rev() cannot be None (working copy)"""
194 # files might have to be traced back to the fctx parent of the last
194 # files might have to be traced back to the fctx parent of the last
195 # one-side-only changeset, but not further back than that
195 # one-side-only changeset, but not further back than that
196 repo = a._repo
196 repo = a._repo
197
197
198 if usechangesetcentricalgo(repo):
198 if usechangesetcentricalgo(repo):
199 return _changesetforwardcopies(a, b, match)
199 return _changesetforwardcopies(a, b, match)
200
200
201 debug = repo.ui.debugflag and repo.ui.configbool('devel', 'debug.copies')
201 debug = repo.ui.debugflag and repo.ui.configbool('devel', 'debug.copies')
202 dbg = repo.ui.debug
202 dbg = repo.ui.debug
203 if debug:
203 if debug:
204 dbg('debug.copies: looking into rename from %s to %s\n'
204 dbg('debug.copies: looking into rename from %s to %s\n'
205 % (a, b))
205 % (a, b))
206 limit = _findlimit(repo, a, b)
206 limit = _findlimit(repo, a, b)
207 if debug:
207 if debug:
208 dbg('debug.copies: search limit: %d\n' % limit)
208 dbg('debug.copies: search limit: %d\n' % limit)
209 am = a.manifest()
209 am = a.manifest()
210
210
211 # find where new files came from
211 # find where new files came from
212 # we currently don't try to find where old files went, too expensive
212 # we currently don't try to find where old files went, too expensive
213 # this means we can miss a case like 'hg rm b; hg cp a b'
213 # this means we can miss a case like 'hg rm b; hg cp a b'
214 cm = {}
214 cm = {}
215
215
216 # Computing the forward missing is quite expensive on large manifests, since
216 # Computing the forward missing is quite expensive on large manifests, since
217 # it compares the entire manifests. We can optimize it in the common use
217 # it compares the entire manifests. We can optimize it in the common use
218 # case of computing what copies are in a commit versus its parent (like
218 # case of computing what copies are in a commit versus its parent (like
219 # during a rebase or histedit). Note, we exclude merge commits from this
219 # during a rebase or histedit). Note, we exclude merge commits from this
220 # optimization, since the ctx.files() for a merge commit is not correct for
220 # optimization, since the ctx.files() for a merge commit is not correct for
221 # this comparison.
221 # this comparison.
222 forwardmissingmatch = match
222 forwardmissingmatch = match
223 if b.p1() == a and b.p2().node() == node.nullid:
223 if b.p1() == a and b.p2().node() == node.nullid:
224 filesmatcher = matchmod.exact(b.files())
224 filesmatcher = matchmod.exact(b.files())
225 forwardmissingmatch = matchmod.intersectmatchers(match, filesmatcher)
225 forwardmissingmatch = matchmod.intersectmatchers(match, filesmatcher)
226 missing = _computeforwardmissing(a, b, match=forwardmissingmatch)
226 missing = _computeforwardmissing(a, b, match=forwardmissingmatch)
227
227
228 ancestrycontext = a._repo.changelog.ancestors([b.rev()], inclusive=True)
228 ancestrycontext = a._repo.changelog.ancestors([b.rev()], inclusive=True)
229
229
230 if debug:
230 if debug:
231 dbg('debug.copies: missing files to search: %d\n' % len(missing))
231 dbg('debug.copies: missing files to search: %d\n' % len(missing))
232
232
233 for f in sorted(missing):
233 for f in sorted(missing):
234 if debug:
234 if debug:
235 dbg('debug.copies: tracing file: %s\n' % f)
235 dbg('debug.copies: tracing file: %s\n' % f)
236 fctx = b[f]
236 fctx = b[f]
237 fctx._ancestrycontext = ancestrycontext
237 fctx._ancestrycontext = ancestrycontext
238
238
239 if debug:
239 if debug:
240 start = util.timer()
240 start = util.timer()
241 opath = _tracefile(fctx, am, limit)
241 opath = _tracefile(fctx, am, limit)
242 if opath:
242 if opath:
243 if debug:
243 if debug:
244 dbg('debug.copies: rename of: %s\n' % opath)
244 dbg('debug.copies: rename of: %s\n' % opath)
245 cm[f] = opath
245 cm[f] = opath
246 if debug:
246 if debug:
247 dbg('debug.copies: time: %f seconds\n'
247 dbg('debug.copies: time: %f seconds\n'
248 % (util.timer() - start))
248 % (util.timer() - start))
249 return cm
249 return cm
250
250
251 def _changesetforwardcopies(a, b, match):
251 def _changesetforwardcopies(a, b, match):
252 if a.rev() == node.nullrev:
252 if a.rev() == node.nullrev:
253 return {}
253 return {}
254
254
255 repo = a.repo()
255 repo = a.repo()
256 children = {}
256 children = {}
257 cl = repo.changelog
257 cl = repo.changelog
258 missingrevs = cl.findmissingrevs(common=[a.rev()], heads=[b.rev()])
258 missingrevs = cl.findmissingrevs(common=[a.rev()], heads=[b.rev()])
259 for r in missingrevs:
259 for r in missingrevs:
260 for p in cl.parentrevs(r):
260 for p in cl.parentrevs(r):
261 if p == node.nullrev:
261 if p == node.nullrev:
262 continue
262 continue
263 if p not in children:
263 if p not in children:
264 children[p] = [r]
264 children[p] = [r]
265 else:
265 else:
266 children[p].append(r)
266 children[p].append(r)
267
267
268 roots = set(children) - set(missingrevs)
268 roots = set(children) - set(missingrevs)
269 # 'work' contains 3-tuples of a (revision number, parent number, copies).
269 # 'work' contains 3-tuples of a (revision number, parent number, copies).
270 # The parent number is only used for knowing which parent the copies dict
270 # The parent number is only used for knowing which parent the copies dict
271 # came from.
271 # came from.
272 # NOTE: To reduce costly copying the 'copies' dicts, we reuse the same
272 # NOTE: To reduce costly copying the 'copies' dicts, we reuse the same
273 # instance for *one* of the child nodes (the last one). Once an instance
273 # instance for *one* of the child nodes (the last one). Once an instance
274 # has been put on the queue, it is thus no longer safe to modify it.
274 # has been put on the queue, it is thus no longer safe to modify it.
275 # Conversely, it *is* safe to modify an instance popped off the queue.
275 # Conversely, it *is* safe to modify an instance popped off the queue.
276 work = [(r, 1, {}) for r in roots]
276 work = [(r, 1, {}) for r in roots]
277 heapq.heapify(work)
277 heapq.heapify(work)
278 alwaysmatch = match.always()
278 alwaysmatch = match.always()
279 while work:
279 while work:
280 r, i1, copies = heapq.heappop(work)
280 r, i1, copies = heapq.heappop(work)
281 if work and work[0][0] == r:
281 if work and work[0][0] == r:
282 # We are tracing copies from both parents
282 # We are tracing copies from both parents
283 r, i2, copies2 = heapq.heappop(work)
283 r, i2, copies2 = heapq.heappop(work)
284 for dst, src in copies2.items():
284 for dst, src in copies2.items():
285 # Unlike when copies are stored in the filelog, we consider
285 # Unlike when copies are stored in the filelog, we consider
286 # it a copy even if the destination already existed on the
286 # it a copy even if the destination already existed on the
287 # other branch. It's simply too expensive to check if the
287 # other branch. It's simply too expensive to check if the
288 # file existed in the manifest.
288 # file existed in the manifest.
289 if dst not in copies:
289 if dst not in copies:
290 # If it was copied on the p1 side, leave it as copied from
290 # If it was copied on the p1 side, leave it as copied from
291 # that side, even if it was also copied on the p2 side.
291 # that side, even if it was also copied on the p2 side.
292 copies[dst] = copies2[dst]
292 copies[dst] = copies2[dst]
293 if r == b.rev():
293 if r == b.rev():
294 _filter(a, b, copies)
294 _filter(a, b, copies)
295 return copies
295 return copies
296 for i, c in enumerate(children[r]):
296 for i, c in enumerate(children[r]):
297 childctx = repo[c]
297 childctx = repo[c]
298 if r == childctx.p1().rev():
298 if r == childctx.p1().rev():
299 parent = 1
299 parent = 1
300 childcopies = childctx.p1copies()
300 childcopies = childctx.p1copies()
301 else:
301 else:
302 assert r == childctx.p2().rev()
302 assert r == childctx.p2().rev()
303 parent = 2
303 parent = 2
304 childcopies = childctx.p2copies()
304 childcopies = childctx.p2copies()
305 if not alwaysmatch:
305 if not alwaysmatch:
306 childcopies = {dst: src for dst, src in childcopies.items()
306 childcopies = {dst: src for dst, src in childcopies.items()
307 if match(dst)}
307 if match(dst)}
308 # Copy the dict only if later iterations will also need it
308 # Copy the dict only if later iterations will also need it
309 if i != len(children[r]) - 1:
309 if i != len(children[r]) - 1:
310 newcopies = copies.copy()
310 newcopies = copies.copy()
311 else:
311 else:
312 newcopies = copies
312 newcopies = copies
313 if childcopies:
313 if childcopies:
314 newcopies = _chain(newcopies, childcopies)
314 newcopies = _chain(newcopies, childcopies)
315 for f in childctx.filesremoved():
315 for f in childctx.filesremoved():
316 if f in newcopies:
316 if f in newcopies:
317 del newcopies[f]
317 del newcopies[f]
318 heapq.heappush(work, (c, parent, newcopies))
318 heapq.heappush(work, (c, parent, newcopies))
319 assert False
319 assert False
320
320
321 def _forwardcopies(a, b, match=None):
321 def _forwardcopies(a, b, match=None):
322 """find {dst@b: src@a} copy mapping where a is an ancestor of b"""
322 """find {dst@b: src@a} copy mapping where a is an ancestor of b"""
323
323
324 match = a.repo().narrowmatch(match)
324 match = a.repo().narrowmatch(match)
325 # check for working copy
325 # check for working copy
326 if b.rev() is None:
326 if b.rev() is None:
327 if a == b.p1():
328 # short-circuit to avoid issues with merge states
329 return _dirstatecopies(b._repo, match)
330
331 cm = _committedforwardcopies(a, b.p1(), match)
327 cm = _committedforwardcopies(a, b.p1(), match)
332 # combine copies from dirstate if necessary
328 # combine copies from dirstate if necessary
333 return _chainandfilter(a, b, cm, _dirstatecopies(b._repo, match))
329 return _chainandfilter(a, b, cm, _dirstatecopies(b._repo, match))
334 return _committedforwardcopies(a, b, match)
330 return _committedforwardcopies(a, b, match)
335
331
336 def _backwardrenames(a, b, match):
332 def _backwardrenames(a, b, match):
337 if a._repo.ui.config('experimental', 'copytrace') == 'off':
333 if a._repo.ui.config('experimental', 'copytrace') == 'off':
338 return {}
334 return {}
339
335
340 # Even though we're not taking copies into account, 1:n rename situations
336 # Even though we're not taking copies into account, 1:n rename situations
341 # can still exist (e.g. hg cp a b; hg mv a c). In those cases we
337 # can still exist (e.g. hg cp a b; hg mv a c). In those cases we
342 # arbitrarily pick one of the renames.
338 # arbitrarily pick one of the renames.
343 # We don't want to pass in "match" here, since that would filter
339 # We don't want to pass in "match" here, since that would filter
344 # the destination by it. Since we're reversing the copies, we want
340 # the destination by it. Since we're reversing the copies, we want
345 # to filter the source instead.
341 # to filter the source instead.
346 f = _forwardcopies(b, a)
342 f = _forwardcopies(b, a)
347 r = {}
343 r = {}
348 for k, v in sorted(f.iteritems()):
344 for k, v in sorted(f.iteritems()):
349 if match and not match(v):
345 if match and not match(v):
350 continue
346 continue
351 # remove copies
347 # remove copies
352 if v in a:
348 if v in a:
353 continue
349 continue
354 r[v] = k
350 r[v] = k
355 return r
351 return r
356
352
357 def pathcopies(x, y, match=None):
353 def pathcopies(x, y, match=None):
358 """find {dst@y: src@x} copy mapping for directed compare"""
354 """find {dst@y: src@x} copy mapping for directed compare"""
359 repo = x._repo
355 repo = x._repo
360 debug = repo.ui.debugflag and repo.ui.configbool('devel', 'debug.copies')
356 debug = repo.ui.debugflag and repo.ui.configbool('devel', 'debug.copies')
361 if debug:
357 if debug:
362 repo.ui.debug('debug.copies: searching copies from %s to %s\n'
358 repo.ui.debug('debug.copies: searching copies from %s to %s\n'
363 % (x, y))
359 % (x, y))
364 if x == y or not x or not y:
360 if x == y or not x or not y:
365 return {}
361 return {}
366 a = y.ancestor(x)
362 a = y.ancestor(x)
367 if a == x:
363 if a == x:
368 if debug:
364 if debug:
369 repo.ui.debug('debug.copies: search mode: forward\n')
365 repo.ui.debug('debug.copies: search mode: forward\n')
366 if y.rev() is None and x == y.p1():
367 # short-circuit to avoid issues with merge states
368 return _dirstatecopies(repo, match)
370 return _forwardcopies(x, y, match=match)
369 return _forwardcopies(x, y, match=match)
371 if a == y:
370 if a == y:
372 if debug:
371 if debug:
373 repo.ui.debug('debug.copies: search mode: backward\n')
372 repo.ui.debug('debug.copies: search mode: backward\n')
374 return _backwardrenames(x, y, match=match)
373 return _backwardrenames(x, y, match=match)
375 if debug:
374 if debug:
376 repo.ui.debug('debug.copies: search mode: combined\n')
375 repo.ui.debug('debug.copies: search mode: combined\n')
377 return _chainandfilter(x, y, _backwardrenames(x, a, match=match),
376 return _chainandfilter(x, y, _backwardrenames(x, a, match=match),
378 _forwardcopies(a, y, match=match))
377 _forwardcopies(a, y, match=match))
379
378
380 def mergecopies(repo, c1, c2, base):
379 def mergecopies(repo, c1, c2, base):
381 """
380 """
382 Finds moves and copies between context c1 and c2 that are relevant for
381 Finds moves and copies between context c1 and c2 that are relevant for
383 merging. 'base' will be used as the merge base.
382 merging. 'base' will be used as the merge base.
384
383
385 Copytracing is used in commands like rebase, merge, unshelve, etc to merge
384 Copytracing is used in commands like rebase, merge, unshelve, etc to merge
386 files that were moved/ copied in one merge parent and modified in another.
385 files that were moved/ copied in one merge parent and modified in another.
387 For example:
386 For example:
388
387
389 o ---> 4 another commit
388 o ---> 4 another commit
390 |
389 |
391 | o ---> 3 commit that modifies a.txt
390 | o ---> 3 commit that modifies a.txt
392 | /
391 | /
393 o / ---> 2 commit that moves a.txt to b.txt
392 o / ---> 2 commit that moves a.txt to b.txt
394 |/
393 |/
395 o ---> 1 merge base
394 o ---> 1 merge base
396
395
397 If we try to rebase revision 3 on revision 4, since there is no a.txt in
396 If we try to rebase revision 3 on revision 4, since there is no a.txt in
398 revision 4, and if user have copytrace disabled, we prints the following
397 revision 4, and if user have copytrace disabled, we prints the following
399 message:
398 message:
400
399
401 ```other changed <file> which local deleted```
400 ```other changed <file> which local deleted```
402
401
403 Returns five dicts: "copy", "movewithdir", "diverge", "renamedelete" and
402 Returns five dicts: "copy", "movewithdir", "diverge", "renamedelete" and
404 "dirmove".
403 "dirmove".
405
404
406 "copy" is a mapping from destination name -> source name,
405 "copy" is a mapping from destination name -> source name,
407 where source is in c1 and destination is in c2 or vice-versa.
406 where source is in c1 and destination is in c2 or vice-versa.
408
407
409 "movewithdir" is a mapping from source name -> destination name,
408 "movewithdir" is a mapping from source name -> destination name,
410 where the file at source present in one context but not the other
409 where the file at source present in one context but not the other
411 needs to be moved to destination by the merge process, because the
410 needs to be moved to destination by the merge process, because the
412 other context moved the directory it is in.
411 other context moved the directory it is in.
413
412
414 "diverge" is a mapping of source name -> list of destination names
413 "diverge" is a mapping of source name -> list of destination names
415 for divergent renames.
414 for divergent renames.
416
415
417 "renamedelete" is a mapping of source name -> list of destination
416 "renamedelete" is a mapping of source name -> list of destination
418 names for files deleted in c1 that were renamed in c2 or vice-versa.
417 names for files deleted in c1 that were renamed in c2 or vice-versa.
419
418
420 "dirmove" is a mapping of detected source dir -> destination dir renames.
419 "dirmove" is a mapping of detected source dir -> destination dir renames.
421 This is needed for handling changes to new files previously grafted into
420 This is needed for handling changes to new files previously grafted into
422 renamed directories.
421 renamed directories.
423
422
424 This function calls different copytracing algorithms based on config.
423 This function calls different copytracing algorithms based on config.
425 """
424 """
426 # avoid silly behavior for update from empty dir
425 # avoid silly behavior for update from empty dir
427 if not c1 or not c2 or c1 == c2:
426 if not c1 or not c2 or c1 == c2:
428 return {}, {}, {}, {}, {}
427 return {}, {}, {}, {}, {}
429
428
430 narrowmatch = c1.repo().narrowmatch()
429 narrowmatch = c1.repo().narrowmatch()
431
430
432 # avoid silly behavior for parent -> working dir
431 # avoid silly behavior for parent -> working dir
433 if c2.node() is None and c1.node() == repo.dirstate.p1():
432 if c2.node() is None and c1.node() == repo.dirstate.p1():
434 return _dirstatecopies(repo, narrowmatch), {}, {}, {}, {}
433 return _dirstatecopies(repo, narrowmatch), {}, {}, {}, {}
435
434
436 copytracing = repo.ui.config('experimental', 'copytrace')
435 copytracing = repo.ui.config('experimental', 'copytrace')
437 if stringutil.parsebool(copytracing) is False:
436 if stringutil.parsebool(copytracing) is False:
438 # stringutil.parsebool() returns None when it is unable to parse the
437 # stringutil.parsebool() returns None when it is unable to parse the
439 # value, so we should rely on making sure copytracing is on such cases
438 # value, so we should rely on making sure copytracing is on such cases
440 return {}, {}, {}, {}, {}
439 return {}, {}, {}, {}, {}
441
440
442 if usechangesetcentricalgo(repo):
441 if usechangesetcentricalgo(repo):
443 # The heuristics don't make sense when we need changeset-centric algos
442 # The heuristics don't make sense when we need changeset-centric algos
444 return _fullcopytracing(repo, c1, c2, base)
443 return _fullcopytracing(repo, c1, c2, base)
445
444
446 # Copy trace disabling is explicitly below the node == p1 logic above
445 # Copy trace disabling is explicitly below the node == p1 logic above
447 # because the logic above is required for a simple copy to be kept across a
446 # because the logic above is required for a simple copy to be kept across a
448 # rebase.
447 # rebase.
449 if copytracing == 'heuristics':
448 if copytracing == 'heuristics':
450 # Do full copytracing if only non-public revisions are involved as
449 # Do full copytracing if only non-public revisions are involved as
451 # that will be fast enough and will also cover the copies which could
450 # that will be fast enough and will also cover the copies which could
452 # be missed by heuristics
451 # be missed by heuristics
453 if _isfullcopytraceable(repo, c1, base):
452 if _isfullcopytraceable(repo, c1, base):
454 return _fullcopytracing(repo, c1, c2, base)
453 return _fullcopytracing(repo, c1, c2, base)
455 return _heuristicscopytracing(repo, c1, c2, base)
454 return _heuristicscopytracing(repo, c1, c2, base)
456 else:
455 else:
457 return _fullcopytracing(repo, c1, c2, base)
456 return _fullcopytracing(repo, c1, c2, base)
458
457
459 def _isfullcopytraceable(repo, c1, base):
458 def _isfullcopytraceable(repo, c1, base):
460 """ Checks that if base, source and destination are all no-public branches,
459 """ Checks that if base, source and destination are all no-public branches,
461 if yes let's use the full copytrace algorithm for increased capabilities
460 if yes let's use the full copytrace algorithm for increased capabilities
462 since it will be fast enough.
461 since it will be fast enough.
463
462
464 `experimental.copytrace.sourcecommitlimit` can be used to set a limit for
463 `experimental.copytrace.sourcecommitlimit` can be used to set a limit for
465 number of changesets from c1 to base such that if number of changesets are
464 number of changesets from c1 to base such that if number of changesets are
466 more than the limit, full copytracing algorithm won't be used.
465 more than the limit, full copytracing algorithm won't be used.
467 """
466 """
468 if c1.rev() is None:
467 if c1.rev() is None:
469 c1 = c1.p1()
468 c1 = c1.p1()
470 if c1.mutable() and base.mutable():
469 if c1.mutable() and base.mutable():
471 sourcecommitlimit = repo.ui.configint('experimental',
470 sourcecommitlimit = repo.ui.configint('experimental',
472 'copytrace.sourcecommitlimit')
471 'copytrace.sourcecommitlimit')
473 commits = len(repo.revs('%d::%d', base.rev(), c1.rev()))
472 commits = len(repo.revs('%d::%d', base.rev(), c1.rev()))
474 return commits < sourcecommitlimit
473 return commits < sourcecommitlimit
475 return False
474 return False
476
475
477 def _checksinglesidecopies(src, dsts1, m1, m2, mb, c2, base,
476 def _checksinglesidecopies(src, dsts1, m1, m2, mb, c2, base,
478 copy, renamedelete):
477 copy, renamedelete):
479 if src not in m2:
478 if src not in m2:
480 # deleted on side 2
479 # deleted on side 2
481 if src not in m1:
480 if src not in m1:
482 # renamed on side 1, deleted on side 2
481 # renamed on side 1, deleted on side 2
483 renamedelete[src] = dsts1
482 renamedelete[src] = dsts1
484 elif m2[src] != mb[src]:
483 elif m2[src] != mb[src]:
485 if not _related(c2[src], base[src]):
484 if not _related(c2[src], base[src]):
486 return
485 return
487 # modified on side 2
486 # modified on side 2
488 for dst in dsts1:
487 for dst in dsts1:
489 if dst not in m2:
488 if dst not in m2:
490 # dst not added on side 2 (handle as regular
489 # dst not added on side 2 (handle as regular
491 # "both created" case in manifestmerge otherwise)
490 # "both created" case in manifestmerge otherwise)
492 copy[dst] = src
491 copy[dst] = src
493
492
494 def _fullcopytracing(repo, c1, c2, base):
493 def _fullcopytracing(repo, c1, c2, base):
495 """ The full copytracing algorithm which finds all the new files that were
494 """ The full copytracing algorithm which finds all the new files that were
496 added from merge base up to the top commit and for each file it checks if
495 added from merge base up to the top commit and for each file it checks if
497 this file was copied from another file.
496 this file was copied from another file.
498
497
499 This is pretty slow when a lot of changesets are involved but will track all
498 This is pretty slow when a lot of changesets are involved but will track all
500 the copies.
499 the copies.
501 """
500 """
502 m1 = c1.manifest()
501 m1 = c1.manifest()
503 m2 = c2.manifest()
502 m2 = c2.manifest()
504 mb = base.manifest()
503 mb = base.manifest()
505
504
506 copies1 = pathcopies(base, c1)
505 copies1 = pathcopies(base, c1)
507 copies2 = pathcopies(base, c2)
506 copies2 = pathcopies(base, c2)
508
507
509 inversecopies1 = {}
508 inversecopies1 = {}
510 inversecopies2 = {}
509 inversecopies2 = {}
511 for dst, src in copies1.items():
510 for dst, src in copies1.items():
512 inversecopies1.setdefault(src, []).append(dst)
511 inversecopies1.setdefault(src, []).append(dst)
513 for dst, src in copies2.items():
512 for dst, src in copies2.items():
514 inversecopies2.setdefault(src, []).append(dst)
513 inversecopies2.setdefault(src, []).append(dst)
515
514
516 copy = {}
515 copy = {}
517 diverge = {}
516 diverge = {}
518 renamedelete = {}
517 renamedelete = {}
519 allsources = set(inversecopies1) | set(inversecopies2)
518 allsources = set(inversecopies1) | set(inversecopies2)
520 for src in allsources:
519 for src in allsources:
521 dsts1 = inversecopies1.get(src)
520 dsts1 = inversecopies1.get(src)
522 dsts2 = inversecopies2.get(src)
521 dsts2 = inversecopies2.get(src)
523 if dsts1 and dsts2:
522 if dsts1 and dsts2:
524 # copied/renamed on both sides
523 # copied/renamed on both sides
525 if src not in m1 and src not in m2:
524 if src not in m1 and src not in m2:
526 # renamed on both sides
525 # renamed on both sides
527 dsts1 = set(dsts1)
526 dsts1 = set(dsts1)
528 dsts2 = set(dsts2)
527 dsts2 = set(dsts2)
529 # If there's some overlap in the rename destinations, we
528 # If there's some overlap in the rename destinations, we
530 # consider it not divergent. For example, if side 1 copies 'a'
529 # consider it not divergent. For example, if side 1 copies 'a'
531 # to 'b' and 'c' and deletes 'a', and side 2 copies 'a' to 'c'
530 # to 'b' and 'c' and deletes 'a', and side 2 copies 'a' to 'c'
532 # and 'd' and deletes 'a'.
531 # and 'd' and deletes 'a'.
533 if dsts1 & dsts2:
532 if dsts1 & dsts2:
534 for dst in (dsts1 & dsts2):
533 for dst in (dsts1 & dsts2):
535 copy[dst] = src
534 copy[dst] = src
536 else:
535 else:
537 diverge[src] = sorted(dsts1 | dsts2)
536 diverge[src] = sorted(dsts1 | dsts2)
538 elif src in m1 and src in m2:
537 elif src in m1 and src in m2:
539 # copied on both sides
538 # copied on both sides
540 dsts1 = set(dsts1)
539 dsts1 = set(dsts1)
541 dsts2 = set(dsts2)
540 dsts2 = set(dsts2)
542 for dst in (dsts1 & dsts2):
541 for dst in (dsts1 & dsts2):
543 copy[dst] = src
542 copy[dst] = src
544 # TODO: Handle cases where it was renamed on one side and copied
543 # TODO: Handle cases where it was renamed on one side and copied
545 # on the other side
544 # on the other side
546 elif dsts1:
545 elif dsts1:
547 # copied/renamed only on side 1
546 # copied/renamed only on side 1
548 _checksinglesidecopies(src, dsts1, m1, m2, mb, c2, base,
547 _checksinglesidecopies(src, dsts1, m1, m2, mb, c2, base,
549 copy, renamedelete)
548 copy, renamedelete)
550 elif dsts2:
549 elif dsts2:
551 # copied/renamed only on side 2
550 # copied/renamed only on side 2
552 _checksinglesidecopies(src, dsts2, m2, m1, mb, c1, base,
551 _checksinglesidecopies(src, dsts2, m2, m1, mb, c1, base,
553 copy, renamedelete)
552 copy, renamedelete)
554
553
555 renamedeleteset = set()
554 renamedeleteset = set()
556 divergeset = set()
555 divergeset = set()
557 for dsts in diverge.values():
556 for dsts in diverge.values():
558 divergeset.update(dsts)
557 divergeset.update(dsts)
559 for dsts in renamedelete.values():
558 for dsts in renamedelete.values():
560 renamedeleteset.update(dsts)
559 renamedeleteset.update(dsts)
561
560
562 # find interesting file sets from manifests
561 # find interesting file sets from manifests
563 addedinm1 = m1.filesnotin(mb, repo.narrowmatch())
562 addedinm1 = m1.filesnotin(mb, repo.narrowmatch())
564 addedinm2 = m2.filesnotin(mb, repo.narrowmatch())
563 addedinm2 = m2.filesnotin(mb, repo.narrowmatch())
565 u1 = sorted(addedinm1 - addedinm2)
564 u1 = sorted(addedinm1 - addedinm2)
566 u2 = sorted(addedinm2 - addedinm1)
565 u2 = sorted(addedinm2 - addedinm1)
567
566
568 header = " unmatched files in %s"
567 header = " unmatched files in %s"
569 if u1:
568 if u1:
570 repo.ui.debug("%s:\n %s\n" % (header % 'local', "\n ".join(u1)))
569 repo.ui.debug("%s:\n %s\n" % (header % 'local', "\n ".join(u1)))
571 if u2:
570 if u2:
572 repo.ui.debug("%s:\n %s\n" % (header % 'other', "\n ".join(u2)))
571 repo.ui.debug("%s:\n %s\n" % (header % 'other', "\n ".join(u2)))
573
572
574 fullcopy = copies1.copy()
573 fullcopy = copies1.copy()
575 fullcopy.update(copies2)
574 fullcopy.update(copies2)
576 if not fullcopy:
575 if not fullcopy:
577 return copy, {}, diverge, renamedelete, {}
576 return copy, {}, diverge, renamedelete, {}
578
577
579 if repo.ui.debugflag:
578 if repo.ui.debugflag:
580 repo.ui.debug(" all copies found (* = to merge, ! = divergent, "
579 repo.ui.debug(" all copies found (* = to merge, ! = divergent, "
581 "% = renamed and deleted):\n")
580 "% = renamed and deleted):\n")
582 for f in sorted(fullcopy):
581 for f in sorted(fullcopy):
583 note = ""
582 note = ""
584 if f in copy:
583 if f in copy:
585 note += "*"
584 note += "*"
586 if f in divergeset:
585 if f in divergeset:
587 note += "!"
586 note += "!"
588 if f in renamedeleteset:
587 if f in renamedeleteset:
589 note += "%"
588 note += "%"
590 repo.ui.debug(" src: '%s' -> dst: '%s' %s\n" % (fullcopy[f], f,
589 repo.ui.debug(" src: '%s' -> dst: '%s' %s\n" % (fullcopy[f], f,
591 note))
590 note))
592 del divergeset
591 del divergeset
593
592
594 repo.ui.debug(" checking for directory renames\n")
593 repo.ui.debug(" checking for directory renames\n")
595
594
596 # generate a directory move map
595 # generate a directory move map
597 d1, d2 = c1.dirs(), c2.dirs()
596 d1, d2 = c1.dirs(), c2.dirs()
598 invalid = set()
597 invalid = set()
599 dirmove = {}
598 dirmove = {}
600
599
601 # examine each file copy for a potential directory move, which is
600 # examine each file copy for a potential directory move, which is
602 # when all the files in a directory are moved to a new directory
601 # when all the files in a directory are moved to a new directory
603 for dst, src in fullcopy.iteritems():
602 for dst, src in fullcopy.iteritems():
604 dsrc, ddst = pathutil.dirname(src), pathutil.dirname(dst)
603 dsrc, ddst = pathutil.dirname(src), pathutil.dirname(dst)
605 if dsrc in invalid:
604 if dsrc in invalid:
606 # already seen to be uninteresting
605 # already seen to be uninteresting
607 continue
606 continue
608 elif dsrc in d1 and ddst in d1:
607 elif dsrc in d1 and ddst in d1:
609 # directory wasn't entirely moved locally
608 # directory wasn't entirely moved locally
610 invalid.add(dsrc)
609 invalid.add(dsrc)
611 elif dsrc in d2 and ddst in d2:
610 elif dsrc in d2 and ddst in d2:
612 # directory wasn't entirely moved remotely
611 # directory wasn't entirely moved remotely
613 invalid.add(dsrc)
612 invalid.add(dsrc)
614 elif dsrc in dirmove and dirmove[dsrc] != ddst:
613 elif dsrc in dirmove and dirmove[dsrc] != ddst:
615 # files from the same directory moved to two different places
614 # files from the same directory moved to two different places
616 invalid.add(dsrc)
615 invalid.add(dsrc)
617 else:
616 else:
618 # looks good so far
617 # looks good so far
619 dirmove[dsrc] = ddst
618 dirmove[dsrc] = ddst
620
619
621 for i in invalid:
620 for i in invalid:
622 if i in dirmove:
621 if i in dirmove:
623 del dirmove[i]
622 del dirmove[i]
624 del d1, d2, invalid
623 del d1, d2, invalid
625
624
626 if not dirmove:
625 if not dirmove:
627 return copy, {}, diverge, renamedelete, {}
626 return copy, {}, diverge, renamedelete, {}
628
627
629 dirmove = {k + "/": v + "/" for k, v in dirmove.iteritems()}
628 dirmove = {k + "/": v + "/" for k, v in dirmove.iteritems()}
630
629
631 for d in dirmove:
630 for d in dirmove:
632 repo.ui.debug(" discovered dir src: '%s' -> dst: '%s'\n" %
631 repo.ui.debug(" discovered dir src: '%s' -> dst: '%s'\n" %
633 (d, dirmove[d]))
632 (d, dirmove[d]))
634
633
635 movewithdir = {}
634 movewithdir = {}
636 # check unaccounted nonoverlapping files against directory moves
635 # check unaccounted nonoverlapping files against directory moves
637 for f in u1 + u2:
636 for f in u1 + u2:
638 if f not in fullcopy:
637 if f not in fullcopy:
639 for d in dirmove:
638 for d in dirmove:
640 if f.startswith(d):
639 if f.startswith(d):
641 # new file added in a directory that was moved, move it
640 # new file added in a directory that was moved, move it
642 df = dirmove[d] + f[len(d):]
641 df = dirmove[d] + f[len(d):]
643 if df not in copy:
642 if df not in copy:
644 movewithdir[f] = df
643 movewithdir[f] = df
645 repo.ui.debug((" pending file src: '%s' -> "
644 repo.ui.debug((" pending file src: '%s' -> "
646 "dst: '%s'\n") % (f, df))
645 "dst: '%s'\n") % (f, df))
647 break
646 break
648
647
649 return copy, movewithdir, diverge, renamedelete, dirmove
648 return copy, movewithdir, diverge, renamedelete, dirmove
650
649
651 def _heuristicscopytracing(repo, c1, c2, base):
650 def _heuristicscopytracing(repo, c1, c2, base):
652 """ Fast copytracing using filename heuristics
651 """ Fast copytracing using filename heuristics
653
652
654 Assumes that moves or renames are of following two types:
653 Assumes that moves or renames are of following two types:
655
654
656 1) Inside a directory only (same directory name but different filenames)
655 1) Inside a directory only (same directory name but different filenames)
657 2) Move from one directory to another
656 2) Move from one directory to another
658 (same filenames but different directory names)
657 (same filenames but different directory names)
659
658
660 Works only when there are no merge commits in the "source branch".
659 Works only when there are no merge commits in the "source branch".
661 Source branch is commits from base up to c2 not including base.
660 Source branch is commits from base up to c2 not including base.
662
661
663 If merge is involved it fallbacks to _fullcopytracing().
662 If merge is involved it fallbacks to _fullcopytracing().
664
663
665 Can be used by setting the following config:
664 Can be used by setting the following config:
666
665
667 [experimental]
666 [experimental]
668 copytrace = heuristics
667 copytrace = heuristics
669
668
670 In some cases the copy/move candidates found by heuristics can be very large
669 In some cases the copy/move candidates found by heuristics can be very large
671 in number and that will make the algorithm slow. The number of possible
670 in number and that will make the algorithm slow. The number of possible
672 candidates to check can be limited by using the config
671 candidates to check can be limited by using the config
673 `experimental.copytrace.movecandidateslimit` which defaults to 100.
672 `experimental.copytrace.movecandidateslimit` which defaults to 100.
674 """
673 """
675
674
676 if c1.rev() is None:
675 if c1.rev() is None:
677 c1 = c1.p1()
676 c1 = c1.p1()
678 if c2.rev() is None:
677 if c2.rev() is None:
679 c2 = c2.p1()
678 c2 = c2.p1()
680
679
681 copies = {}
680 copies = {}
682
681
683 changedfiles = set()
682 changedfiles = set()
684 m1 = c1.manifest()
683 m1 = c1.manifest()
685 if not repo.revs('%d::%d', base.rev(), c2.rev()):
684 if not repo.revs('%d::%d', base.rev(), c2.rev()):
686 # If base is not in c2 branch, we switch to fullcopytracing
685 # If base is not in c2 branch, we switch to fullcopytracing
687 repo.ui.debug("switching to full copytracing as base is not "
686 repo.ui.debug("switching to full copytracing as base is not "
688 "an ancestor of c2\n")
687 "an ancestor of c2\n")
689 return _fullcopytracing(repo, c1, c2, base)
688 return _fullcopytracing(repo, c1, c2, base)
690
689
691 ctx = c2
690 ctx = c2
692 while ctx != base:
691 while ctx != base:
693 if len(ctx.parents()) == 2:
692 if len(ctx.parents()) == 2:
694 # To keep things simple let's not handle merges
693 # To keep things simple let's not handle merges
695 repo.ui.debug("switching to full copytracing because of merges\n")
694 repo.ui.debug("switching to full copytracing because of merges\n")
696 return _fullcopytracing(repo, c1, c2, base)
695 return _fullcopytracing(repo, c1, c2, base)
697 changedfiles.update(ctx.files())
696 changedfiles.update(ctx.files())
698 ctx = ctx.p1()
697 ctx = ctx.p1()
699
698
700 cp = _forwardcopies(base, c2)
699 cp = _forwardcopies(base, c2)
701 for dst, src in cp.iteritems():
700 for dst, src in cp.iteritems():
702 if src in m1:
701 if src in m1:
703 copies[dst] = src
702 copies[dst] = src
704
703
705 # file is missing if it isn't present in the destination, but is present in
704 # file is missing if it isn't present in the destination, but is present in
706 # the base and present in the source.
705 # the base and present in the source.
707 # Presence in the base is important to exclude added files, presence in the
706 # Presence in the base is important to exclude added files, presence in the
708 # source is important to exclude removed files.
707 # source is important to exclude removed files.
709 filt = lambda f: f not in m1 and f in base and f in c2
708 filt = lambda f: f not in m1 and f in base and f in c2
710 missingfiles = [f for f in changedfiles if filt(f)]
709 missingfiles = [f for f in changedfiles if filt(f)]
711
710
712 if missingfiles:
711 if missingfiles:
713 basenametofilename = collections.defaultdict(list)
712 basenametofilename = collections.defaultdict(list)
714 dirnametofilename = collections.defaultdict(list)
713 dirnametofilename = collections.defaultdict(list)
715
714
716 for f in m1.filesnotin(base.manifest()):
715 for f in m1.filesnotin(base.manifest()):
717 basename = os.path.basename(f)
716 basename = os.path.basename(f)
718 dirname = os.path.dirname(f)
717 dirname = os.path.dirname(f)
719 basenametofilename[basename].append(f)
718 basenametofilename[basename].append(f)
720 dirnametofilename[dirname].append(f)
719 dirnametofilename[dirname].append(f)
721
720
722 for f in missingfiles:
721 for f in missingfiles:
723 basename = os.path.basename(f)
722 basename = os.path.basename(f)
724 dirname = os.path.dirname(f)
723 dirname = os.path.dirname(f)
725 samebasename = basenametofilename[basename]
724 samebasename = basenametofilename[basename]
726 samedirname = dirnametofilename[dirname]
725 samedirname = dirnametofilename[dirname]
727 movecandidates = samebasename + samedirname
726 movecandidates = samebasename + samedirname
728 # f is guaranteed to be present in c2, that's why
727 # f is guaranteed to be present in c2, that's why
729 # c2.filectx(f) won't fail
728 # c2.filectx(f) won't fail
730 f2 = c2.filectx(f)
729 f2 = c2.filectx(f)
731 # we can have a lot of candidates which can slow down the heuristics
730 # we can have a lot of candidates which can slow down the heuristics
732 # config value to limit the number of candidates moves to check
731 # config value to limit the number of candidates moves to check
733 maxcandidates = repo.ui.configint('experimental',
732 maxcandidates = repo.ui.configint('experimental',
734 'copytrace.movecandidateslimit')
733 'copytrace.movecandidateslimit')
735
734
736 if len(movecandidates) > maxcandidates:
735 if len(movecandidates) > maxcandidates:
737 repo.ui.status(_("skipping copytracing for '%s', more "
736 repo.ui.status(_("skipping copytracing for '%s', more "
738 "candidates than the limit: %d\n")
737 "candidates than the limit: %d\n")
739 % (f, len(movecandidates)))
738 % (f, len(movecandidates)))
740 continue
739 continue
741
740
742 for candidate in movecandidates:
741 for candidate in movecandidates:
743 f1 = c1.filectx(candidate)
742 f1 = c1.filectx(candidate)
744 if _related(f1, f2):
743 if _related(f1, f2):
745 # if there are a few related copies then we'll merge
744 # if there are a few related copies then we'll merge
746 # changes into all of them. This matches the behaviour
745 # changes into all of them. This matches the behaviour
747 # of upstream copytracing
746 # of upstream copytracing
748 copies[candidate] = f
747 copies[candidate] = f
749
748
750 return copies, {}, {}, {}, {}
749 return copies, {}, {}, {}, {}
751
750
752 def _related(f1, f2):
751 def _related(f1, f2):
753 """return True if f1 and f2 filectx have a common ancestor
752 """return True if f1 and f2 filectx have a common ancestor
754
753
755 Walk back to common ancestor to see if the two files originate
754 Walk back to common ancestor to see if the two files originate
756 from the same file. Since workingfilectx's rev() is None it messes
755 from the same file. Since workingfilectx's rev() is None it messes
757 up the integer comparison logic, hence the pre-step check for
756 up the integer comparison logic, hence the pre-step check for
758 None (f1 and f2 can only be workingfilectx's initially).
757 None (f1 and f2 can only be workingfilectx's initially).
759 """
758 """
760
759
761 if f1 == f2:
760 if f1 == f2:
762 return True # a match
761 return True # a match
763
762
764 g1, g2 = f1.ancestors(), f2.ancestors()
763 g1, g2 = f1.ancestors(), f2.ancestors()
765 try:
764 try:
766 f1r, f2r = f1.linkrev(), f2.linkrev()
765 f1r, f2r = f1.linkrev(), f2.linkrev()
767
766
768 if f1r is None:
767 if f1r is None:
769 f1 = next(g1)
768 f1 = next(g1)
770 if f2r is None:
769 if f2r is None:
771 f2 = next(g2)
770 f2 = next(g2)
772
771
773 while True:
772 while True:
774 f1r, f2r = f1.linkrev(), f2.linkrev()
773 f1r, f2r = f1.linkrev(), f2.linkrev()
775 if f1r > f2r:
774 if f1r > f2r:
776 f1 = next(g1)
775 f1 = next(g1)
777 elif f2r > f1r:
776 elif f2r > f1r:
778 f2 = next(g2)
777 f2 = next(g2)
779 else: # f1 and f2 point to files in the same linkrev
778 else: # f1 and f2 point to files in the same linkrev
780 return f1 == f2 # true if they point to the same file
779 return f1 == f2 # true if they point to the same file
781 except StopIteration:
780 except StopIteration:
782 return False
781 return False
783
782
784 def duplicatecopies(repo, wctx, rev, fromrev, skiprev=None):
783 def duplicatecopies(repo, wctx, rev, fromrev, skiprev=None):
785 """reproduce copies from fromrev to rev in the dirstate
784 """reproduce copies from fromrev to rev in the dirstate
786
785
787 If skiprev is specified, it's a revision that should be used to
786 If skiprev is specified, it's a revision that should be used to
788 filter copy records. Any copies that occur between fromrev and
787 filter copy records. Any copies that occur between fromrev and
789 skiprev will not be duplicated, even if they appear in the set of
788 skiprev will not be duplicated, even if they appear in the set of
790 copies between fromrev and rev.
789 copies between fromrev and rev.
791 """
790 """
792 exclude = {}
791 exclude = {}
793 ctraceconfig = repo.ui.config('experimental', 'copytrace')
792 ctraceconfig = repo.ui.config('experimental', 'copytrace')
794 bctrace = stringutil.parsebool(ctraceconfig)
793 bctrace = stringutil.parsebool(ctraceconfig)
795 if (skiprev is not None and
794 if (skiprev is not None and
796 (ctraceconfig == 'heuristics' or bctrace or bctrace is None)):
795 (ctraceconfig == 'heuristics' or bctrace or bctrace is None)):
797 # copytrace='off' skips this line, but not the entire function because
796 # copytrace='off' skips this line, but not the entire function because
798 # the line below is O(size of the repo) during a rebase, while the rest
797 # the line below is O(size of the repo) during a rebase, while the rest
799 # of the function is much faster (and is required for carrying copy
798 # of the function is much faster (and is required for carrying copy
800 # metadata across the rebase anyway).
799 # metadata across the rebase anyway).
801 exclude = pathcopies(repo[fromrev], repo[skiprev])
800 exclude = pathcopies(repo[fromrev], repo[skiprev])
802 for dst, src in pathcopies(repo[fromrev], repo[rev]).iteritems():
801 for dst, src in pathcopies(repo[fromrev], repo[rev]).iteritems():
803 if dst in exclude:
802 if dst in exclude:
804 continue
803 continue
805 if dst in wctx:
804 if dst in wctx:
806 wctx[dst].markcopied(src)
805 wctx[dst].markcopied(src)
General Comments 0
You need to be logged in to leave comments. Login now