Show More
@@ -1,472 +1,480 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 | import util |
|
8 | import util | |
9 | import heapq |
|
9 | import heapq | |
10 |
|
10 | |||
11 | def _nonoverlap(d1, d2, d3): |
|
11 | def _nonoverlap(d1, d2, d3): | |
12 | "Return list of elements in d1 not in d2 or d3" |
|
12 | "Return list of elements in d1 not in d2 or d3" | |
13 | return sorted([d for d in d1 if d not in d3 and d not in d2]) |
|
13 | return sorted([d for d in d1 if d not in d3 and d not in d2]) | |
14 |
|
14 | |||
15 | def _dirname(f): |
|
15 | def _dirname(f): | |
16 | s = f.rfind("/") |
|
16 | s = f.rfind("/") | |
17 | if s == -1: |
|
17 | if s == -1: | |
18 | return "" |
|
18 | return "" | |
19 | return f[:s] |
|
19 | return f[:s] | |
20 |
|
20 | |||
21 | def _findlimit(repo, a, b): |
|
21 | def _findlimit(repo, a, b): | |
22 | """ |
|
22 | """ | |
23 | Find the last revision that needs to be checked to ensure that a full |
|
23 | Find the last revision that needs to be checked to ensure that a full | |
24 | transitive closure for file copies can be properly calculated. |
|
24 | transitive closure for file copies can be properly calculated. | |
25 | Generally, this means finding the earliest revision number that's an |
|
25 | Generally, this means finding the earliest revision number that's an | |
26 | ancestor of a or b but not both, except when a or b is a direct descendent |
|
26 | ancestor of a or b but not both, except when a or b is a direct descendent | |
27 | of the other, in which case we can return the minimum revnum of a and b. |
|
27 | of the other, in which case we can return the minimum revnum of a and b. | |
28 | None if no such revision exists. |
|
28 | None if no such revision exists. | |
29 | """ |
|
29 | """ | |
30 |
|
30 | |||
31 | # basic idea: |
|
31 | # basic idea: | |
32 | # - mark a and b with different sides |
|
32 | # - mark a and b with different sides | |
33 | # - if a parent's children are all on the same side, the parent is |
|
33 | # - if a parent's children are all on the same side, the parent is | |
34 | # on that side, otherwise it is on no side |
|
34 | # on that side, otherwise it is on no side | |
35 | # - walk the graph in topological order with the help of a heap; |
|
35 | # - walk the graph in topological order with the help of a heap; | |
36 | # - add unseen parents to side map |
|
36 | # - add unseen parents to side map | |
37 | # - clear side of any parent that has children on different sides |
|
37 | # - clear side of any parent that has children on different sides | |
38 | # - track number of interesting revs that might still be on a side |
|
38 | # - track number of interesting revs that might still be on a side | |
39 | # - track the lowest interesting rev seen |
|
39 | # - track the lowest interesting rev seen | |
40 | # - quit when interesting revs is zero |
|
40 | # - quit when interesting revs is zero | |
41 |
|
41 | |||
42 | cl = repo.changelog |
|
42 | cl = repo.changelog | |
43 | working = len(cl) # pseudo rev for the working directory |
|
43 | working = len(cl) # pseudo rev for the working directory | |
44 | if a is None: |
|
44 | if a is None: | |
45 | a = working |
|
45 | a = working | |
46 | if b is None: |
|
46 | if b is None: | |
47 | b = working |
|
47 | b = working | |
48 |
|
48 | |||
49 | side = {a: -1, b: 1} |
|
49 | side = {a: -1, b: 1} | |
50 | visit = [-a, -b] |
|
50 | visit = [-a, -b] | |
51 | heapq.heapify(visit) |
|
51 | heapq.heapify(visit) | |
52 | interesting = len(visit) |
|
52 | interesting = len(visit) | |
53 | hascommonancestor = False |
|
53 | hascommonancestor = False | |
54 | limit = working |
|
54 | limit = working | |
55 |
|
55 | |||
56 | while interesting: |
|
56 | while interesting: | |
57 | r = -heapq.heappop(visit) |
|
57 | r = -heapq.heappop(visit) | |
58 | if r == working: |
|
58 | if r == working: | |
59 | parents = [cl.rev(p) for p in repo.dirstate.parents()] |
|
59 | parents = [cl.rev(p) for p in repo.dirstate.parents()] | |
60 | else: |
|
60 | else: | |
61 | parents = cl.parentrevs(r) |
|
61 | parents = cl.parentrevs(r) | |
62 | for p in parents: |
|
62 | for p in parents: | |
63 | if p < 0: |
|
63 | if p < 0: | |
64 | continue |
|
64 | continue | |
65 | if p not in side: |
|
65 | if p not in side: | |
66 | # first time we see p; add it to visit |
|
66 | # first time we see p; add it to visit | |
67 | side[p] = side[r] |
|
67 | side[p] = side[r] | |
68 | if side[p]: |
|
68 | if side[p]: | |
69 | interesting += 1 |
|
69 | interesting += 1 | |
70 | heapq.heappush(visit, -p) |
|
70 | heapq.heappush(visit, -p) | |
71 | elif side[p] and side[p] != side[r]: |
|
71 | elif side[p] and side[p] != side[r]: | |
72 | # p was interesting but now we know better |
|
72 | # p was interesting but now we know better | |
73 | side[p] = 0 |
|
73 | side[p] = 0 | |
74 | interesting -= 1 |
|
74 | interesting -= 1 | |
75 | hascommonancestor = True |
|
75 | hascommonancestor = True | |
76 | if side[r]: |
|
76 | if side[r]: | |
77 | limit = r # lowest rev visited |
|
77 | limit = r # lowest rev visited | |
78 | interesting -= 1 |
|
78 | interesting -= 1 | |
79 |
|
79 | |||
80 | if not hascommonancestor: |
|
80 | if not hascommonancestor: | |
81 | return None |
|
81 | return None | |
82 |
|
82 | |||
83 | # Consider the following flow (see test-commit-amend.t under issue4405): |
|
83 | # Consider the following flow (see test-commit-amend.t under issue4405): | |
84 | # 1/ File 'a0' committed |
|
84 | # 1/ File 'a0' committed | |
85 | # 2/ File renamed from 'a0' to 'a1' in a new commit (call it 'a1') |
|
85 | # 2/ File renamed from 'a0' to 'a1' in a new commit (call it 'a1') | |
86 | # 3/ Move back to first commit |
|
86 | # 3/ Move back to first commit | |
87 | # 4/ Create a new commit via revert to contents of 'a1' (call it 'a1-amend') |
|
87 | # 4/ Create a new commit via revert to contents of 'a1' (call it 'a1-amend') | |
88 | # 5/ Rename file from 'a1' to 'a2' and commit --amend 'a1-msg' |
|
88 | # 5/ Rename file from 'a1' to 'a2' and commit --amend 'a1-msg' | |
89 | # |
|
89 | # | |
90 | # During the amend in step five, we will be in this state: |
|
90 | # During the amend in step five, we will be in this state: | |
91 | # |
|
91 | # | |
92 | # @ 3 temporary amend commit for a1-amend |
|
92 | # @ 3 temporary amend commit for a1-amend | |
93 | # | |
|
93 | # | | |
94 | # o 2 a1-amend |
|
94 | # o 2 a1-amend | |
95 | # | |
|
95 | # | | |
96 | # | o 1 a1 |
|
96 | # | o 1 a1 | |
97 | # |/ |
|
97 | # |/ | |
98 | # o 0 a0 |
|
98 | # o 0 a0 | |
99 | # |
|
99 | # | |
100 | # When _findlimit is called, a and b are revs 3 and 0, so limit will be 2, |
|
100 | # When _findlimit is called, a and b are revs 3 and 0, so limit will be 2, | |
101 | # yet the filelog has the copy information in rev 1 and we will not look |
|
101 | # yet the filelog has the copy information in rev 1 and we will not look | |
102 | # back far enough unless we also look at the a and b as candidates. |
|
102 | # back far enough unless we also look at the a and b as candidates. | |
103 | # This only occurs when a is a descendent of b or visa-versa. |
|
103 | # This only occurs when a is a descendent of b or visa-versa. | |
104 | return min(limit, a, b) |
|
104 | return min(limit, a, b) | |
105 |
|
105 | |||
106 | def _chain(src, dst, a, b): |
|
106 | def _chain(src, dst, a, b): | |
107 | '''chain two sets of copies a->b''' |
|
107 | '''chain two sets of copies a->b''' | |
108 | t = a.copy() |
|
108 | t = a.copy() | |
109 | for k, v in b.iteritems(): |
|
109 | for k, v in b.iteritems(): | |
110 | if v in t: |
|
110 | if v in t: | |
111 | # found a chain |
|
111 | # found a chain | |
112 | if t[v] != k: |
|
112 | if t[v] != k: | |
113 | # file wasn't renamed back to itself |
|
113 | # file wasn't renamed back to itself | |
114 | t[k] = t[v] |
|
114 | t[k] = t[v] | |
115 | if v not in dst: |
|
115 | if v not in dst: | |
116 | # chain was a rename, not a copy |
|
116 | # chain was a rename, not a copy | |
117 | del t[v] |
|
117 | del t[v] | |
118 | if v in src: |
|
118 | if v in src: | |
119 | # file is a copy of an existing file |
|
119 | # file is a copy of an existing file | |
120 | t[k] = v |
|
120 | t[k] = v | |
121 |
|
121 | |||
122 | # remove criss-crossed copies |
|
122 | # remove criss-crossed copies | |
123 | for k, v in t.items(): |
|
123 | for k, v in t.items(): | |
124 | if k in src and v in dst: |
|
124 | if k in src and v in dst: | |
125 | del t[k] |
|
125 | del t[k] | |
126 |
|
126 | |||
127 | return t |
|
127 | return t | |
128 |
|
128 | |||
129 | def _tracefile(fctx, am, limit=-1): |
|
129 | def _tracefile(fctx, am, limit=-1): | |
130 | '''return file context that is the ancestor of fctx present in ancestor |
|
130 | '''return file context that is the ancestor of fctx present in ancestor | |
131 | manifest am, stopping after the first ancestor lower than limit''' |
|
131 | manifest am, stopping after the first ancestor lower than limit''' | |
132 |
|
132 | |||
133 | for f in fctx.ancestors(): |
|
133 | for f in fctx.ancestors(): | |
134 | if am.get(f.path(), None) == f.filenode(): |
|
134 | if am.get(f.path(), None) == f.filenode(): | |
135 | return f |
|
135 | return f | |
136 | if limit >= 0 and f.linkrev() < limit and f.rev() < limit: |
|
136 | if limit >= 0 and f.linkrev() < limit and f.rev() < limit: | |
137 | return None |
|
137 | return None | |
138 |
|
138 | |||
139 | def _dirstatecopies(d): |
|
139 | def _dirstatecopies(d): | |
140 | ds = d._repo.dirstate |
|
140 | ds = d._repo.dirstate | |
141 | c = ds.copies().copy() |
|
141 | c = ds.copies().copy() | |
142 | for k in c.keys(): |
|
142 | for k in c.keys(): | |
143 | if ds[k] not in 'anm': |
|
143 | if ds[k] not in 'anm': | |
144 | del c[k] |
|
144 | del c[k] | |
145 | return c |
|
145 | return c | |
146 |
|
146 | |||
147 | def _forwardcopies(a, b): |
|
147 | def _forwardcopies(a, b): | |
148 | '''find {dst@b: src@a} copy mapping where a is an ancestor of b''' |
|
148 | '''find {dst@b: src@a} copy mapping where a is an ancestor of b''' | |
149 |
|
149 | |||
150 | # check for working copy |
|
150 | # check for working copy | |
151 | w = None |
|
151 | w = None | |
152 | if b.rev() is None: |
|
152 | if b.rev() is None: | |
153 | w = b |
|
153 | w = b | |
154 | b = w.p1() |
|
154 | b = w.p1() | |
155 | if a == b: |
|
155 | if a == b: | |
156 | # short-circuit to avoid issues with merge states |
|
156 | # short-circuit to avoid issues with merge states | |
157 | return _dirstatecopies(w) |
|
157 | return _dirstatecopies(w) | |
158 |
|
158 | |||
159 | # files might have to be traced back to the fctx parent of the last |
|
159 | # files might have to be traced back to the fctx parent of the last | |
160 | # one-side-only changeset, but not further back than that |
|
160 | # one-side-only changeset, but not further back than that | |
161 | limit = _findlimit(a._repo, a.rev(), b.rev()) |
|
161 | limit = _findlimit(a._repo, a.rev(), b.rev()) | |
162 | if limit is None: |
|
162 | if limit is None: | |
163 | limit = -1 |
|
163 | limit = -1 | |
164 | am = a.manifest() |
|
164 | am = a.manifest() | |
165 |
|
165 | |||
166 | # find where new files came from |
|
166 | # find where new files came from | |
167 | # we currently don't try to find where old files went, too expensive |
|
167 | # we currently don't try to find where old files went, too expensive | |
168 | # this means we can miss a case like 'hg rm b; hg cp a b' |
|
168 | # this means we can miss a case like 'hg rm b; hg cp a b' | |
169 | cm = {} |
|
169 | cm = {} | |
170 | missing = set(b.manifest().iterkeys()) |
|
170 | missing = set(b.manifest().iterkeys()) | |
171 | missing.difference_update(a.manifest().iterkeys()) |
|
171 | missing.difference_update(a.manifest().iterkeys()) | |
172 |
|
172 | |||
173 | ancestrycontext = a._repo.changelog.ancestors([b.rev()], inclusive=True) |
|
173 | ancestrycontext = a._repo.changelog.ancestors([b.rev()], inclusive=True) | |
174 | for f in missing: |
|
174 | for f in missing: | |
175 | fctx = b[f] |
|
175 | fctx = b[f] | |
176 | fctx._ancestrycontext = ancestrycontext |
|
176 | fctx._ancestrycontext = ancestrycontext | |
177 | ofctx = _tracefile(fctx, am, limit) |
|
177 | ofctx = _tracefile(fctx, am, limit) | |
178 | if ofctx: |
|
178 | if ofctx: | |
179 | cm[f] = ofctx.path() |
|
179 | cm[f] = ofctx.path() | |
180 |
|
180 | |||
181 | # combine copies from dirstate if necessary |
|
181 | # combine copies from dirstate if necessary | |
182 | if w is not None: |
|
182 | if w is not None: | |
183 | cm = _chain(a, w, cm, _dirstatecopies(w)) |
|
183 | cm = _chain(a, w, cm, _dirstatecopies(w)) | |
184 |
|
184 | |||
185 | return cm |
|
185 | return cm | |
186 |
|
186 | |||
187 | def _backwardrenames(a, b): |
|
187 | def _backwardrenames(a, b): | |
188 | # Even though we're not taking copies into account, 1:n rename situations |
|
188 | # Even though we're not taking copies into account, 1:n rename situations | |
189 | # can still exist (e.g. hg cp a b; hg mv a c). In those cases we |
|
189 | # can still exist (e.g. hg cp a b; hg mv a c). In those cases we | |
190 | # arbitrarily pick one of the renames. |
|
190 | # arbitrarily pick one of the renames. | |
191 | f = _forwardcopies(b, a) |
|
191 | f = _forwardcopies(b, a) | |
192 | r = {} |
|
192 | r = {} | |
193 | for k, v in sorted(f.iteritems()): |
|
193 | for k, v in sorted(f.iteritems()): | |
194 | # remove copies |
|
194 | # remove copies | |
195 | if v in a: |
|
195 | if v in a: | |
196 | continue |
|
196 | continue | |
197 | r[v] = k |
|
197 | r[v] = k | |
198 | return r |
|
198 | return r | |
199 |
|
199 | |||
200 | def pathcopies(x, y): |
|
200 | def pathcopies(x, y): | |
201 | '''find {dst@y: src@x} copy mapping for directed compare''' |
|
201 | '''find {dst@y: src@x} copy mapping for directed compare''' | |
202 | if x == y or not x or not y: |
|
202 | if x == y or not x or not y: | |
203 | return {} |
|
203 | return {} | |
204 | a = y.ancestor(x) |
|
204 | a = y.ancestor(x) | |
205 | if a == x: |
|
205 | if a == x: | |
206 | return _forwardcopies(x, y) |
|
206 | return _forwardcopies(x, y) | |
207 | if a == y: |
|
207 | if a == y: | |
208 | return _backwardrenames(x, y) |
|
208 | return _backwardrenames(x, y) | |
209 | return _chain(x, y, _backwardrenames(x, a), _forwardcopies(a, y)) |
|
209 | return _chain(x, y, _backwardrenames(x, a), _forwardcopies(a, y)) | |
210 |
|
210 | |||
|
211 | def _computenonoverlap(repo, m1, m2, ma): | |||
|
212 | """Computes the files exclusive to m1 and m2. | |||
|
213 | This is its own function so extensions can easily wrap this call to see what | |||
|
214 | files mergecopies is about to process. | |||
|
215 | """ | |||
|
216 | u1 = _nonoverlap(m1, m2, ma) | |||
|
217 | u2 = _nonoverlap(m2, m1, ma) | |||
|
218 | ||||
|
219 | if u1: | |||
|
220 | repo.ui.debug(" unmatched files in local:\n %s\n" | |||
|
221 | % "\n ".join(u1)) | |||
|
222 | if u2: | |||
|
223 | repo.ui.debug(" unmatched files in other:\n %s\n" | |||
|
224 | % "\n ".join(u2)) | |||
|
225 | return u1, u2 | |||
|
226 | ||||
211 | def mergecopies(repo, c1, c2, ca): |
|
227 | def mergecopies(repo, c1, c2, ca): | |
212 | """ |
|
228 | """ | |
213 | Find moves and copies between context c1 and c2 that are relevant |
|
229 | Find moves and copies between context c1 and c2 that are relevant | |
214 | for merging. |
|
230 | for merging. | |
215 |
|
231 | |||
216 | Returns four dicts: "copy", "movewithdir", "diverge", and |
|
232 | Returns four dicts: "copy", "movewithdir", "diverge", and | |
217 | "renamedelete". |
|
233 | "renamedelete". | |
218 |
|
234 | |||
219 | "copy" is a mapping from destination name -> source name, |
|
235 | "copy" is a mapping from destination name -> source name, | |
220 | where source is in c1 and destination is in c2 or vice-versa. |
|
236 | where source is in c1 and destination is in c2 or vice-versa. | |
221 |
|
237 | |||
222 | "movewithdir" is a mapping from source name -> destination name, |
|
238 | "movewithdir" is a mapping from source name -> destination name, | |
223 | where the file at source present in one context but not the other |
|
239 | where the file at source present in one context but not the other | |
224 | needs to be moved to destination by the merge process, because the |
|
240 | needs to be moved to destination by the merge process, because the | |
225 | other context moved the directory it is in. |
|
241 | other context moved the directory it is in. | |
226 |
|
242 | |||
227 | "diverge" is a mapping of source name -> list of destination names |
|
243 | "diverge" is a mapping of source name -> list of destination names | |
228 | for divergent renames. |
|
244 | for divergent renames. | |
229 |
|
245 | |||
230 | "renamedelete" is a mapping of source name -> list of destination |
|
246 | "renamedelete" is a mapping of source name -> list of destination | |
231 | names for files deleted in c1 that were renamed in c2 or vice-versa. |
|
247 | names for files deleted in c1 that were renamed in c2 or vice-versa. | |
232 | """ |
|
248 | """ | |
233 | # avoid silly behavior for update from empty dir |
|
249 | # avoid silly behavior for update from empty dir | |
234 | if not c1 or not c2 or c1 == c2: |
|
250 | if not c1 or not c2 or c1 == c2: | |
235 | return {}, {}, {}, {} |
|
251 | return {}, {}, {}, {} | |
236 |
|
252 | |||
237 | # avoid silly behavior for parent -> working dir |
|
253 | # avoid silly behavior for parent -> working dir | |
238 | if c2.node() is None and c1.node() == repo.dirstate.p1(): |
|
254 | if c2.node() is None and c1.node() == repo.dirstate.p1(): | |
239 | return repo.dirstate.copies(), {}, {}, {} |
|
255 | return repo.dirstate.copies(), {}, {}, {} | |
240 |
|
256 | |||
241 | limit = _findlimit(repo, c1.rev(), c2.rev()) |
|
257 | limit = _findlimit(repo, c1.rev(), c2.rev()) | |
242 | if limit is None: |
|
258 | if limit is None: | |
243 | # no common ancestor, no copies |
|
259 | # no common ancestor, no copies | |
244 | return {}, {}, {}, {} |
|
260 | return {}, {}, {}, {} | |
245 | m1 = c1.manifest() |
|
261 | m1 = c1.manifest() | |
246 | m2 = c2.manifest() |
|
262 | m2 = c2.manifest() | |
247 | ma = ca.manifest() |
|
263 | ma = ca.manifest() | |
248 |
|
264 | |||
249 | def makectx(f, n): |
|
265 | def makectx(f, n): | |
250 | if len(n) != 20: # in a working context? |
|
266 | if len(n) != 20: # in a working context? | |
251 | if c1.rev() is None: |
|
267 | if c1.rev() is None: | |
252 | return c1.filectx(f) |
|
268 | return c1.filectx(f) | |
253 | return c2.filectx(f) |
|
269 | return c2.filectx(f) | |
254 | return repo.filectx(f, fileid=n) |
|
270 | return repo.filectx(f, fileid=n) | |
255 |
|
271 | |||
256 | ctx = util.lrucachefunc(makectx) |
|
272 | ctx = util.lrucachefunc(makectx) | |
257 | copy = {} |
|
273 | copy = {} | |
258 | movewithdir = {} |
|
274 | movewithdir = {} | |
259 | fullcopy = {} |
|
275 | fullcopy = {} | |
260 | diverge = {} |
|
276 | diverge = {} | |
261 |
|
277 | |||
262 | repo.ui.debug(" searching for copies back to rev %d\n" % limit) |
|
278 | repo.ui.debug(" searching for copies back to rev %d\n" % limit) | |
263 |
|
279 | |||
264 | u1 = _nonoverlap(m1, m2, ma) |
|
280 | u1, u2 = _computenonoverlap(repo, m1, m2, ma) | |
265 | u2 = _nonoverlap(m2, m1, ma) |
|
|||
266 |
|
||||
267 | if u1: |
|
|||
268 | repo.ui.debug(" unmatched files in local:\n %s\n" |
|
|||
269 | % "\n ".join(u1)) |
|
|||
270 | if u2: |
|
|||
271 | repo.ui.debug(" unmatched files in other:\n %s\n" |
|
|||
272 | % "\n ".join(u2)) |
|
|||
273 |
|
281 | |||
274 | for f in u1: |
|
282 | for f in u1: | |
275 | checkcopies(ctx, f, m1, m2, ca, limit, diverge, copy, fullcopy) |
|
283 | checkcopies(ctx, f, m1, m2, ca, limit, diverge, copy, fullcopy) | |
276 |
|
284 | |||
277 | for f in u2: |
|
285 | for f in u2: | |
278 | checkcopies(ctx, f, m2, m1, ca, limit, diverge, copy, fullcopy) |
|
286 | checkcopies(ctx, f, m2, m1, ca, limit, diverge, copy, fullcopy) | |
279 |
|
287 | |||
280 | renamedelete = {} |
|
288 | renamedelete = {} | |
281 | renamedelete2 = set() |
|
289 | renamedelete2 = set() | |
282 | diverge2 = set() |
|
290 | diverge2 = set() | |
283 | for of, fl in diverge.items(): |
|
291 | for of, fl in diverge.items(): | |
284 | if len(fl) == 1 or of in c1 or of in c2: |
|
292 | if len(fl) == 1 or of in c1 or of in c2: | |
285 | del diverge[of] # not actually divergent, or not a rename |
|
293 | del diverge[of] # not actually divergent, or not a rename | |
286 | if of not in c1 and of not in c2: |
|
294 | if of not in c1 and of not in c2: | |
287 | # renamed on one side, deleted on the other side, but filter |
|
295 | # renamed on one side, deleted on the other side, but filter | |
288 | # out files that have been renamed and then deleted |
|
296 | # out files that have been renamed and then deleted | |
289 | renamedelete[of] = [f for f in fl if f in c1 or f in c2] |
|
297 | renamedelete[of] = [f for f in fl if f in c1 or f in c2] | |
290 | renamedelete2.update(fl) # reverse map for below |
|
298 | renamedelete2.update(fl) # reverse map for below | |
291 | else: |
|
299 | else: | |
292 | diverge2.update(fl) # reverse map for below |
|
300 | diverge2.update(fl) # reverse map for below | |
293 |
|
301 | |||
294 | bothnew = sorted([d for d in m1 if d in m2 and d not in ma]) |
|
302 | bothnew = sorted([d for d in m1 if d in m2 and d not in ma]) | |
295 | if bothnew: |
|
303 | if bothnew: | |
296 | repo.ui.debug(" unmatched files new in both:\n %s\n" |
|
304 | repo.ui.debug(" unmatched files new in both:\n %s\n" | |
297 | % "\n ".join(bothnew)) |
|
305 | % "\n ".join(bothnew)) | |
298 | bothdiverge, _copy, _fullcopy = {}, {}, {} |
|
306 | bothdiverge, _copy, _fullcopy = {}, {}, {} | |
299 | for f in bothnew: |
|
307 | for f in bothnew: | |
300 | checkcopies(ctx, f, m1, m2, ca, limit, bothdiverge, _copy, _fullcopy) |
|
308 | checkcopies(ctx, f, m1, m2, ca, limit, bothdiverge, _copy, _fullcopy) | |
301 | checkcopies(ctx, f, m2, m1, ca, limit, bothdiverge, _copy, _fullcopy) |
|
309 | checkcopies(ctx, f, m2, m1, ca, limit, bothdiverge, _copy, _fullcopy) | |
302 | for of, fl in bothdiverge.items(): |
|
310 | for of, fl in bothdiverge.items(): | |
303 | if len(fl) == 2 and fl[0] == fl[1]: |
|
311 | if len(fl) == 2 and fl[0] == fl[1]: | |
304 | copy[fl[0]] = of # not actually divergent, just matching renames |
|
312 | copy[fl[0]] = of # not actually divergent, just matching renames | |
305 |
|
313 | |||
306 | if fullcopy and repo.ui.debugflag: |
|
314 | if fullcopy and repo.ui.debugflag: | |
307 | repo.ui.debug(" all copies found (* = to merge, ! = divergent, " |
|
315 | repo.ui.debug(" all copies found (* = to merge, ! = divergent, " | |
308 | "% = renamed and deleted):\n") |
|
316 | "% = renamed and deleted):\n") | |
309 | for f in sorted(fullcopy): |
|
317 | for f in sorted(fullcopy): | |
310 | note = "" |
|
318 | note = "" | |
311 | if f in copy: |
|
319 | if f in copy: | |
312 | note += "*" |
|
320 | note += "*" | |
313 | if f in diverge2: |
|
321 | if f in diverge2: | |
314 | note += "!" |
|
322 | note += "!" | |
315 | if f in renamedelete2: |
|
323 | if f in renamedelete2: | |
316 | note += "%" |
|
324 | note += "%" | |
317 | repo.ui.debug(" src: '%s' -> dst: '%s' %s\n" % (fullcopy[f], f, |
|
325 | repo.ui.debug(" src: '%s' -> dst: '%s' %s\n" % (fullcopy[f], f, | |
318 | note)) |
|
326 | note)) | |
319 | del diverge2 |
|
327 | del diverge2 | |
320 |
|
328 | |||
321 | if not fullcopy: |
|
329 | if not fullcopy: | |
322 | return copy, movewithdir, diverge, renamedelete |
|
330 | return copy, movewithdir, diverge, renamedelete | |
323 |
|
331 | |||
324 | repo.ui.debug(" checking for directory renames\n") |
|
332 | repo.ui.debug(" checking for directory renames\n") | |
325 |
|
333 | |||
326 | # generate a directory move map |
|
334 | # generate a directory move map | |
327 | d1, d2 = c1.dirs(), c2.dirs() |
|
335 | d1, d2 = c1.dirs(), c2.dirs() | |
328 | d1.addpath('/') |
|
336 | d1.addpath('/') | |
329 | d2.addpath('/') |
|
337 | d2.addpath('/') | |
330 | invalid = set() |
|
338 | invalid = set() | |
331 | dirmove = {} |
|
339 | dirmove = {} | |
332 |
|
340 | |||
333 | # examine each file copy for a potential directory move, which is |
|
341 | # examine each file copy for a potential directory move, which is | |
334 | # when all the files in a directory are moved to a new directory |
|
342 | # when all the files in a directory are moved to a new directory | |
335 | for dst, src in fullcopy.iteritems(): |
|
343 | for dst, src in fullcopy.iteritems(): | |
336 | dsrc, ddst = _dirname(src), _dirname(dst) |
|
344 | dsrc, ddst = _dirname(src), _dirname(dst) | |
337 | if dsrc in invalid: |
|
345 | if dsrc in invalid: | |
338 | # already seen to be uninteresting |
|
346 | # already seen to be uninteresting | |
339 | continue |
|
347 | continue | |
340 | elif dsrc in d1 and ddst in d1: |
|
348 | elif dsrc in d1 and ddst in d1: | |
341 | # directory wasn't entirely moved locally |
|
349 | # directory wasn't entirely moved locally | |
342 | invalid.add(dsrc) |
|
350 | invalid.add(dsrc) | |
343 | elif dsrc in d2 and ddst in d2: |
|
351 | elif dsrc in d2 and ddst in d2: | |
344 | # directory wasn't entirely moved remotely |
|
352 | # directory wasn't entirely moved remotely | |
345 | invalid.add(dsrc) |
|
353 | invalid.add(dsrc) | |
346 | elif dsrc in dirmove and dirmove[dsrc] != ddst: |
|
354 | elif dsrc in dirmove and dirmove[dsrc] != ddst: | |
347 | # files from the same directory moved to two different places |
|
355 | # files from the same directory moved to two different places | |
348 | invalid.add(dsrc) |
|
356 | invalid.add(dsrc) | |
349 | else: |
|
357 | else: | |
350 | # looks good so far |
|
358 | # looks good so far | |
351 | dirmove[dsrc + "/"] = ddst + "/" |
|
359 | dirmove[dsrc + "/"] = ddst + "/" | |
352 |
|
360 | |||
353 | for i in invalid: |
|
361 | for i in invalid: | |
354 | if i in dirmove: |
|
362 | if i in dirmove: | |
355 | del dirmove[i] |
|
363 | del dirmove[i] | |
356 | del d1, d2, invalid |
|
364 | del d1, d2, invalid | |
357 |
|
365 | |||
358 | if not dirmove: |
|
366 | if not dirmove: | |
359 | return copy, movewithdir, diverge, renamedelete |
|
367 | return copy, movewithdir, diverge, renamedelete | |
360 |
|
368 | |||
361 | for d in dirmove: |
|
369 | for d in dirmove: | |
362 | repo.ui.debug(" discovered dir src: '%s' -> dst: '%s'\n" % |
|
370 | repo.ui.debug(" discovered dir src: '%s' -> dst: '%s'\n" % | |
363 | (d, dirmove[d])) |
|
371 | (d, dirmove[d])) | |
364 |
|
372 | |||
365 | # check unaccounted nonoverlapping files against directory moves |
|
373 | # check unaccounted nonoverlapping files against directory moves | |
366 | for f in u1 + u2: |
|
374 | for f in u1 + u2: | |
367 | if f not in fullcopy: |
|
375 | if f not in fullcopy: | |
368 | for d in dirmove: |
|
376 | for d in dirmove: | |
369 | if f.startswith(d): |
|
377 | if f.startswith(d): | |
370 | # new file added in a directory that was moved, move it |
|
378 | # new file added in a directory that was moved, move it | |
371 | df = dirmove[d] + f[len(d):] |
|
379 | df = dirmove[d] + f[len(d):] | |
372 | if df not in copy: |
|
380 | if df not in copy: | |
373 | movewithdir[f] = df |
|
381 | movewithdir[f] = df | |
374 | repo.ui.debug((" pending file src: '%s' -> " |
|
382 | repo.ui.debug((" pending file src: '%s' -> " | |
375 | "dst: '%s'\n") % (f, df)) |
|
383 | "dst: '%s'\n") % (f, df)) | |
376 | break |
|
384 | break | |
377 |
|
385 | |||
378 | return copy, movewithdir, diverge, renamedelete |
|
386 | return copy, movewithdir, diverge, renamedelete | |
379 |
|
387 | |||
380 | def checkcopies(ctx, f, m1, m2, ca, limit, diverge, copy, fullcopy): |
|
388 | def checkcopies(ctx, f, m1, m2, ca, limit, diverge, copy, fullcopy): | |
381 | """ |
|
389 | """ | |
382 | check possible copies of f from m1 to m2 |
|
390 | check possible copies of f from m1 to m2 | |
383 |
|
391 | |||
384 | ctx = function accepting (filename, node) that returns a filectx. |
|
392 | ctx = function accepting (filename, node) that returns a filectx. | |
385 | f = the filename to check |
|
393 | f = the filename to check | |
386 | m1 = the source manifest |
|
394 | m1 = the source manifest | |
387 | m2 = the destination manifest |
|
395 | m2 = the destination manifest | |
388 | ca = the changectx of the common ancestor |
|
396 | ca = the changectx of the common ancestor | |
389 | limit = the rev number to not search beyond |
|
397 | limit = the rev number to not search beyond | |
390 | diverge = record all diverges in this dict |
|
398 | diverge = record all diverges in this dict | |
391 | copy = record all non-divergent copies in this dict |
|
399 | copy = record all non-divergent copies in this dict | |
392 | fullcopy = record all copies in this dict |
|
400 | fullcopy = record all copies in this dict | |
393 | """ |
|
401 | """ | |
394 |
|
402 | |||
395 | ma = ca.manifest() |
|
403 | ma = ca.manifest() | |
396 |
|
404 | |||
397 | def _related(f1, f2, limit): |
|
405 | def _related(f1, f2, limit): | |
398 | # Walk back to common ancestor to see if the two files originate |
|
406 | # Walk back to common ancestor to see if the two files originate | |
399 | # from the same file. Since workingfilectx's rev() is None it messes |
|
407 | # from the same file. Since workingfilectx's rev() is None it messes | |
400 | # up the integer comparison logic, hence the pre-step check for |
|
408 | # up the integer comparison logic, hence the pre-step check for | |
401 | # None (f1 and f2 can only be workingfilectx's initially). |
|
409 | # None (f1 and f2 can only be workingfilectx's initially). | |
402 |
|
410 | |||
403 | if f1 == f2: |
|
411 | if f1 == f2: | |
404 | return f1 # a match |
|
412 | return f1 # a match | |
405 |
|
413 | |||
406 | g1, g2 = f1.ancestors(), f2.ancestors() |
|
414 | g1, g2 = f1.ancestors(), f2.ancestors() | |
407 | try: |
|
415 | try: | |
408 | f1r, f2r = f1.rev(), f2.rev() |
|
416 | f1r, f2r = f1.rev(), f2.rev() | |
409 |
|
417 | |||
410 | if f1r is None: |
|
418 | if f1r is None: | |
411 | f1 = g1.next() |
|
419 | f1 = g1.next() | |
412 | if f2r is None: |
|
420 | if f2r is None: | |
413 | f2 = g2.next() |
|
421 | f2 = g2.next() | |
414 |
|
422 | |||
415 | while True: |
|
423 | while True: | |
416 | f1r, f2r = f1.rev(), f2.rev() |
|
424 | f1r, f2r = f1.rev(), f2.rev() | |
417 | if f1r > f2r: |
|
425 | if f1r > f2r: | |
418 | f1 = g1.next() |
|
426 | f1 = g1.next() | |
419 | elif f2r > f1r: |
|
427 | elif f2r > f1r: | |
420 | f2 = g2.next() |
|
428 | f2 = g2.next() | |
421 | elif f1 == f2: |
|
429 | elif f1 == f2: | |
422 | return f1 # a match |
|
430 | return f1 # a match | |
423 | elif f1r == f2r or f1r < limit or f2r < limit: |
|
431 | elif f1r == f2r or f1r < limit or f2r < limit: | |
424 | return False # copy no longer relevant |
|
432 | return False # copy no longer relevant | |
425 | except StopIteration: |
|
433 | except StopIteration: | |
426 | return False |
|
434 | return False | |
427 |
|
435 | |||
428 | of = None |
|
436 | of = None | |
429 | seen = set([f]) |
|
437 | seen = set([f]) | |
430 | for oc in ctx(f, m1[f]).ancestors(): |
|
438 | for oc in ctx(f, m1[f]).ancestors(): | |
431 | ocr = oc.rev() |
|
439 | ocr = oc.rev() | |
432 | of = oc.path() |
|
440 | of = oc.path() | |
433 | if of in seen: |
|
441 | if of in seen: | |
434 | # check limit late - grab last rename before |
|
442 | # check limit late - grab last rename before | |
435 | if ocr < limit: |
|
443 | if ocr < limit: | |
436 | break |
|
444 | break | |
437 | continue |
|
445 | continue | |
438 | seen.add(of) |
|
446 | seen.add(of) | |
439 |
|
447 | |||
440 | fullcopy[f] = of # remember for dir rename detection |
|
448 | fullcopy[f] = of # remember for dir rename detection | |
441 | if of not in m2: |
|
449 | if of not in m2: | |
442 | continue # no match, keep looking |
|
450 | continue # no match, keep looking | |
443 | if m2[of] == ma.get(of): |
|
451 | if m2[of] == ma.get(of): | |
444 | break # no merge needed, quit early |
|
452 | break # no merge needed, quit early | |
445 | c2 = ctx(of, m2[of]) |
|
453 | c2 = ctx(of, m2[of]) | |
446 | cr = _related(oc, c2, ca.rev()) |
|
454 | cr = _related(oc, c2, ca.rev()) | |
447 | if cr and (of == f or of == c2.path()): # non-divergent |
|
455 | if cr and (of == f or of == c2.path()): # non-divergent | |
448 | copy[f] = of |
|
456 | copy[f] = of | |
449 | of = None |
|
457 | of = None | |
450 | break |
|
458 | break | |
451 |
|
459 | |||
452 | if of in ma: |
|
460 | if of in ma: | |
453 | diverge.setdefault(of, []).append(f) |
|
461 | diverge.setdefault(of, []).append(f) | |
454 |
|
462 | |||
455 | def duplicatecopies(repo, rev, fromrev, skiprev=None): |
|
463 | def duplicatecopies(repo, rev, fromrev, skiprev=None): | |
456 | '''reproduce copies from fromrev to rev in the dirstate |
|
464 | '''reproduce copies from fromrev to rev in the dirstate | |
457 |
|
465 | |||
458 | If skiprev is specified, it's a revision that should be used to |
|
466 | If skiprev is specified, it's a revision that should be used to | |
459 | filter copy records. Any copies that occur between fromrev and |
|
467 | filter copy records. Any copies that occur between fromrev and | |
460 | skiprev will not be duplicated, even if they appear in the set of |
|
468 | skiprev will not be duplicated, even if they appear in the set of | |
461 | copies between fromrev and rev. |
|
469 | copies between fromrev and rev. | |
462 | ''' |
|
470 | ''' | |
463 | exclude = {} |
|
471 | exclude = {} | |
464 | if skiprev is not None: |
|
472 | if skiprev is not None: | |
465 | exclude = pathcopies(repo[fromrev], repo[skiprev]) |
|
473 | exclude = pathcopies(repo[fromrev], repo[skiprev]) | |
466 | for dst, src in pathcopies(repo[fromrev], repo[rev]).iteritems(): |
|
474 | for dst, src in pathcopies(repo[fromrev], repo[rev]).iteritems(): | |
467 | # copies.pathcopies returns backward renames, so dst might not |
|
475 | # copies.pathcopies returns backward renames, so dst might not | |
468 | # actually be in the dirstate |
|
476 | # actually be in the dirstate | |
469 | if dst in exclude: |
|
477 | if dst in exclude: | |
470 | continue |
|
478 | continue | |
471 | if repo.dirstate[dst] in "nma": |
|
479 | if repo.dirstate[dst] in "nma": | |
472 | repo.dirstate.copy(src, dst) |
|
480 | repo.dirstate.copy(src, dst) |
General Comments 0
You need to be logged in to leave comments.
Login now