Show More
@@ -1,315 +1,313 b'' | |||||
1 | # convert.py Foreign SCM converter |
|
1 | # convert.py Foreign SCM converter | |
2 | # |
|
2 | # | |
3 | # Copyright 2005, 2006 Matt Mackall <mpm@selenic.com> |
|
3 | # Copyright 2005, 2006 Matt Mackall <mpm@selenic.com> | |
4 | # |
|
4 | # | |
5 | # This software may be used and distributed according to the terms |
|
5 | # This software may be used and distributed according to the terms | |
6 | # of the GNU General Public License, incorporated herein by reference. |
|
6 | # of the GNU General Public License, incorporated herein by reference. | |
7 |
|
7 | |||
8 | from common import NoRepo |
|
8 | from common import NoRepo | |
9 | from cvs import convert_cvs |
|
9 | from cvs import convert_cvs | |
10 | from git import convert_git |
|
10 | from git import convert_git | |
11 | from hg import convert_mercurial |
|
11 | from hg import convert_mercurial | |
12 |
|
12 | |||
13 | import os |
|
13 | import os | |
14 | from mercurial import hg, ui, util, commands |
|
14 | from mercurial import hg, ui, util, commands | |
15 |
|
15 | |||
16 | commands.norepo += " convert" |
|
16 | commands.norepo += " convert" | |
17 |
|
17 | |||
18 | converters = [convert_cvs, convert_git, convert_mercurial] |
|
18 | converters = [convert_cvs, convert_git, convert_mercurial] | |
19 |
|
19 | |||
20 | def converter(ui, path): |
|
20 | def converter(ui, path): | |
21 | if not os.path.isdir(path): |
|
21 | if not os.path.isdir(path): | |
22 | raise util.Abort("%s: not a directory" % path) |
|
22 | raise util.Abort("%s: not a directory" % path) | |
23 | for c in converters: |
|
23 | for c in converters: | |
24 | try: |
|
24 | try: | |
25 | return c(ui, path) |
|
25 | return c(ui, path) | |
26 | except NoRepo: |
|
26 | except NoRepo: | |
27 | pass |
|
27 | pass | |
28 | raise util.Abort("%s: unknown repository type" % path) |
|
28 | raise util.Abort("%s: unknown repository type" % path) | |
29 |
|
29 | |||
30 | class convert(object): |
|
30 | class convert(object): | |
31 | def __init__(self, ui, source, dest, mapfile, opts): |
|
31 | def __init__(self, ui, source, dest, mapfile, opts): | |
32 |
|
32 | |||
33 | self.source = source |
|
33 | self.source = source | |
34 | self.dest = dest |
|
34 | self.dest = dest | |
35 | self.ui = ui |
|
35 | self.ui = ui | |
36 | self.opts = opts |
|
36 | self.opts = opts | |
37 | self.commitcache = {} |
|
37 | self.commitcache = {} | |
38 | self.mapfile = mapfile |
|
38 | self.mapfile = mapfile | |
39 | self.mapfilefd = None |
|
39 | self.mapfilefd = None | |
40 | self.authors = {} |
|
40 | self.authors = {} | |
41 | self.authorfile = None |
|
41 | self.authorfile = None | |
42 |
|
42 | |||
43 | self.map = {} |
|
43 | self.map = {} | |
44 | try: |
|
44 | try: | |
45 | origmapfile = open(self.mapfile, 'r') |
|
45 | origmapfile = open(self.mapfile, 'r') | |
46 | for l in origmapfile: |
|
46 | for l in origmapfile: | |
47 | sv, dv = l[:-1].split() |
|
47 | sv, dv = l[:-1].split() | |
48 | self.map[sv] = dv |
|
48 | self.map[sv] = dv | |
49 | origmapfile.close() |
|
49 | origmapfile.close() | |
50 | except IOError: |
|
50 | except IOError: | |
51 | pass |
|
51 | pass | |
52 |
|
52 | |||
53 | # Read first the dst author map if any |
|
53 | # Read first the dst author map if any | |
54 | authorfile = self.dest.authorfile() |
|
54 | authorfile = self.dest.authorfile() | |
55 | if authorfile and os.path.exists(authorfile): |
|
55 | if authorfile and os.path.exists(authorfile): | |
56 | self.readauthormap(authorfile) |
|
56 | self.readauthormap(authorfile) | |
57 | # Extend/Override with new author map if necessary |
|
57 | # Extend/Override with new author map if necessary | |
58 | if opts.get('authors'): |
|
58 | if opts.get('authors'): | |
59 | import pdb |
|
|||
60 | pdb.set_trace() |
|
|||
61 | self.readauthormap(opts.get('authors')) |
|
59 | self.readauthormap(opts.get('authors')) | |
62 | self.authorfile = self.dest.authorfile() |
|
60 | self.authorfile = self.dest.authorfile() | |
63 |
|
61 | |||
64 | def walktree(self, heads): |
|
62 | def walktree(self, heads): | |
65 | visit = heads |
|
63 | visit = heads | |
66 | known = {} |
|
64 | known = {} | |
67 | parents = {} |
|
65 | parents = {} | |
68 | while visit: |
|
66 | while visit: | |
69 | n = visit.pop(0) |
|
67 | n = visit.pop(0) | |
70 | if n in known or n in self.map: continue |
|
68 | if n in known or n in self.map: continue | |
71 | known[n] = 1 |
|
69 | known[n] = 1 | |
72 | self.commitcache[n] = self.source.getcommit(n) |
|
70 | self.commitcache[n] = self.source.getcommit(n) | |
73 | cp = self.commitcache[n].parents |
|
71 | cp = self.commitcache[n].parents | |
74 | for p in cp: |
|
72 | for p in cp: | |
75 | parents.setdefault(n, []).append(p) |
|
73 | parents.setdefault(n, []).append(p) | |
76 | visit.append(p) |
|
74 | visit.append(p) | |
77 |
|
75 | |||
78 | return parents |
|
76 | return parents | |
79 |
|
77 | |||
80 | def toposort(self, parents): |
|
78 | def toposort(self, parents): | |
81 | visit = parents.keys() |
|
79 | visit = parents.keys() | |
82 | seen = {} |
|
80 | seen = {} | |
83 | children = {} |
|
81 | children = {} | |
84 |
|
82 | |||
85 | while visit: |
|
83 | while visit: | |
86 | n = visit.pop(0) |
|
84 | n = visit.pop(0) | |
87 | if n in seen: continue |
|
85 | if n in seen: continue | |
88 | seen[n] = 1 |
|
86 | seen[n] = 1 | |
89 | pc = 0 |
|
87 | pc = 0 | |
90 | if n in parents: |
|
88 | if n in parents: | |
91 | for p in parents[n]: |
|
89 | for p in parents[n]: | |
92 | if p not in self.map: pc += 1 |
|
90 | if p not in self.map: pc += 1 | |
93 | visit.append(p) |
|
91 | visit.append(p) | |
94 | children.setdefault(p, []).append(n) |
|
92 | children.setdefault(p, []).append(n) | |
95 | if not pc: root = n |
|
93 | if not pc: root = n | |
96 |
|
94 | |||
97 | s = [] |
|
95 | s = [] | |
98 | removed = {} |
|
96 | removed = {} | |
99 | visit = children.keys() |
|
97 | visit = children.keys() | |
100 | while visit: |
|
98 | while visit: | |
101 | n = visit.pop(0) |
|
99 | n = visit.pop(0) | |
102 | if n in removed: continue |
|
100 | if n in removed: continue | |
103 | dep = 0 |
|
101 | dep = 0 | |
104 | if n in parents: |
|
102 | if n in parents: | |
105 | for p in parents[n]: |
|
103 | for p in parents[n]: | |
106 | if p in self.map: continue |
|
104 | if p in self.map: continue | |
107 | if p not in removed: |
|
105 | if p not in removed: | |
108 | # we're still dependent |
|
106 | # we're still dependent | |
109 | visit.append(n) |
|
107 | visit.append(n) | |
110 | dep = 1 |
|
108 | dep = 1 | |
111 | break |
|
109 | break | |
112 |
|
110 | |||
113 | if not dep: |
|
111 | if not dep: | |
114 | # all n's parents are in the list |
|
112 | # all n's parents are in the list | |
115 | removed[n] = 1 |
|
113 | removed[n] = 1 | |
116 | if n not in self.map: |
|
114 | if n not in self.map: | |
117 | s.append(n) |
|
115 | s.append(n) | |
118 | if n in children: |
|
116 | if n in children: | |
119 | for c in children[n]: |
|
117 | for c in children[n]: | |
120 | visit.insert(0, c) |
|
118 | visit.insert(0, c) | |
121 |
|
119 | |||
122 | if self.opts.get('datesort'): |
|
120 | if self.opts.get('datesort'): | |
123 | depth = {} |
|
121 | depth = {} | |
124 | for n in s: |
|
122 | for n in s: | |
125 | depth[n] = 0 |
|
123 | depth[n] = 0 | |
126 | pl = [p for p in self.commitcache[n].parents |
|
124 | pl = [p for p in self.commitcache[n].parents | |
127 | if p not in self.map] |
|
125 | if p not in self.map] | |
128 | if pl: |
|
126 | if pl: | |
129 | depth[n] = max([depth[p] for p in pl]) + 1 |
|
127 | depth[n] = max([depth[p] for p in pl]) + 1 | |
130 |
|
128 | |||
131 | s = [(depth[n], self.commitcache[n].date, n) for n in s] |
|
129 | s = [(depth[n], self.commitcache[n].date, n) for n in s] | |
132 | s.sort() |
|
130 | s.sort() | |
133 | s = [e[2] for e in s] |
|
131 | s = [e[2] for e in s] | |
134 |
|
132 | |||
135 | return s |
|
133 | return s | |
136 |
|
134 | |||
137 | def mapentry(self, src, dst): |
|
135 | def mapentry(self, src, dst): | |
138 | if self.mapfilefd is None: |
|
136 | if self.mapfilefd is None: | |
139 | try: |
|
137 | try: | |
140 | self.mapfilefd = open(self.mapfile, "a") |
|
138 | self.mapfilefd = open(self.mapfile, "a") | |
141 | except IOError, (errno, strerror): |
|
139 | except IOError, (errno, strerror): | |
142 | raise util.Abort("Could not open map file %s: %s, %s\n" % (self.mapfile, errno, strerror)) |
|
140 | raise util.Abort("Could not open map file %s: %s, %s\n" % (self.mapfile, errno, strerror)) | |
143 | self.map[src] = dst |
|
141 | self.map[src] = dst | |
144 | self.mapfilefd.write("%s %s\n" % (src, dst)) |
|
142 | self.mapfilefd.write("%s %s\n" % (src, dst)) | |
145 | self.mapfilefd.flush() |
|
143 | self.mapfilefd.flush() | |
146 |
|
144 | |||
147 | def writeauthormap(self): |
|
145 | def writeauthormap(self): | |
148 | authorfile = self.authorfile |
|
146 | authorfile = self.authorfile | |
149 | if authorfile: |
|
147 | if authorfile: | |
150 | self.ui.status('Writing author map file %s\n' % authorfile) |
|
148 | self.ui.status('Writing author map file %s\n' % authorfile) | |
151 | ofile = open(authorfile, 'w+') |
|
149 | ofile = open(authorfile, 'w+') | |
152 | for author in self.authors: |
|
150 | for author in self.authors: | |
153 | ofile.write("%s=%s\n" % (author, self.authors[author])) |
|
151 | ofile.write("%s=%s\n" % (author, self.authors[author])) | |
154 | ofile.close() |
|
152 | ofile.close() | |
155 |
|
153 | |||
156 | def readauthormap(self, authorfile): |
|
154 | def readauthormap(self, authorfile): | |
157 | afile = open(authorfile, 'r') |
|
155 | afile = open(authorfile, 'r') | |
158 | for line in afile: |
|
156 | for line in afile: | |
159 | try: |
|
157 | try: | |
160 | srcauthor = line.split('=')[0].strip() |
|
158 | srcauthor = line.split('=')[0].strip() | |
161 | dstauthor = line.split('=')[1].strip() |
|
159 | dstauthor = line.split('=')[1].strip() | |
162 | if srcauthor in self.authors and dstauthor != self.authors[srcauthor]: |
|
160 | if srcauthor in self.authors and dstauthor != self.authors[srcauthor]: | |
163 | self.ui.status( |
|
161 | self.ui.status( | |
164 | 'Overriding mapping for author %s, was %s, will be %s\n' |
|
162 | 'Overriding mapping for author %s, was %s, will be %s\n' | |
165 | % (srcauthor, self.authors[srcauthor], dstauthor)) |
|
163 | % (srcauthor, self.authors[srcauthor], dstauthor)) | |
166 | else: |
|
164 | else: | |
167 | self.ui.debug('Mapping author %s to %s\n' |
|
165 | self.ui.debug('Mapping author %s to %s\n' | |
168 | % (srcauthor, dstauthor)) |
|
166 | % (srcauthor, dstauthor)) | |
169 | self.authors[srcauthor] = dstauthor |
|
167 | self.authors[srcauthor] = dstauthor | |
170 | except IndexError: |
|
168 | except IndexError: | |
171 | self.ui.warn( |
|
169 | self.ui.warn( | |
172 | 'Ignoring bad line in author file map %s: %s\n' |
|
170 | 'Ignoring bad line in author file map %s: %s\n' | |
173 | % (authorfile, line)) |
|
171 | % (authorfile, line)) | |
174 | afile.close() |
|
172 | afile.close() | |
175 |
|
173 | |||
176 | def copy(self, rev): |
|
174 | def copy(self, rev): | |
177 | c = self.commitcache[rev] |
|
175 | c = self.commitcache[rev] | |
178 | files = self.source.getchanges(rev) |
|
176 | files = self.source.getchanges(rev) | |
179 |
|
177 | |||
180 | for f, v in files: |
|
178 | for f, v in files: | |
181 | try: |
|
179 | try: | |
182 | data = self.source.getfile(f, v) |
|
180 | data = self.source.getfile(f, v) | |
183 | except IOError, inst: |
|
181 | except IOError, inst: | |
184 | self.dest.delfile(f) |
|
182 | self.dest.delfile(f) | |
185 | else: |
|
183 | else: | |
186 | e = self.source.getmode(f, v) |
|
184 | e = self.source.getmode(f, v) | |
187 | self.dest.putfile(f, e, data) |
|
185 | self.dest.putfile(f, e, data) | |
188 |
|
186 | |||
189 | r = [self.map[v] for v in c.parents] |
|
187 | r = [self.map[v] for v in c.parents] | |
190 | f = [f for f, v in files] |
|
188 | f = [f for f, v in files] | |
191 | newnode = self.dest.putcommit(f, r, c) |
|
189 | newnode = self.dest.putcommit(f, r, c) | |
192 | self.mapentry(rev, newnode) |
|
190 | self.mapentry(rev, newnode) | |
193 |
|
191 | |||
194 | def convert(self): |
|
192 | def convert(self): | |
195 | try: |
|
193 | try: | |
196 | self.ui.status("scanning source...\n") |
|
194 | self.ui.status("scanning source...\n") | |
197 | heads = self.source.getheads() |
|
195 | heads = self.source.getheads() | |
198 | parents = self.walktree(heads) |
|
196 | parents = self.walktree(heads) | |
199 | self.ui.status("sorting...\n") |
|
197 | self.ui.status("sorting...\n") | |
200 | t = self.toposort(parents) |
|
198 | t = self.toposort(parents) | |
201 | num = len(t) |
|
199 | num = len(t) | |
202 | c = None |
|
200 | c = None | |
203 |
|
201 | |||
204 | self.ui.status("converting...\n") |
|
202 | self.ui.status("converting...\n") | |
205 | for c in t: |
|
203 | for c in t: | |
206 | num -= 1 |
|
204 | num -= 1 | |
207 | desc = self.commitcache[c].desc |
|
205 | desc = self.commitcache[c].desc | |
208 | if "\n" in desc: |
|
206 | if "\n" in desc: | |
209 | desc = desc.splitlines()[0] |
|
207 | desc = desc.splitlines()[0] | |
210 | author = self.commitcache[c].author |
|
208 | author = self.commitcache[c].author | |
211 | author = self.authors.get(author, author) |
|
209 | author = self.authors.get(author, author) | |
212 | self.commitcache[c].author = author |
|
210 | self.commitcache[c].author = author | |
213 | self.ui.status("%d %s\n" % (num, desc)) |
|
211 | self.ui.status("%d %s\n" % (num, desc)) | |
214 | self.copy(c) |
|
212 | self.copy(c) | |
215 |
|
213 | |||
216 | tags = self.source.gettags() |
|
214 | tags = self.source.gettags() | |
217 | ctags = {} |
|
215 | ctags = {} | |
218 | for k in tags: |
|
216 | for k in tags: | |
219 | v = tags[k] |
|
217 | v = tags[k] | |
220 | if v in self.map: |
|
218 | if v in self.map: | |
221 | ctags[k] = self.map[v] |
|
219 | ctags[k] = self.map[v] | |
222 |
|
220 | |||
223 | if c and ctags: |
|
221 | if c and ctags: | |
224 | nrev = self.dest.puttags(ctags) |
|
222 | nrev = self.dest.puttags(ctags) | |
225 | # write another hash correspondence to override the previous |
|
223 | # write another hash correspondence to override the previous | |
226 | # one so we don't end up with extra tag heads |
|
224 | # one so we don't end up with extra tag heads | |
227 | if nrev: |
|
225 | if nrev: | |
228 | self.mapentry(c, nrev) |
|
226 | self.mapentry(c, nrev) | |
229 |
|
227 | |||
230 | self.writeauthormap() |
|
228 | self.writeauthormap() | |
231 | finally: |
|
229 | finally: | |
232 | self.cleanup() |
|
230 | self.cleanup() | |
233 |
|
231 | |||
234 | def cleanup(self): |
|
232 | def cleanup(self): | |
235 | if self.mapfilefd: |
|
233 | if self.mapfilefd: | |
236 | self.mapfilefd.close() |
|
234 | self.mapfilefd.close() | |
237 |
|
235 | |||
238 | def _convert(ui, src, dest=None, mapfile=None, **opts): |
|
236 | def _convert(ui, src, dest=None, mapfile=None, **opts): | |
239 | '''Convert a foreign SCM repository to a Mercurial one. |
|
237 | '''Convert a foreign SCM repository to a Mercurial one. | |
240 |
|
238 | |||
241 | Accepted source formats: |
|
239 | Accepted source formats: | |
242 | - GIT |
|
240 | - GIT | |
243 | - CVS |
|
241 | - CVS | |
244 |
|
242 | |||
245 | Accepted destination formats: |
|
243 | Accepted destination formats: | |
246 | - Mercurial |
|
244 | - Mercurial | |
247 |
|
245 | |||
248 | If destination isn't given, a new Mercurial repo named <src>-hg will |
|
246 | If destination isn't given, a new Mercurial repo named <src>-hg will | |
249 | be created. If <mapfile> isn't given, it will be put in a default |
|
247 | be created. If <mapfile> isn't given, it will be put in a default | |
250 | location (<dest>/.hg/shamap by default) |
|
248 | location (<dest>/.hg/shamap by default) | |
251 |
|
249 | |||
252 | The <mapfile> is a simple text file that maps each source commit ID to |
|
250 | The <mapfile> is a simple text file that maps each source commit ID to | |
253 | the destination ID for that revision, like so: |
|
251 | the destination ID for that revision, like so: | |
254 | <source ID> <destination ID> |
|
252 | <source ID> <destination ID> | |
255 |
|
253 | |||
256 | If the file doesn't exist, it's automatically created. It's updated |
|
254 | If the file doesn't exist, it's automatically created. It's updated | |
257 | on each commit copied, so convert-repo can be interrupted and can |
|
255 | on each commit copied, so convert-repo can be interrupted and can | |
258 | be run repeatedly to copy new commits. |
|
256 | be run repeatedly to copy new commits. | |
259 |
|
257 | |||
260 | The [username mapping] file is a simple text file that maps each source |
|
258 | The [username mapping] file is a simple text file that maps each source | |
261 | commit author to a destination commit author. It is handy for source SCMs |
|
259 | commit author to a destination commit author. It is handy for source SCMs | |
262 | that use unix logins to identify authors (eg: CVS). One line per author |
|
260 | that use unix logins to identify authors (eg: CVS). One line per author | |
263 | mapping and the line format is: |
|
261 | mapping and the line format is: | |
264 | srcauthor=whatever string you want |
|
262 | srcauthor=whatever string you want | |
265 | ''' |
|
263 | ''' | |
266 |
|
264 | |||
267 | srcc = converter(ui, src) |
|
265 | srcc = converter(ui, src) | |
268 | if not hasattr(srcc, "getcommit"): |
|
266 | if not hasattr(srcc, "getcommit"): | |
269 | raise util.Abort("%s: can't read from this repo type" % src) |
|
267 | raise util.Abort("%s: can't read from this repo type" % src) | |
270 |
|
268 | |||
271 | if not dest: |
|
269 | if not dest: | |
272 | dest = src + "-hg" |
|
270 | dest = src + "-hg" | |
273 | ui.status("assuming destination %s\n" % dest) |
|
271 | ui.status("assuming destination %s\n" % dest) | |
274 |
|
272 | |||
275 | # Try to be smart and initalize things when required |
|
273 | # Try to be smart and initalize things when required | |
276 | if os.path.isdir(dest): |
|
274 | if os.path.isdir(dest): | |
277 | if len(os.listdir(dest)) > 0: |
|
275 | if len(os.listdir(dest)) > 0: | |
278 | try: |
|
276 | try: | |
279 | hg.repository(ui, dest) |
|
277 | hg.repository(ui, dest) | |
280 | ui.status("destination %s is a Mercurial repository\n" % dest) |
|
278 | ui.status("destination %s is a Mercurial repository\n" % dest) | |
281 | except hg.RepoError: |
|
279 | except hg.RepoError: | |
282 | raise util.Abort( |
|
280 | raise util.Abort( | |
283 | "destination directory %s is not empty.\n" |
|
281 | "destination directory %s is not empty.\n" | |
284 | "Please specify an empty directory to be initialized\n" |
|
282 | "Please specify an empty directory to be initialized\n" | |
285 | "or an already initialized mercurial repository" |
|
283 | "or an already initialized mercurial repository" | |
286 | % dest) |
|
284 | % dest) | |
287 | else: |
|
285 | else: | |
288 | ui.status("initializing destination %s repository\n" % dest) |
|
286 | ui.status("initializing destination %s repository\n" % dest) | |
289 | hg.repository(ui, dest, create=True) |
|
287 | hg.repository(ui, dest, create=True) | |
290 | elif os.path.exists(dest): |
|
288 | elif os.path.exists(dest): | |
291 | raise util.Abort("destination %s exists and is not a directory" % dest) |
|
289 | raise util.Abort("destination %s exists and is not a directory" % dest) | |
292 | else: |
|
290 | else: | |
293 | ui.status("initializing destination %s repository\n" % dest) |
|
291 | ui.status("initializing destination %s repository\n" % dest) | |
294 | hg.repository(ui, dest, create=True) |
|
292 | hg.repository(ui, dest, create=True) | |
295 |
|
293 | |||
296 | destc = converter(ui, dest) |
|
294 | destc = converter(ui, dest) | |
297 | if not hasattr(destc, "putcommit"): |
|
295 | if not hasattr(destc, "putcommit"): | |
298 | raise util.Abort("%s: can't write to this repo type" % src) |
|
296 | raise util.Abort("%s: can't write to this repo type" % src) | |
299 |
|
297 | |||
300 | if not mapfile: |
|
298 | if not mapfile: | |
301 | try: |
|
299 | try: | |
302 | mapfile = destc.mapfile() |
|
300 | mapfile = destc.mapfile() | |
303 | except: |
|
301 | except: | |
304 | mapfile = os.path.join(destc, "map") |
|
302 | mapfile = os.path.join(destc, "map") | |
305 |
|
303 | |||
306 | c = convert(ui, srcc, destc, mapfile, opts) |
|
304 | c = convert(ui, srcc, destc, mapfile, opts) | |
307 | c.convert() |
|
305 | c.convert() | |
308 |
|
306 | |||
309 | cmdtable = { |
|
307 | cmdtable = { | |
310 | "convert": |
|
308 | "convert": | |
311 | (_convert, |
|
309 | (_convert, | |
312 | [('A', 'authors', '', 'username mapping filename'), |
|
310 | [('A', 'authors', '', 'username mapping filename'), | |
313 | ('', 'datesort', None, 'try to sort changesets by date')], |
|
311 | ('', 'datesort', None, 'try to sort changesets by date')], | |
314 | 'hg convert [OPTION]... SOURCE [DEST [MAPFILE]]'), |
|
312 | 'hg convert [OPTION]... SOURCE [DEST [MAPFILE]]'), | |
315 | } |
|
313 | } |
General Comments 0
You need to be logged in to leave comments.
Login now