Show More
@@ -1,409 +1,431 b'' | |||
|
1 | 1 | # common.py - common code for the convert extension |
|
2 | 2 | # |
|
3 | 3 | # Copyright 2005-2009 Matt Mackall <mpm@selenic.com> and others |
|
4 | 4 | # |
|
5 | 5 | # This software may be used and distributed according to the terms of the |
|
6 | 6 | # GNU General Public License version 2 or any later version. |
|
7 | 7 | |
|
8 | 8 | import base64, errno |
|
9 | 9 | import os |
|
10 | 10 | import cPickle as pickle |
|
11 | 11 | from mercurial import util |
|
12 | 12 | from mercurial.i18n import _ |
|
13 | 13 | |
|
14 | 14 | propertycache = util.propertycache |
|
15 | 15 | |
|
16 | 16 | def encodeargs(args): |
|
17 | 17 | def encodearg(s): |
|
18 | 18 | lines = base64.encodestring(s) |
|
19 | 19 | lines = [l.splitlines()[0] for l in lines] |
|
20 | 20 | return ''.join(lines) |
|
21 | 21 | |
|
22 | 22 | s = pickle.dumps(args) |
|
23 | 23 | return encodearg(s) |
|
24 | 24 | |
|
25 | 25 | def decodeargs(s): |
|
26 | 26 | s = base64.decodestring(s) |
|
27 | 27 | return pickle.loads(s) |
|
28 | 28 | |
|
29 | 29 | class MissingTool(Exception): |
|
30 | 30 | pass |
|
31 | 31 | |
|
32 | 32 | def checktool(exe, name=None, abort=True): |
|
33 | 33 | name = name or exe |
|
34 | 34 | if not util.findexe(exe): |
|
35 | 35 | exc = abort and util.Abort or MissingTool |
|
36 | 36 | raise exc(_('cannot find required "%s" tool') % name) |
|
37 | 37 | |
|
38 | 38 | class NoRepo(Exception): |
|
39 | 39 | pass |
|
40 | 40 | |
|
41 | 41 | SKIPREV = 'SKIP' |
|
42 | 42 | |
|
43 | 43 | class commit(object): |
|
44 | 44 | def __init__(self, author, date, desc, parents, branch=None, rev=None, |
|
45 | 45 | extra={}, sortkey=None): |
|
46 | 46 | self.author = author or 'unknown' |
|
47 | 47 | self.date = date or '0 0' |
|
48 | 48 | self.desc = desc |
|
49 | 49 | self.parents = parents |
|
50 | 50 | self.branch = branch |
|
51 | 51 | self.rev = rev |
|
52 | 52 | self.extra = extra |
|
53 | 53 | self.sortkey = sortkey |
|
54 | 54 | |
|
55 | 55 | class converter_source(object): |
|
56 | 56 | """Conversion source interface""" |
|
57 | 57 | |
|
58 | 58 | def __init__(self, ui, path=None, rev=None): |
|
59 | 59 | """Initialize conversion source (or raise NoRepo("message") |
|
60 | 60 | exception if path is not a valid repository)""" |
|
61 | 61 | self.ui = ui |
|
62 | 62 | self.path = path |
|
63 | 63 | self.rev = rev |
|
64 | 64 | |
|
65 | 65 | self.encoding = 'utf-8' |
|
66 | 66 | |
|
67 | 67 | def before(self): |
|
68 | 68 | pass |
|
69 | 69 | |
|
70 | 70 | def after(self): |
|
71 | 71 | pass |
|
72 | 72 | |
|
73 | 73 | def setrevmap(self, revmap): |
|
74 | 74 | """set the map of already-converted revisions""" |
|
75 | 75 | pass |
|
76 | 76 | |
|
77 | 77 | def getheads(self): |
|
78 | 78 | """Return a list of this repository's heads""" |
|
79 | 79 | raise NotImplementedError() |
|
80 | 80 | |
|
81 | 81 | def getfile(self, name, rev): |
|
82 | 82 | """Return a pair (data, mode) where data is the file content |
|
83 | 83 | as a string and mode one of '', 'x' or 'l'. rev is the |
|
84 | 84 | identifier returned by a previous call to getchanges(). Raise |
|
85 | 85 | IOError to indicate that name was deleted in rev. |
|
86 | 86 | """ |
|
87 | 87 | raise NotImplementedError() |
|
88 | 88 | |
|
89 | 89 | def getchanges(self, version): |
|
90 | 90 | """Returns a tuple of (files, copies). |
|
91 | 91 | |
|
92 | 92 | files is a sorted list of (filename, id) tuples for all files |
|
93 | 93 | changed between version and its first parent returned by |
|
94 | 94 | getcommit(). id is the source revision id of the file. |
|
95 | 95 | |
|
96 | 96 | copies is a dictionary of dest: source |
|
97 | 97 | """ |
|
98 | 98 | raise NotImplementedError() |
|
99 | 99 | |
|
100 | 100 | def getcommit(self, version): |
|
101 | 101 | """Return the commit object for version""" |
|
102 | 102 | raise NotImplementedError() |
|
103 | 103 | |
|
104 | 104 | def gettags(self): |
|
105 | 105 | """Return the tags as a dictionary of name: revision |
|
106 | 106 | |
|
107 | 107 | Tag names must be UTF-8 strings. |
|
108 | 108 | """ |
|
109 | 109 | raise NotImplementedError() |
|
110 | 110 | |
|
111 | 111 | def recode(self, s, encoding=None): |
|
112 | 112 | if not encoding: |
|
113 | 113 | encoding = self.encoding or 'utf-8' |
|
114 | 114 | |
|
115 | 115 | if isinstance(s, unicode): |
|
116 | 116 | return s.encode("utf-8") |
|
117 | 117 | try: |
|
118 | 118 | return s.decode(encoding).encode("utf-8") |
|
119 | 119 | except: |
|
120 | 120 | try: |
|
121 | 121 | return s.decode("latin-1").encode("utf-8") |
|
122 | 122 | except: |
|
123 | 123 | return s.decode(encoding, "replace").encode("utf-8") |
|
124 | 124 | |
|
125 | 125 | def getchangedfiles(self, rev, i): |
|
126 | 126 | """Return the files changed by rev compared to parent[i]. |
|
127 | 127 | |
|
128 | 128 | i is an index selecting one of the parents of rev. The return |
|
129 | 129 | value should be the list of files that are different in rev and |
|
130 | 130 | this parent. |
|
131 | 131 | |
|
132 | 132 | If rev has no parents, i is None. |
|
133 | 133 | |
|
134 | 134 | This function is only needed to support --filemap |
|
135 | 135 | """ |
|
136 | 136 | raise NotImplementedError() |
|
137 | 137 | |
|
138 | 138 | def converted(self, rev, sinkrev): |
|
139 | 139 | '''Notify the source that a revision has been converted.''' |
|
140 | 140 | pass |
|
141 | 141 | |
|
142 | 142 | def hasnativeorder(self): |
|
143 | 143 | """Return true if this source has a meaningful, native revision |
|
144 | 144 | order. For instance, Mercurial revisions are store sequentially |
|
145 | 145 | while there is no such global ordering with Darcs. |
|
146 | 146 | """ |
|
147 | 147 | return False |
|
148 | 148 | |
|
149 | 149 | def lookuprev(self, rev): |
|
150 | 150 | """If rev is a meaningful revision reference in source, return |
|
151 | 151 | the referenced identifier in the same format used by getcommit(). |
|
152 | 152 | return None otherwise. |
|
153 | 153 | """ |
|
154 | 154 | return None |
|
155 | 155 | |
|
156 | 156 | def getbookmarks(self): |
|
157 | 157 | """Return the bookmarks as a dictionary of name: revision |
|
158 | 158 | |
|
159 | 159 | Bookmark names are to be UTF-8 strings. |
|
160 | 160 | """ |
|
161 | 161 | return {} |
|
162 | 162 | |
|
163 | 163 | class converter_sink(object): |
|
164 | 164 | """Conversion sink (target) interface""" |
|
165 | 165 | |
|
166 | 166 | def __init__(self, ui, path): |
|
167 | 167 | """Initialize conversion sink (or raise NoRepo("message") |
|
168 | 168 | exception if path is not a valid repository) |
|
169 | 169 | |
|
170 | 170 | created is a list of paths to remove if a fatal error occurs |
|
171 | 171 | later""" |
|
172 | 172 | self.ui = ui |
|
173 | 173 | self.path = path |
|
174 | 174 | self.created = [] |
|
175 | 175 | |
|
176 | 176 | def getheads(self): |
|
177 | 177 | """Return a list of this repository's heads""" |
|
178 | 178 | raise NotImplementedError() |
|
179 | 179 | |
|
180 | 180 | def revmapfile(self): |
|
181 | 181 | """Path to a file that will contain lines |
|
182 | 182 | source_rev_id sink_rev_id |
|
183 | 183 | mapping equivalent revision identifiers for each system.""" |
|
184 | 184 | raise NotImplementedError() |
|
185 | 185 | |
|
186 | 186 | def authorfile(self): |
|
187 | 187 | """Path to a file that will contain lines |
|
188 | 188 | srcauthor=dstauthor |
|
189 | 189 | mapping equivalent authors identifiers for each system.""" |
|
190 | 190 | return None |
|
191 | 191 | |
|
192 | 192 | def putcommit(self, files, copies, parents, commit, source, revmap): |
|
193 | 193 | """Create a revision with all changed files listed in 'files' |
|
194 | 194 | and having listed parents. 'commit' is a commit object |
|
195 | 195 | containing at a minimum the author, date, and message for this |
|
196 | 196 | changeset. 'files' is a list of (path, version) tuples, |
|
197 | 197 | 'copies' is a dictionary mapping destinations to sources, |
|
198 | 198 | 'source' is the source repository, and 'revmap' is a mapfile |
|
199 | 199 | of source revisions to converted revisions. Only getfile() and |
|
200 | 200 | lookuprev() should be called on 'source'. |
|
201 | 201 | |
|
202 | 202 | Note that the sink repository is not told to update itself to |
|
203 | 203 | a particular revision (or even what that revision would be) |
|
204 | 204 | before it receives the file data. |
|
205 | 205 | """ |
|
206 | 206 | raise NotImplementedError() |
|
207 | 207 | |
|
208 | 208 | def puttags(self, tags): |
|
209 | 209 | """Put tags into sink. |
|
210 | 210 | |
|
211 | 211 | tags: {tagname: sink_rev_id, ...} where tagname is an UTF-8 string. |
|
212 | 212 | Return a pair (tag_revision, tag_parent_revision), or (None, None) |
|
213 | 213 | if nothing was changed. |
|
214 | 214 | """ |
|
215 | 215 | raise NotImplementedError() |
|
216 | 216 | |
|
217 | 217 | def setbranch(self, branch, pbranches): |
|
218 | 218 | """Set the current branch name. Called before the first putcommit |
|
219 | 219 | on the branch. |
|
220 | 220 | branch: branch name for subsequent commits |
|
221 | 221 | pbranches: (converted parent revision, parent branch) tuples""" |
|
222 | 222 | pass |
|
223 | 223 | |
|
224 | 224 | def setfilemapmode(self, active): |
|
225 | 225 | """Tell the destination that we're using a filemap |
|
226 | 226 | |
|
227 | 227 | Some converter_sources (svn in particular) can claim that a file |
|
228 | 228 | was changed in a revision, even if there was no change. This method |
|
229 | 229 | tells the destination that we're using a filemap and that it should |
|
230 | 230 | filter empty revisions. |
|
231 | 231 | """ |
|
232 | 232 | pass |
|
233 | 233 | |
|
234 | 234 | def before(self): |
|
235 | 235 | pass |
|
236 | 236 | |
|
237 | 237 | def after(self): |
|
238 | 238 | pass |
|
239 | 239 | |
|
240 | 240 | def putbookmarks(self, bookmarks): |
|
241 | 241 | """Put bookmarks into sink. |
|
242 | 242 | |
|
243 | 243 | bookmarks: {bookmarkname: sink_rev_id, ...} |
|
244 | 244 | where bookmarkname is an UTF-8 string. |
|
245 | 245 | """ |
|
246 | 246 | pass |
|
247 | 247 | |
|
248 | 248 | class commandline(object): |
|
249 | 249 | def __init__(self, ui, command): |
|
250 | 250 | self.ui = ui |
|
251 | 251 | self.command = command |
|
252 | 252 | |
|
253 | 253 | def prerun(self): |
|
254 | 254 | pass |
|
255 | 255 | |
|
256 | 256 | def postrun(self): |
|
257 | 257 | pass |
|
258 | 258 | |
|
259 | 259 | def _cmdline(self, cmd, closestdin, *args, **kwargs): |
|
260 | 260 | cmdline = [self.command, cmd] + list(args) |
|
261 | 261 | for k, v in kwargs.iteritems(): |
|
262 | 262 | if len(k) == 1: |
|
263 | 263 | cmdline.append('-' + k) |
|
264 | 264 | else: |
|
265 | 265 | cmdline.append('--' + k.replace('_', '-')) |
|
266 | 266 | try: |
|
267 | 267 | if len(k) == 1: |
|
268 | 268 | cmdline.append('' + v) |
|
269 | 269 | else: |
|
270 | 270 | cmdline[-1] += '=' + v |
|
271 | 271 | except TypeError: |
|
272 | 272 | pass |
|
273 | 273 | cmdline = [util.shellquote(arg) for arg in cmdline] |
|
274 | 274 | if not self.ui.debugflag: |
|
275 | 275 | cmdline += ['2>', util.nulldev] |
|
276 | 276 | if closestdin: |
|
277 | 277 | cmdline += ['<', util.nulldev] |
|
278 | 278 | cmdline = ' '.join(cmdline) |
|
279 | 279 | return cmdline |
|
280 | 280 | |
|
281 | 281 | def _run(self, cmd, *args, **kwargs): |
|
282 | 282 | return self._dorun(util.popen, cmd, True, *args, **kwargs) |
|
283 | 283 | |
|
284 | 284 | def _run2(self, cmd, *args, **kwargs): |
|
285 | 285 | return self._dorun(util.popen2, cmd, False, *args, **kwargs) |
|
286 | 286 | |
|
287 | 287 | def _dorun(self, openfunc, cmd, closestdin, *args, **kwargs): |
|
288 | 288 | cmdline = self._cmdline(cmd, closestdin, *args, **kwargs) |
|
289 | 289 | self.ui.debug('running: %s\n' % (cmdline,)) |
|
290 | 290 | self.prerun() |
|
291 | 291 | try: |
|
292 | 292 | return openfunc(cmdline) |
|
293 | 293 | finally: |
|
294 | 294 | self.postrun() |
|
295 | 295 | |
|
296 | 296 | def run(self, cmd, *args, **kwargs): |
|
297 | 297 | fp = self._run(cmd, *args, **kwargs) |
|
298 | 298 | output = fp.read() |
|
299 | 299 | self.ui.debug(output) |
|
300 | 300 | return output, fp.close() |
|
301 | 301 | |
|
302 | 302 | def runlines(self, cmd, *args, **kwargs): |
|
303 | 303 | fp = self._run(cmd, *args, **kwargs) |
|
304 | 304 | output = fp.readlines() |
|
305 | 305 | self.ui.debug(''.join(output)) |
|
306 | 306 | return output, fp.close() |
|
307 | 307 | |
|
308 | 308 | def checkexit(self, status, output=''): |
|
309 | 309 | if status: |
|
310 | 310 | if output: |
|
311 | 311 | self.ui.warn(_('%s error:\n') % self.command) |
|
312 | 312 | self.ui.warn(output) |
|
313 | 313 | msg = util.explainexit(status)[0] |
|
314 | 314 | raise util.Abort('%s %s' % (self.command, msg)) |
|
315 | 315 | |
|
316 | 316 | def run0(self, cmd, *args, **kwargs): |
|
317 | 317 | output, status = self.run(cmd, *args, **kwargs) |
|
318 | 318 | self.checkexit(status, output) |
|
319 | 319 | return output |
|
320 | 320 | |
|
321 | 321 | def runlines0(self, cmd, *args, **kwargs): |
|
322 | 322 | output, status = self.runlines(cmd, *args, **kwargs) |
|
323 | 323 | self.checkexit(status, ''.join(output)) |
|
324 | 324 | return output |
|
325 | 325 | |
|
326 | 326 | @propertycache |
|
327 | 327 | def argmax(self): |
|
328 | 328 | # POSIX requires at least 4096 bytes for ARG_MAX |
|
329 | 329 | argmax = 4096 |
|
330 | 330 | try: |
|
331 | 331 | argmax = os.sysconf("SC_ARG_MAX") |
|
332 | 332 | except: |
|
333 | 333 | pass |
|
334 | 334 | |
|
335 | 335 | # Windows shells impose their own limits on command line length, |
|
336 | 336 | # down to 2047 bytes for cmd.exe under Windows NT/2k and 2500 bytes |
|
337 | 337 | # for older 4nt.exe. See http://support.microsoft.com/kb/830473 for |
|
338 | 338 | # details about cmd.exe limitations. |
|
339 | 339 | |
|
340 | 340 | # Since ARG_MAX is for command line _and_ environment, lower our limit |
|
341 | 341 | # (and make happy Windows shells while doing this). |
|
342 | 342 | return argmax // 2 - 1 |
|
343 | 343 | |
|
344 | 344 | def limit_arglist(self, arglist, cmd, closestdin, *args, **kwargs): |
|
345 | 345 | cmdlen = len(self._cmdline(cmd, closestdin, *args, **kwargs)) |
|
346 | 346 | limit = self.argmax - cmdlen |
|
347 | 347 | bytes = 0 |
|
348 | 348 | fl = [] |
|
349 | 349 | for fn in arglist: |
|
350 | 350 | b = len(fn) + 3 |
|
351 | 351 | if bytes + b < limit or len(fl) == 0: |
|
352 | 352 | fl.append(fn) |
|
353 | 353 | bytes += b |
|
354 | 354 | else: |
|
355 | 355 | yield fl |
|
356 | 356 | fl = [fn] |
|
357 | 357 | bytes = b |
|
358 | 358 | if fl: |
|
359 | 359 | yield fl |
|
360 | 360 | |
|
361 | 361 | def xargs(self, arglist, cmd, *args, **kwargs): |
|
362 | 362 | for l in self.limit_arglist(arglist, cmd, True, *args, **kwargs): |
|
363 | 363 | self.run0(cmd, *(list(args) + l), **kwargs) |
|
364 | 364 | |
|
365 | 365 | class mapfile(dict): |
|
366 | 366 | def __init__(self, ui, path): |
|
367 | 367 | super(mapfile, self).__init__() |
|
368 | 368 | self.ui = ui |
|
369 | 369 | self.path = path |
|
370 | 370 | self.fp = None |
|
371 | 371 | self.order = [] |
|
372 | 372 | self._read() |
|
373 | 373 | |
|
374 | 374 | def _read(self): |
|
375 | 375 | if not self.path: |
|
376 | 376 | return |
|
377 | 377 | try: |
|
378 | 378 | fp = open(self.path, 'r') |
|
379 | 379 | except IOError, err: |
|
380 | 380 | if err.errno != errno.ENOENT: |
|
381 | 381 | raise |
|
382 | 382 | return |
|
383 | 383 | for i, line in enumerate(fp): |
|
384 | 384 | try: |
|
385 | 385 | key, value = line.splitlines()[0].rstrip().rsplit(' ', 1) |
|
386 | 386 | except ValueError: |
|
387 | 387 | raise util.Abort( |
|
388 | 388 | _('syntax error in %s(%d): key/value pair expected') |
|
389 | 389 | % (self.path, i + 1)) |
|
390 | 390 | if key not in self: |
|
391 | 391 | self.order.append(key) |
|
392 | 392 | super(mapfile, self).__setitem__(key, value) |
|
393 | 393 | fp.close() |
|
394 | 394 | |
|
395 | 395 | def __setitem__(self, key, value): |
|
396 | 396 | if self.fp is None: |
|
397 | 397 | try: |
|
398 | 398 | self.fp = open(self.path, 'a') |
|
399 | 399 | except IOError, err: |
|
400 | 400 | raise util.Abort(_('could not open map file %r: %s') % |
|
401 | 401 | (self.path, err.strerror)) |
|
402 | 402 | self.fp.write('%s %s\n' % (key, value)) |
|
403 | 403 | self.fp.flush() |
|
404 | 404 | super(mapfile, self).__setitem__(key, value) |
|
405 | 405 | |
|
406 | 406 | def close(self): |
|
407 | 407 | if self.fp: |
|
408 | 408 | self.fp.close() |
|
409 | 409 | self.fp = None |
|
410 | ||
|
411 | def parsesplicemap(path): | |
|
412 | """Parse a splicemap, return a child/parents dictionary.""" | |
|
413 | m = {} | |
|
414 | try: | |
|
415 | fp = open(path, 'r') | |
|
416 | for i, line in enumerate(fp): | |
|
417 | try: | |
|
418 | child, parents = line.splitlines()[0].rstrip().rsplit(' ', 1) | |
|
419 | parents = parents.replace(',', ' ').split() | |
|
420 | except ValueError: | |
|
421 | raise util.Abort(_('syntax error in %s(%d): child parent1' | |
|
422 | '[,parent2] expected') % (path, i + 1)) | |
|
423 | pp = [] | |
|
424 | for p in parents: | |
|
425 | if p not in pp: | |
|
426 | pp.append(p) | |
|
427 | m[child] = pp | |
|
428 | except IOError, e: | |
|
429 | if e.errno != errno.ENOENT: | |
|
430 | raise | |
|
431 | return m |
@@ -1,446 +1,446 b'' | |||
|
1 | 1 | # convcmd - convert extension commands definition |
|
2 | 2 | # |
|
3 | 3 | # Copyright 2005-2007 Matt Mackall <mpm@selenic.com> |
|
4 | 4 | # |
|
5 | 5 | # This software may be used and distributed according to the terms of the |
|
6 | 6 | # GNU General Public License version 2 or any later version. |
|
7 | 7 | |
|
8 | 8 | from common import NoRepo, MissingTool, SKIPREV, mapfile |
|
9 | 9 | from cvs import convert_cvs |
|
10 | 10 | from darcs import darcs_source |
|
11 | 11 | from git import convert_git |
|
12 | 12 | from hg import mercurial_source, mercurial_sink |
|
13 | 13 | from subversion import svn_source, svn_sink |
|
14 | 14 | from monotone import monotone_source |
|
15 | 15 | from gnuarch import gnuarch_source |
|
16 | 16 | from bzr import bzr_source |
|
17 | 17 | from p4 import p4_source |
|
18 | import filemap | |
|
18 | import filemap, common | |
|
19 | 19 | |
|
20 | 20 | import os, shutil |
|
21 | 21 | from mercurial import hg, util, encoding |
|
22 | 22 | from mercurial.i18n import _ |
|
23 | 23 | |
|
24 | 24 | orig_encoding = 'ascii' |
|
25 | 25 | |
|
26 | 26 | def recode(s): |
|
27 | 27 | if isinstance(s, unicode): |
|
28 | 28 | return s.encode(orig_encoding, 'replace') |
|
29 | 29 | else: |
|
30 | 30 | return s.decode('utf-8').encode(orig_encoding, 'replace') |
|
31 | 31 | |
|
32 | 32 | source_converters = [ |
|
33 | 33 | ('cvs', convert_cvs, 'branchsort'), |
|
34 | 34 | ('git', convert_git, 'branchsort'), |
|
35 | 35 | ('svn', svn_source, 'branchsort'), |
|
36 | 36 | ('hg', mercurial_source, 'sourcesort'), |
|
37 | 37 | ('darcs', darcs_source, 'branchsort'), |
|
38 | 38 | ('mtn', monotone_source, 'branchsort'), |
|
39 | 39 | ('gnuarch', gnuarch_source, 'branchsort'), |
|
40 | 40 | ('bzr', bzr_source, 'branchsort'), |
|
41 | 41 | ('p4', p4_source, 'branchsort'), |
|
42 | 42 | ] |
|
43 | 43 | |
|
44 | 44 | sink_converters = [ |
|
45 | 45 | ('hg', mercurial_sink), |
|
46 | 46 | ('svn', svn_sink), |
|
47 | 47 | ] |
|
48 | 48 | |
|
49 | 49 | def convertsource(ui, path, type, rev): |
|
50 | 50 | exceptions = [] |
|
51 | 51 | if type and type not in [s[0] for s in source_converters]: |
|
52 | 52 | raise util.Abort(_('%s: invalid source repository type') % type) |
|
53 | 53 | for name, source, sortmode in source_converters: |
|
54 | 54 | try: |
|
55 | 55 | if not type or name == type: |
|
56 | 56 | return source(ui, path, rev), sortmode |
|
57 | 57 | except (NoRepo, MissingTool), inst: |
|
58 | 58 | exceptions.append(inst) |
|
59 | 59 | if not ui.quiet: |
|
60 | 60 | for inst in exceptions: |
|
61 | 61 | ui.write("%s\n" % inst) |
|
62 | 62 | raise util.Abort(_('%s: missing or unsupported repository') % path) |
|
63 | 63 | |
|
64 | 64 | def convertsink(ui, path, type): |
|
65 | 65 | if type and type not in [s[0] for s in sink_converters]: |
|
66 | 66 | raise util.Abort(_('%s: invalid destination repository type') % type) |
|
67 | 67 | for name, sink in sink_converters: |
|
68 | 68 | try: |
|
69 | 69 | if not type or name == type: |
|
70 | 70 | return sink(ui, path) |
|
71 | 71 | except NoRepo, inst: |
|
72 | 72 | ui.note(_("convert: %s\n") % inst) |
|
73 | 73 | except MissingTool, inst: |
|
74 | 74 | raise util.Abort('%s\n' % inst) |
|
75 | 75 | raise util.Abort(_('%s: unknown repository type') % path) |
|
76 | 76 | |
|
77 | 77 | class progresssource(object): |
|
78 | 78 | def __init__(self, ui, source, filecount): |
|
79 | 79 | self.ui = ui |
|
80 | 80 | self.source = source |
|
81 | 81 | self.filecount = filecount |
|
82 | 82 | self.retrieved = 0 |
|
83 | 83 | |
|
84 | 84 | def getfile(self, file, rev): |
|
85 | 85 | self.retrieved += 1 |
|
86 | 86 | self.ui.progress(_('getting files'), self.retrieved, |
|
87 | 87 | item=file, total=self.filecount) |
|
88 | 88 | return self.source.getfile(file, rev) |
|
89 | 89 | |
|
90 | 90 | def lookuprev(self, rev): |
|
91 | 91 | return self.source.lookuprev(rev) |
|
92 | 92 | |
|
93 | 93 | def close(self): |
|
94 | 94 | self.ui.progress(_('getting files'), None) |
|
95 | 95 | |
|
96 | 96 | class converter(object): |
|
97 | 97 | def __init__(self, ui, source, dest, revmapfile, opts): |
|
98 | 98 | |
|
99 | 99 | self.source = source |
|
100 | 100 | self.dest = dest |
|
101 | 101 | self.ui = ui |
|
102 | 102 | self.opts = opts |
|
103 | 103 | self.commitcache = {} |
|
104 | 104 | self.authors = {} |
|
105 | 105 | self.authorfile = None |
|
106 | 106 | |
|
107 | 107 | # Record converted revisions persistently: maps source revision |
|
108 | 108 | # ID to target revision ID (both strings). (This is how |
|
109 | 109 | # incremental conversions work.) |
|
110 | 110 | self.map = mapfile(ui, revmapfile) |
|
111 | 111 | |
|
112 | 112 | # Read first the dst author map if any |
|
113 | 113 | authorfile = self.dest.authorfile() |
|
114 | 114 | if authorfile and os.path.exists(authorfile): |
|
115 | 115 | self.readauthormap(authorfile) |
|
116 | 116 | # Extend/Override with new author map if necessary |
|
117 | 117 | if opts.get('authormap'): |
|
118 | 118 | self.readauthormap(opts.get('authormap')) |
|
119 | 119 | self.authorfile = self.dest.authorfile() |
|
120 | 120 | |
|
121 |
self.splicemap = |
|
|
121 | self.splicemap = common.parsesplicemap(opts.get('splicemap')) | |
|
122 | 122 | self.branchmap = mapfile(ui, opts.get('branchmap')) |
|
123 | 123 | |
|
124 | 124 | def walktree(self, heads): |
|
125 | 125 | '''Return a mapping that identifies the uncommitted parents of every |
|
126 | 126 | uncommitted changeset.''' |
|
127 | 127 | visit = heads |
|
128 | 128 | known = set() |
|
129 | 129 | parents = {} |
|
130 | 130 | while visit: |
|
131 | 131 | n = visit.pop(0) |
|
132 | 132 | if n in known or n in self.map: |
|
133 | 133 | continue |
|
134 | 134 | known.add(n) |
|
135 | 135 | self.ui.progress(_('scanning'), len(known), unit=_('revisions')) |
|
136 | 136 | commit = self.cachecommit(n) |
|
137 | 137 | parents[n] = [] |
|
138 | 138 | for p in commit.parents: |
|
139 | 139 | parents[n].append(p) |
|
140 | 140 | visit.append(p) |
|
141 | 141 | self.ui.progress(_('scanning'), None) |
|
142 | 142 | |
|
143 | 143 | return parents |
|
144 | 144 | |
|
145 | 145 | def toposort(self, parents, sortmode): |
|
146 | 146 | '''Return an ordering such that every uncommitted changeset is |
|
147 | 147 | preceeded by all its uncommitted ancestors.''' |
|
148 | 148 | |
|
149 | 149 | def mapchildren(parents): |
|
150 | 150 | """Return a (children, roots) tuple where 'children' maps parent |
|
151 | 151 | revision identifiers to children ones, and 'roots' is the list of |
|
152 | 152 | revisions without parents. 'parents' must be a mapping of revision |
|
153 | 153 | identifier to its parents ones. |
|
154 | 154 | """ |
|
155 | 155 | visit = parents.keys() |
|
156 | 156 | seen = set() |
|
157 | 157 | children = {} |
|
158 | 158 | roots = [] |
|
159 | 159 | |
|
160 | 160 | while visit: |
|
161 | 161 | n = visit.pop(0) |
|
162 | 162 | if n in seen: |
|
163 | 163 | continue |
|
164 | 164 | seen.add(n) |
|
165 | 165 | # Ensure that nodes without parents are present in the |
|
166 | 166 | # 'children' mapping. |
|
167 | 167 | children.setdefault(n, []) |
|
168 | 168 | hasparent = False |
|
169 | 169 | for p in parents[n]: |
|
170 | 170 | if not p in self.map: |
|
171 | 171 | visit.append(p) |
|
172 | 172 | hasparent = True |
|
173 | 173 | children.setdefault(p, []).append(n) |
|
174 | 174 | if not hasparent: |
|
175 | 175 | roots.append(n) |
|
176 | 176 | |
|
177 | 177 | return children, roots |
|
178 | 178 | |
|
179 | 179 | # Sort functions are supposed to take a list of revisions which |
|
180 | 180 | # can be converted immediately and pick one |
|
181 | 181 | |
|
182 | 182 | def makebranchsorter(): |
|
183 | 183 | """If the previously converted revision has a child in the |
|
184 | 184 | eligible revisions list, pick it. Return the list head |
|
185 | 185 | otherwise. Branch sort attempts to minimize branch |
|
186 | 186 | switching, which is harmful for Mercurial backend |
|
187 | 187 | compression. |
|
188 | 188 | """ |
|
189 | 189 | prev = [None] |
|
190 | 190 | def picknext(nodes): |
|
191 | 191 | next = nodes[0] |
|
192 | 192 | for n in nodes: |
|
193 | 193 | if prev[0] in parents[n]: |
|
194 | 194 | next = n |
|
195 | 195 | break |
|
196 | 196 | prev[0] = next |
|
197 | 197 | return next |
|
198 | 198 | return picknext |
|
199 | 199 | |
|
200 | 200 | def makesourcesorter(): |
|
201 | 201 | """Source specific sort.""" |
|
202 | 202 | keyfn = lambda n: self.commitcache[n].sortkey |
|
203 | 203 | def picknext(nodes): |
|
204 | 204 | return sorted(nodes, key=keyfn)[0] |
|
205 | 205 | return picknext |
|
206 | 206 | |
|
207 | 207 | def makedatesorter(): |
|
208 | 208 | """Sort revisions by date.""" |
|
209 | 209 | dates = {} |
|
210 | 210 | def getdate(n): |
|
211 | 211 | if n not in dates: |
|
212 | 212 | dates[n] = util.parsedate(self.commitcache[n].date) |
|
213 | 213 | return dates[n] |
|
214 | 214 | |
|
215 | 215 | def picknext(nodes): |
|
216 | 216 | return min([(getdate(n), n) for n in nodes])[1] |
|
217 | 217 | |
|
218 | 218 | return picknext |
|
219 | 219 | |
|
220 | 220 | if sortmode == 'branchsort': |
|
221 | 221 | picknext = makebranchsorter() |
|
222 | 222 | elif sortmode == 'datesort': |
|
223 | 223 | picknext = makedatesorter() |
|
224 | 224 | elif sortmode == 'sourcesort': |
|
225 | 225 | picknext = makesourcesorter() |
|
226 | 226 | else: |
|
227 | 227 | raise util.Abort(_('unknown sort mode: %s') % sortmode) |
|
228 | 228 | |
|
229 | 229 | children, actives = mapchildren(parents) |
|
230 | 230 | |
|
231 | 231 | s = [] |
|
232 | 232 | pendings = {} |
|
233 | 233 | while actives: |
|
234 | 234 | n = picknext(actives) |
|
235 | 235 | actives.remove(n) |
|
236 | 236 | s.append(n) |
|
237 | 237 | |
|
238 | 238 | # Update dependents list |
|
239 | 239 | for c in children.get(n, []): |
|
240 | 240 | if c not in pendings: |
|
241 | 241 | pendings[c] = [p for p in parents[c] if p not in self.map] |
|
242 | 242 | try: |
|
243 | 243 | pendings[c].remove(n) |
|
244 | 244 | except ValueError: |
|
245 | 245 | raise util.Abort(_('cycle detected between %s and %s') |
|
246 | 246 | % (recode(c), recode(n))) |
|
247 | 247 | if not pendings[c]: |
|
248 | 248 | # Parents are converted, node is eligible |
|
249 | 249 | actives.insert(0, c) |
|
250 | 250 | pendings[c] = None |
|
251 | 251 | |
|
252 | 252 | if len(s) != len(parents): |
|
253 | 253 | raise util.Abort(_("not all revisions were sorted")) |
|
254 | 254 | |
|
255 | 255 | return s |
|
256 | 256 | |
|
257 | 257 | def writeauthormap(self): |
|
258 | 258 | authorfile = self.authorfile |
|
259 | 259 | if authorfile: |
|
260 | 260 | self.ui.status(_('Writing author map file %s\n') % authorfile) |
|
261 | 261 | ofile = open(authorfile, 'w+') |
|
262 | 262 | for author in self.authors: |
|
263 | 263 | ofile.write("%s=%s\n" % (author, self.authors[author])) |
|
264 | 264 | ofile.close() |
|
265 | 265 | |
|
266 | 266 | def readauthormap(self, authorfile): |
|
267 | 267 | afile = open(authorfile, 'r') |
|
268 | 268 | for line in afile: |
|
269 | 269 | |
|
270 | 270 | line = line.strip() |
|
271 | 271 | if not line or line.startswith('#'): |
|
272 | 272 | continue |
|
273 | 273 | |
|
274 | 274 | try: |
|
275 | 275 | srcauthor, dstauthor = line.split('=', 1) |
|
276 | 276 | except ValueError: |
|
277 | 277 | msg = _('Ignoring bad line in author map file %s: %s\n') |
|
278 | 278 | self.ui.warn(msg % (authorfile, line.rstrip())) |
|
279 | 279 | continue |
|
280 | 280 | |
|
281 | 281 | srcauthor = srcauthor.strip() |
|
282 | 282 | dstauthor = dstauthor.strip() |
|
283 | 283 | if self.authors.get(srcauthor) in (None, dstauthor): |
|
284 | 284 | msg = _('mapping author %s to %s\n') |
|
285 | 285 | self.ui.debug(msg % (srcauthor, dstauthor)) |
|
286 | 286 | self.authors[srcauthor] = dstauthor |
|
287 | 287 | continue |
|
288 | 288 | |
|
289 | 289 | m = _('overriding mapping for author %s, was %s, will be %s\n') |
|
290 | 290 | self.ui.status(m % (srcauthor, self.authors[srcauthor], dstauthor)) |
|
291 | 291 | |
|
292 | 292 | afile.close() |
|
293 | 293 | |
|
294 | 294 | def cachecommit(self, rev): |
|
295 | 295 | commit = self.source.getcommit(rev) |
|
296 | 296 | commit.author = self.authors.get(commit.author, commit.author) |
|
297 | 297 | commit.branch = self.branchmap.get(commit.branch, commit.branch) |
|
298 | 298 | self.commitcache[rev] = commit |
|
299 | 299 | return commit |
|
300 | 300 | |
|
301 | 301 | def copy(self, rev): |
|
302 | 302 | commit = self.commitcache[rev] |
|
303 | 303 | |
|
304 | 304 | changes = self.source.getchanges(rev) |
|
305 | 305 | if isinstance(changes, basestring): |
|
306 | 306 | if changes == SKIPREV: |
|
307 | 307 | dest = SKIPREV |
|
308 | 308 | else: |
|
309 | 309 | dest = self.map[changes] |
|
310 | 310 | self.map[rev] = dest |
|
311 | 311 | return |
|
312 | 312 | files, copies = changes |
|
313 | 313 | pbranches = [] |
|
314 | 314 | if commit.parents: |
|
315 | 315 | for prev in commit.parents: |
|
316 | 316 | if prev not in self.commitcache: |
|
317 | 317 | self.cachecommit(prev) |
|
318 | 318 | pbranches.append((self.map[prev], |
|
319 | 319 | self.commitcache[prev].branch)) |
|
320 | 320 | self.dest.setbranch(commit.branch, pbranches) |
|
321 | 321 | try: |
|
322 |
parents = self.splicemap[rev] |
|
|
322 | parents = self.splicemap[rev] | |
|
323 | 323 | self.ui.status(_('spliced in %s as parents of %s\n') % |
|
324 | 324 | (parents, rev)) |
|
325 | 325 | parents = [self.map.get(p, p) for p in parents] |
|
326 | 326 | except KeyError: |
|
327 | 327 | parents = [b[0] for b in pbranches] |
|
328 | 328 | source = progresssource(self.ui, self.source, len(files)) |
|
329 | 329 | newnode = self.dest.putcommit(files, copies, parents, commit, |
|
330 | 330 | source, self.map) |
|
331 | 331 | source.close() |
|
332 | 332 | self.source.converted(rev, newnode) |
|
333 | 333 | self.map[rev] = newnode |
|
334 | 334 | |
|
335 | 335 | def convert(self, sortmode): |
|
336 | 336 | try: |
|
337 | 337 | self.source.before() |
|
338 | 338 | self.dest.before() |
|
339 | 339 | self.source.setrevmap(self.map) |
|
340 | 340 | self.ui.status(_("scanning source...\n")) |
|
341 | 341 | heads = self.source.getheads() |
|
342 | 342 | parents = self.walktree(heads) |
|
343 | 343 | self.ui.status(_("sorting...\n")) |
|
344 | 344 | t = self.toposort(parents, sortmode) |
|
345 | 345 | num = len(t) |
|
346 | 346 | c = None |
|
347 | 347 | |
|
348 | 348 | self.ui.status(_("converting...\n")) |
|
349 | 349 | for i, c in enumerate(t): |
|
350 | 350 | num -= 1 |
|
351 | 351 | desc = self.commitcache[c].desc |
|
352 | 352 | if "\n" in desc: |
|
353 | 353 | desc = desc.splitlines()[0] |
|
354 | 354 | # convert log message to local encoding without using |
|
355 | 355 | # tolocal() because the encoding.encoding convert() |
|
356 | 356 | # uses is 'utf-8' |
|
357 | 357 | self.ui.status("%d %s\n" % (num, recode(desc))) |
|
358 | 358 | self.ui.note(_("source: %s\n") % recode(c)) |
|
359 | 359 | self.ui.progress(_('converting'), i, unit=_('revisions'), |
|
360 | 360 | total=len(t)) |
|
361 | 361 | self.copy(c) |
|
362 | 362 | self.ui.progress(_('converting'), None) |
|
363 | 363 | |
|
364 | 364 | tags = self.source.gettags() |
|
365 | 365 | ctags = {} |
|
366 | 366 | for k in tags: |
|
367 | 367 | v = tags[k] |
|
368 | 368 | if self.map.get(v, SKIPREV) != SKIPREV: |
|
369 | 369 | ctags[k] = self.map[v] |
|
370 | 370 | |
|
371 | 371 | if c and ctags: |
|
372 | 372 | nrev, tagsparent = self.dest.puttags(ctags) |
|
373 | 373 | if nrev and tagsparent: |
|
374 | 374 | # write another hash correspondence to override the previous |
|
375 | 375 | # one so we don't end up with extra tag heads |
|
376 | 376 | tagsparents = [e for e in self.map.iteritems() |
|
377 | 377 | if e[1] == tagsparent] |
|
378 | 378 | if tagsparents: |
|
379 | 379 | self.map[tagsparents[0][0]] = nrev |
|
380 | 380 | |
|
381 | 381 | bookmarks = self.source.getbookmarks() |
|
382 | 382 | cbookmarks = {} |
|
383 | 383 | for k in bookmarks: |
|
384 | 384 | v = bookmarks[k] |
|
385 | 385 | if self.map.get(v, SKIPREV) != SKIPREV: |
|
386 | 386 | cbookmarks[k] = self.map[v] |
|
387 | 387 | |
|
388 | 388 | if c and cbookmarks: |
|
389 | 389 | self.dest.putbookmarks(cbookmarks) |
|
390 | 390 | |
|
391 | 391 | self.writeauthormap() |
|
392 | 392 | finally: |
|
393 | 393 | self.cleanup() |
|
394 | 394 | |
|
395 | 395 | def cleanup(self): |
|
396 | 396 | try: |
|
397 | 397 | self.dest.after() |
|
398 | 398 | finally: |
|
399 | 399 | self.source.after() |
|
400 | 400 | self.map.close() |
|
401 | 401 | |
|
402 | 402 | def convert(ui, src, dest=None, revmapfile=None, **opts): |
|
403 | 403 | global orig_encoding |
|
404 | 404 | orig_encoding = encoding.encoding |
|
405 | 405 | encoding.encoding = 'UTF-8' |
|
406 | 406 | |
|
407 | 407 | # support --authors as an alias for --authormap |
|
408 | 408 | if not opts.get('authormap'): |
|
409 | 409 | opts['authormap'] = opts.get('authors') |
|
410 | 410 | |
|
411 | 411 | if not dest: |
|
412 | 412 | dest = hg.defaultdest(src) + "-hg" |
|
413 | 413 | ui.status(_("assuming destination %s\n") % dest) |
|
414 | 414 | |
|
415 | 415 | destc = convertsink(ui, dest, opts.get('dest_type')) |
|
416 | 416 | |
|
417 | 417 | try: |
|
418 | 418 | srcc, defaultsort = convertsource(ui, src, opts.get('source_type'), |
|
419 | 419 | opts.get('rev')) |
|
420 | 420 | except Exception: |
|
421 | 421 | for path in destc.created: |
|
422 | 422 | shutil.rmtree(path, True) |
|
423 | 423 | raise |
|
424 | 424 | |
|
425 | 425 | sortmodes = ('branchsort', 'datesort', 'sourcesort') |
|
426 | 426 | sortmode = [m for m in sortmodes if opts.get(m)] |
|
427 | 427 | if len(sortmode) > 1: |
|
428 | 428 | raise util.Abort(_('more than one sort mode specified')) |
|
429 | 429 | sortmode = sortmode and sortmode[0] or defaultsort |
|
430 | 430 | if sortmode == 'sourcesort' and not srcc.hasnativeorder(): |
|
431 | 431 | raise util.Abort(_('--sourcesort is not supported by this data source')) |
|
432 | 432 | |
|
433 | 433 | fmap = opts.get('filemap') |
|
434 | 434 | if fmap: |
|
435 | 435 | srcc = filemap.filemap_source(ui, srcc, fmap) |
|
436 | 436 | destc.setfilemapmode(True) |
|
437 | 437 | |
|
438 | 438 | if not revmapfile: |
|
439 | 439 | try: |
|
440 | 440 | revmapfile = destc.revmapfile() |
|
441 | 441 | except: |
|
442 | 442 | revmapfile = os.path.join(destc, "map") |
|
443 | 443 | |
|
444 | 444 | c = converter(ui, srcc, destc, revmapfile, opts) |
|
445 | 445 | c.convert(sortmode) |
|
446 | 446 |
@@ -1,79 +1,79 b'' | |||
|
1 | 1 | |
|
2 | 2 | $ echo "[extensions]" >> $HGRCPATH |
|
3 | 3 | $ echo "convert=" >> $HGRCPATH |
|
4 | 4 | $ echo 'graphlog =' >> $HGRCPATH |
|
5 | 5 | $ glog() |
|
6 | 6 | > { |
|
7 | 7 | > hg glog --template '{rev} "{desc|firstline}" files: {files}\n' "$@" |
|
8 | 8 | > } |
|
9 | 9 | $ hg init repo1 |
|
10 | 10 | $ cd repo1 |
|
11 | 11 | $ echo a > a |
|
12 | 12 | $ hg ci -Am adda |
|
13 | 13 | adding a |
|
14 | 14 | $ echo b > b |
|
15 | 15 | $ echo a >> a |
|
16 | 16 | $ hg ci -Am addb |
|
17 | 17 | adding b |
|
18 | 18 | $ PARENTID1=`hg id --debug -i` |
|
19 | 19 | $ echo c > c |
|
20 | 20 | $ hg ci -Am addc |
|
21 | 21 | adding c |
|
22 | 22 | $ PARENTID2=`hg id --debug -i` |
|
23 | 23 | $ cd .. |
|
24 | 24 | $ hg init repo2 |
|
25 | 25 | $ cd repo2 |
|
26 | 26 | $ echo b > a |
|
27 | 27 | $ echo d > d |
|
28 | 28 | $ hg ci -Am addaandd |
|
29 | 29 | adding a |
|
30 | 30 | adding d |
|
31 | 31 | $ CHILDID1=`hg id --debug -i` |
|
32 | 32 | $ echo d >> d |
|
33 | 33 | $ hg ci -Am changed |
|
34 | 34 | $ CHILDID2=`hg id --debug -i` |
|
35 | 35 | $ echo e > e |
|
36 | 36 | $ hg ci -Am adde |
|
37 | 37 | adding e |
|
38 | 38 | $ cd .. |
|
39 | 39 | |
|
40 | 40 | test invalid splicemap |
|
41 | 41 | |
|
42 | 42 | $ cat > splicemap <<EOF |
|
43 | 43 | > $CHILDID2 |
|
44 | 44 | > EOF |
|
45 | 45 | $ hg convert --splicemap splicemap repo2 repo1 |
|
46 |
abort: syntax error in splicemap(1): |
|
|
46 | abort: syntax error in splicemap(1): child parent1[,parent2] expected | |
|
47 | 47 | [255] |
|
48 | 48 | |
|
49 | 49 | splice repo2 on repo1 |
|
50 | 50 | |
|
51 | 51 | $ cat > splicemap <<EOF |
|
52 | 52 | > $CHILDID1 $PARENTID1 |
|
53 | 53 | > $CHILDID2 $PARENTID2,$CHILDID1 |
|
54 | 54 | > EOF |
|
55 | 55 | $ hg clone repo1 target1 |
|
56 | 56 | updating to branch default |
|
57 | 57 | 3 files updated, 0 files merged, 0 files removed, 0 files unresolved |
|
58 | 58 | $ hg convert --splicemap splicemap repo2 target1 |
|
59 | 59 | scanning source... |
|
60 | 60 | sorting... |
|
61 | 61 | converting... |
|
62 | 62 | 2 addaandd |
|
63 | 63 | spliced in ['6d4c2037ddc2cb2627ac3a244ecce35283268f8e'] as parents of 527cdedf31fbd5ea708aa14eeecf53d4676f38db |
|
64 | 64 | 1 changed |
|
65 | 65 | spliced in ['e55c719b85b60e5102fac26110ba626e7cb6b7dc', '527cdedf31fbd5ea708aa14eeecf53d4676f38db'] as parents of e4ea00df91897da3079a10fab658c1eddba6617b |
|
66 | 66 | 0 adde |
|
67 | 67 | $ glog -R target1 |
|
68 | 68 | o 5 "adde" files: e |
|
69 | 69 | | |
|
70 | 70 | o 4 "changed" files: d |
|
71 | 71 | |\ |
|
72 | 72 | | o 3 "addaandd" files: a d |
|
73 | 73 | | | |
|
74 | 74 | @ | 2 "addc" files: c |
|
75 | 75 | |/ |
|
76 | 76 | o 1 "addb" files: a b |
|
77 | 77 | | |
|
78 | 78 | o 0 "adda" files: a |
|
79 | 79 |
General Comments 0
You need to be logged in to leave comments.
Login now