1
# Copyright (C) 2005 Canonical Ltd
1
# Copyright (C) 2005, 2006 Canonical Ltd
3
3
# This program is free software; you can redistribute it and/or modify
4
4
# it under the terms of the GNU General Public License as published by
5
5
# the Free Software Foundation; either version 2 of the License, or
6
6
# (at your option) any later version.
8
8
# This program is distributed in the hope that it will be useful,
9
9
# but WITHOUT ANY WARRANTY; without even the implied warranty of
10
10
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11
11
# GNU General Public License for more details.
13
13
# You should have received a copy of the GNU General Public License
14
14
# along with this program; if not, write to the Free Software
15
15
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
18
## XXX: Can we do any better about making interrupted commits change
21
## XXX: If we merged two versions of a file then we still need to
22
## create a new version representing that merge, even if it didn't
23
## change from the parent.
25
## TODO: Read back the just-generated changeset, and make sure it
26
## applies and recreates the right state.
18
# XXX: Can we do any better about making interrupted commits change
19
# nothing? Perhaps the best approach is to integrate commit of
20
# AtomicFiles with releasing the lock on the Branch.
22
# TODO: Separate 'prepare' phase where we find a list of potentially
23
# committed files. We then can then pause the commit to prompt for a
24
# commit message, knowing the summary will be the same as what's
25
# actually used for the commit. (But perhaps simpler to simply get
26
# the tree status, then use that for a selective commit?)
28
# The newly committed revision is going to have a shape corresponding
29
# to that of the working inventory. Files that are not in the
30
# working tree and that were in the predecessor are reported as
31
# removed --- this can include files that were either removed from the
32
# inventory or deleted in the working tree. If they were only
33
# deleted from disk, they are removed from the working inventory.
35
# We then consider the remaining entries, which will be in the new
36
# version. Directory entries are simply copied across. File entries
37
# must be checked to see if a new version of the file should be
38
# recorded. For each parent revision inventory, we check to see what
39
# version of the file was present. If the file was present in at
40
# least one tree, and if it was the same version in all the trees,
41
# then we can just refer to that version. Otherwise, a new version
42
# representing the merger of the file versions must be added.
44
# TODO: Update hashcache before and after - or does the WorkingTree
47
# TODO: Rather than mashing together the ancestry and storing it back,
48
# perhaps the weave should have single method which does it all in one
49
# go, avoiding a lot of redundant work.
51
# TODO: Perhaps give a warning if one of the revisions marked as
52
# merged is already in the ancestry, and then don't record it as a
55
# TODO: If the file is newly merged but unchanged from the version it
56
# merges from, then it should still be reported as newly added
57
# relative to the basis revision.
59
# TODO: Do checks that the tree can be committed *before* running the
60
# editor; this should include checks for a pointless commit and for
61
# unknown or missing files.
63
# TODO: If commit fails, leave the message in a file somewhere.
37
from binascii import hexlify
38
72
from cStringIO import StringIO
40
from bzrlib.osutils import (local_time_offset, username,
41
rand_bytes, compact_date, user_email,
42
kind_marker, is_inside_any, quotefn,
43
sha_string, sha_file, isdir, isfile)
44
from bzrlib.branch import gen_file_id, INVENTORY_FILEID, ANCESTRY_FILEID
45
from bzrlib.errors import BzrError, PointlessCommit
46
from bzrlib.revision import Revision, RevisionReference
47
from bzrlib.trace import mutter, note
74
from bzrlib.atomicfile import AtomicFile
76
import bzrlib.errors as errors
77
from bzrlib.errors import (BzrError, PointlessCommit,
82
from bzrlib.osutils import (kind_marker, isdir,isfile, is_inside_any,
83
is_inside_or_parent_of_any,
84
quotefn, sha_file, split_lines)
85
from bzrlib.testament import Testament
86
from bzrlib.trace import mutter, note, warning
48
87
from bzrlib.xml5 import serializer_v5
49
from bzrlib.inventory import Inventory
50
from bzrlib.delta import compare_trees
51
from bzrlib.weave import Weave
52
from bzrlib.weavefile import read_weave, write_weave_v5
53
from bzrlib.atomicfile import AtomicFile
88
from bzrlib.inventory import Inventory, ROOT_ID, InventoryEntry
89
from bzrlib.symbol_versioning import *
90
from bzrlib.workingtree import WorkingTree
93
@deprecated_function(zero_seven)
56
94
def commit(*args, **kwargs):
57
95
"""Commit a new revision to a branch.
99
165
working inventory.
101
167
def __init__(self,
103
170
if reporter is not None:
104
171
self.reporter = reporter
106
173
self.reporter = NullCommitReporter()
174
if config is not None:
180
branch=DEPRECATED_PARAMETER, message=None,
114
184
specific_files=None,
116
allow_pointless=True):
186
allow_pointless=True,
117
194
"""Commit working copy as a new revision.
119
The basic approach is to add all the file texts into the
120
store, then the inventory, then make a new revision pointing
121
to that inventory and store that.
123
This is not quite safe if the working copy changes during the
124
commit; for the moment that is simply not allowed. A better
125
approach is to make a temporary copy of the files before
126
computing their hashes, and then add those hashes in turn to
127
the inventory. This should mean at least that there are no
128
broken hash pointers. There is no way we can get a snapshot
129
of the whole directory at an instant. This would also have to
130
be robust against files disappearing, moving, etc. So the
131
whole thing is a bit hard.
133
This raises PointlessCommit if there are no changes, no new merges,
134
and allow_pointless is false.
196
branch -- the deprecated branch to commit to. New callers should pass in
199
message -- the commit message, a mandatory parameter
136
201
timestamp -- if not None, seconds-since-epoch for a
137
202
postdated/predated commit.
140
If true, commit only those files.
204
specific_files -- If true, commit only those files.
143
If set, use this as the new revision id.
206
rev_id -- If set, use this as the new revision id.
144
207
Useful for test or import commands that need to tightly
145
208
control what revisions are assigned. If you duplicate
146
209
a revision id that exists elsewhere it is your own fault.
147
210
If null (default), a time/random revision id is generated.
212
allow_pointless -- If true (default), commit even if nothing
213
has changed and no merges are recorded.
215
strict -- If true, don't allow a commit if the working tree
216
contains unknown files.
218
revprops -- Properties for new revision
219
:param local: Perform a local only commit.
151
self.branch.lock_write()
221
mutter('preparing to commit')
223
if deprecated_passed(branch):
224
warn("Commit.commit (branch, ...): The branch parameter is "
225
"deprecated as of bzr 0.8. Please use working_tree= instead.",
226
DeprecationWarning, stacklevel=2)
228
self.work_tree = self.branch.bzrdir.open_workingtree()
229
elif working_tree is None:
230
raise BzrError("One of branch and working_tree must be passed into commit().")
232
self.work_tree = working_tree
233
self.branch = self.work_tree.branch
235
raise BzrError("The message keyword parameter is required for commit().")
237
self.weave_store = self.branch.repository.weave_store
238
self.bound_branch = None
240
self.master_branch = None
241
self.master_locked = False
153
243
self.specific_files = specific_files
154
244
self.allow_pointless = allow_pointless
156
if timestamp is None:
157
self.timestamp = time.time()
159
self.timestamp = long(timestamp)
161
if committer is None:
162
self.committer = username(self.branch)
164
assert isinstance(committer, basestring), type(committer)
165
self.committer = committer
168
self.timezone = local_time_offset()
170
self.timezone = int(timezone)
172
assert isinstance(message, basestring), type(message)
173
self.message = message
246
if reporter is None and self.reporter is None:
247
self.reporter = NullCommitReporter()
248
elif reporter is not None:
249
self.reporter = reporter
251
self.work_tree.lock_write()
252
self.pb = bzrlib.ui.ui_factory.nested_progress_bar()
176
# First walk over the working inventory; and both update that
177
# and also build a new revision inventory. The revision
178
# inventory needs to hold the text-id, sha1 and size of the
179
# actual file versions committed in the revision. (These are
180
# not present in the working inventory.) We also need to
181
# detect missing/deleted files, and remove them from the
184
self.work_tree = self.branch.working_tree()
254
# Cannot commit with conflicts present.
255
if len(self.work_tree.conflicts())>0:
256
raise ConflictsInTree
258
# setup the bound branch variables as needed.
259
self._check_bound_branch()
261
# check for out of date working trees
262
# if we are bound, then self.branch is the master branch and this
263
# test is thus all we need.
264
if self.work_tree.last_revision() != self.master_branch.last_revision():
265
raise errors.OutOfDateTree(self.work_tree)
268
# raise an exception as soon as we find a single unknown.
269
for unknown in self.work_tree.unknowns():
270
raise StrictCommitFailed()
272
if self.config is None:
273
self.config = bzrlib.config.BranchConfig(self.branch)
275
if isinstance(message, str):
276
message = message.decode(bzrlib.user_encoding)
277
assert isinstance(message, unicode), type(message)
278
self.message = message
279
self._escape_commit_message()
185
281
self.work_inv = self.work_tree.inventory
186
self.basis_tree = self.branch.basis_tree()
282
self.basis_tree = self.work_tree.basis_tree()
187
283
self.basis_inv = self.basis_tree.inventory
284
# one to finish, one for rev and inventory, and one for each
285
# inventory entry, and the same for the new inventory.
286
# note that this estimate is too long when we do a partial tree
287
# commit which excludes some new files from being considered.
288
# The estimate is corrected when we populate the new inv.
289
self.pb_total = len(self.work_inv) + 5
189
292
self._gather_parents()
191
if self.rev_id is None:
192
self.rev_id = _gen_revision_id(self.branch, time.time())
194
# TODO: update hashcache
195
self.delta = compare_trees(self.basis_tree, self.work_tree,
196
specific_files=self.specific_files)
198
if not (self.delta.has_changed()
199
or self.allow_pointless
200
or len(self.parents) != 1):
293
if len(self.parents) > 1 and self.specific_files:
294
raise NotImplementedError('selected-file commit of merges is not supported yet: files %r',
296
self._check_parents_present()
297
self.builder = self.branch.get_commit_builder(self.parents,
298
self.config, timestamp, timezone, committer, revprops, rev_id)
300
self._remove_deleted()
301
self._populate_new_inv()
302
self._report_deletes()
304
if not (self.allow_pointless
305
or len(self.parents) > 1
306
or self.builder.new_inventory != self.basis_inv):
201
307
raise PointlessCommit()
203
self.new_inv = self.basis_inv.copy()
205
self.delta.show(sys.stdout)
207
self._remove_deleted()
210
self.branch._write_inventory(self.work_inv)
211
self._record_inventory()
212
self._record_ancestry()
214
self._make_revision()
215
note('committted r%d {%s}', (self.branch.revno() + 1),
309
self._emit_progress_update()
310
# TODO: Now the new inventory is known, check for conflicts and prompt the
311
# user for a commit message.
312
self.builder.finish_inventory()
313
self._emit_progress_update()
314
self.rev_id = self.builder.commit(self.message)
315
self._emit_progress_update()
316
# revision data is in the local branch now.
318
# upload revision data to the master.
319
# this will propogate merged revisions too if needed.
320
if self.bound_branch:
321
self.master_branch.repository.fetch(self.branch.repository,
322
revision_id=self.rev_id)
323
# now the master has the revision data
324
# 'commit' to the master first so a timeout here causes the local
325
# branch to be out of date
326
self.master_branch.append_revision(self.rev_id)
328
# and now do the commit locally.
217
329
self.branch.append_revision(self.rev_id)
218
self.branch.set_pending_merges([])
331
self.work_tree.set_pending_merges([])
332
self.work_tree.set_last_revision(self.rev_id)
333
# now the work tree is up to date with the branch
335
self.reporter.completed(self.branch.revno(), self.rev_id)
336
if self.config.post_commit() is not None:
337
hooks = self.config.post_commit().split(' ')
338
# this would be nicer with twisted.python.reflect.namedAny
340
result = eval(hook + '(branch, rev_id)',
341
{'branch':self.branch,
343
'rev_id':self.rev_id})
344
self._emit_progress_update()
223
def _record_inventory(self):
224
"""Store the inventory for the new revision."""
226
serializer_v5.write_inventory(self.new_inv, inv_tmp)
228
self.inv_sha1 = sha_string(inv_tmp.getvalue())
229
inv_lines = inv_tmp.readlines()
230
self.branch.weave_store.add_text(INVENTORY_FILEID, self.rev_id,
231
inv_lines, self.parents)
234
def _record_ancestry(self):
235
"""Append merged revision ancestry to the ancestry file."""
236
if len(self.parents) > 1:
237
raise NotImplementedError("sorry, can't commit merges yet")
238
w = self.branch.weave_store.get_weave_or_empty(ANCESTRY_FILEID)
240
lines = w.get(w.lookup(self.parents[0]))
243
lines.append(self.rev_id + '\n')
244
parent_idxs = map(w.lookup, self.parents)
245
w.add(self.rev_id, parent_idxs, lines)
246
self.branch.weave_store.put_weave(ANCESTRY_FILEID, w)
348
def _check_bound_branch(self):
349
"""Check to see if the local branch is bound.
351
If it is bound, then most of the commit will actually be
352
done using the remote branch as the target branch.
353
Only at the end will the local branch be updated.
355
if self.local and not self.branch.get_bound_location():
356
raise errors.LocalRequiresBoundBranch()
359
self.master_branch = self.branch.get_master_branch()
361
if not self.master_branch:
362
# make this branch the reference branch for out of date checks.
363
self.master_branch = self.branch
366
# If the master branch is bound, we must fail
367
master_bound_location = self.master_branch.get_bound_location()
368
if master_bound_location:
369
raise errors.CommitToDoubleBoundBranch(self.branch,
370
self.master_branch, master_bound_location)
372
# TODO: jam 20051230 We could automatically push local
373
# commits to the remote branch if they would fit.
374
# But for now, just require remote to be identical
377
# Make sure the local branch is identical to the master
378
master_rh = self.master_branch.revision_history()
379
local_rh = self.branch.revision_history()
380
if local_rh != master_rh:
381
raise errors.BoundBranchOutOfDate(self.branch,
384
# Now things are ready to change the master branch
386
self.bound_branch = self.branch
387
self.master_branch.lock_write()
388
self.master_locked = True
390
#### # Check to see if we have any pending merges. If we do
391
#### # those need to be pushed into the master branch
392
#### pending_merges = self.work_tree.pending_merges()
393
#### if pending_merges:
394
#### for revision_id in pending_merges:
395
#### self.master_branch.repository.fetch(self.bound_branch.repository,
396
#### revision_id=revision_id)
399
"""Cleanup any open locks, progress bars etc."""
400
cleanups = [self._cleanup_bound_branch,
401
self.work_tree.unlock,
403
found_exception = None
404
for cleanup in cleanups:
407
# we want every cleanup to run no matter what.
408
# so we have a catchall here, but we will raise the
409
# last encountered exception up the stack: and
410
# typically this will be useful enough.
413
if found_exception is not None:
414
# dont do a plan raise, because the last exception may have been
415
# trashed, e is our sure-to-work exception even though it loses the
416
# full traceback. XXX: RBC 20060421 perhaps we could check the
417
# exc_info and if its the same one do a plain raise otherwise
418
# 'raise e' as we do now.
421
def _cleanup_bound_branch(self):
422
"""Executed at the end of a try/finally to cleanup a bound branch.
424
If the branch wasn't bound, this is a no-op.
425
If it was, it resents self.branch to the local branch, instead
428
if not self.bound_branch:
430
if self.master_locked:
431
self.master_branch.unlock()
433
def _escape_commit_message(self):
434
"""Replace xml-incompatible control characters."""
435
# FIXME: RBC 20060419 this should be done by the revision
436
# serialiser not by commit. Then we can also add an unescaper
437
# in the deserializer and start roundtripping revision messages
438
# precisely. See repository_implementations/test_repository.py
440
# Python strings can include characters that can't be
441
# represented in well-formed XML; escape characters that
442
# aren't listed in the XML specification
443
# (http://www.w3.org/TR/REC-xml/#NT-Char).
444
self.message, escape_count = re.subn(
445
u'[^\x09\x0A\x0D\u0020-\uD7FF\uE000-\uFFFD]+',
446
lambda match: match.group(0).encode('unicode_escape'),
449
self.reporter.escaped(escape_count, self.message)
249
451
def _gather_parents(self):
250
pending_merges = self.branch.pending_merges()
252
raise NotImplementedError("sorry, can't commit merges to the weave format yet")
452
"""Record the parents of a merge for merge detection."""
453
# TODO: Make sure that this list doesn't contain duplicate
454
# entries and the order is preserved when doing this.
455
pending_merges = self.work_tree.pending_merges()
253
456
self.parents = []
254
precursor_id = self.branch.last_patch()
457
self.parent_invs = []
458
precursor_id = self.branch.last_revision()
256
460
self.parents.append(precursor_id)
257
461
self.parents += pending_merges
260
def _make_revision(self):
261
"""Record a new revision object for this commit."""
262
self.rev = Revision(timestamp=self.timestamp,
263
timezone=self.timezone,
264
committer=self.committer,
265
message=self.message,
266
inventory_sha1=self.inv_sha1,
267
revision_id=self.rev_id)
268
self.rev.parents = map(RevisionReference, self.parents)
269
rev_tmp = tempfile.TemporaryFile()
270
serializer_v5.write_revision(self.rev, rev_tmp)
272
self.branch.revision_store.add(rev_tmp, self.rev_id)
273
mutter('new revision_id is {%s}', self.rev_id)
462
for revision in self.parents:
463
if self.branch.repository.has_revision(revision):
464
inventory = self.branch.repository.get_inventory(revision)
465
self.parent_invs.append(inventory)
467
def _check_parents_present(self):
468
for parent_id in self.parents:
469
mutter('commit parent revision {%s}', parent_id)
470
if not self.branch.repository.has_revision(parent_id):
471
if parent_id == self.branch.last_revision():
472
warning("parent is missing %r", parent_id)
473
raise HistoryMissing(self.branch, 'revision', parent_id)
475
mutter("commit will ghost revision %r", parent_id)
276
477
def _remove_deleted(self):
277
"""Remove deleted files from the working and stored inventories."""
278
for path, id, kind in self.delta.removed:
279
if self.work_inv.has_id(id):
280
del self.work_inv[id]
281
if self.new_inv.has_id(id):
286
def _store_files(self):
287
"""Store new texts of modified/added files."""
288
# We must make sure that directories are added before anything
289
# inside them is added. the files within the delta report are
290
# sorted by path so we know the directory will come before its
292
for path, file_id, kind in self.delta.added:
294
ie = self.work_inv[file_id].copy()
297
self._store_file_text(file_id)
299
for path, file_id, kind in self.delta.modified:
302
self._store_file_text(file_id)
304
for old_path, new_path, file_id, kind, text_modified in self.delta.renamed:
307
if not text_modified:
309
self._store_file_text(file_id)
312
def _store_file_text(self, file_id):
313
"""Store updated text for one modified or added file."""
314
note('store new text for {%s} in revision {%s}',
315
file_id, self.rev_id)
316
new_lines = self.work_tree.get_file(file_id).readlines()
317
if file_id in self.new_inv: # was in basis inventory
318
ie = self.new_inv[file_id]
319
assert ie.file_id == file_id
320
assert file_id in self.basis_inv
321
assert self.basis_inv[file_id].kind == 'file'
322
old_version = self.basis_inv[file_id].text_version
323
file_parents = [old_version]
324
else: # new in this revision
325
ie = self.work_inv[file_id].copy()
327
assert file_id not in self.basis_inv
329
assert ie.kind == 'file'
330
self._add_text_to_weave(file_id, new_lines, file_parents)
331
# make a new inventory entry for this file, using whatever
332
# it had in the working copy, plus details on the new text
333
ie.text_sha1 = _sha_strings(new_lines)
334
ie.text_size = sum(map(len, new_lines))
335
ie.text_version = self.rev_id
336
ie.entry_version = self.rev_id
339
def _add_text_to_weave(self, file_id, new_lines, parents):
340
if file_id.startswith('__'):
341
raise ValueError('illegal file-id %r for text file' % file_id)
342
self.branch.weave_store.add_text(file_id, self.rev_id, new_lines, parents)
345
def _gen_revision_id(branch, when):
346
"""Return new revision-id."""
347
s = '%s-%s-' % (user_email(branch), compact_date(when))
348
s += hexlify(rand_bytes(8))
352
def _sha_strings(strings):
353
"""Return the sha-1 of concatenation of strings"""
355
map(s.update, strings)
478
"""Remove deleted files from the working inventories.
480
This is done prior to taking the working inventory as the
481
basis for the new committed inventory.
483
This returns true if any files
484
*that existed in the basis inventory* were deleted.
485
Files that were added and deleted
486
in the working copy don't matter.
488
specific = self.specific_files
490
for path, ie in self.work_inv.iter_entries():
491
if specific and not is_inside_any(specific, path):
493
if not self.work_tree.has_filename(path):
494
self.reporter.missing(path)
495
deleted_ids.append((path, ie.file_id))
497
deleted_ids.sort(reverse=True)
498
for path, file_id in deleted_ids:
499
del self.work_inv[file_id]
500
self.work_tree._write_inventory(self.work_inv)
502
def _populate_new_inv(self):
503
"""Build revision inventory.
505
This creates a new empty inventory. Depending on
506
which files are selected for commit, and what is present in the
507
current tree, the new inventory is populated. inventory entries
508
which are candidates for modification have their revision set to
509
None; inventory entries that are carried over untouched have their
510
revision set to their prior value.
512
mutter("Selecting files for commit with filter %s", self.specific_files)
513
# iter_entries does not visit the ROOT_ID node so we need to call
514
# self._emit_progress_update once by hand.
515
self._emit_progress_update()
516
for path, new_ie in self.work_inv.iter_entries():
517
self._emit_progress_update()
518
file_id = new_ie.file_id
519
mutter('check %s {%s}', path, file_id)
520
if (not self.specific_files or
521
is_inside_or_parent_of_any(self.specific_files, path)):
522
mutter('%s selected for commit', path)
526
mutter('%s not selected for commit', path)
527
if self.basis_inv.has_id(file_id):
528
ie = self.basis_inv[file_id].copy()
530
# this entry is new and not being committed
533
self.builder.record_entry_contents(ie, self.parent_invs,
534
path, self.work_tree)
535
# describe the nature of the change that has occured relative to
536
# the basis inventory.
537
if (self.basis_inv.has_id(ie.file_id)):
538
basis_ie = self.basis_inv[ie.file_id]
541
change = ie.describe_change(basis_ie, ie)
542
if change in (InventoryEntry.RENAMED,
543
InventoryEntry.MODIFIED_AND_RENAMED):
544
old_path = self.basis_inv.id2path(ie.file_id)
545
self.reporter.renamed(change, old_path, path)
547
self.reporter.snapshot_change(change, path)
549
def _emit_progress_update(self):
550
"""Emit an update to the progress bar."""
551
self.pb.update("Committing", self.pb_count, self.pb_total)
554
def _report_deletes(self):
555
for path, ie in self.basis_inv.iter_entries():
556
if ie.file_id not in self.builder.new_inventory:
557
self.reporter.deleted(path)
added
removed
bzrlib/commit.py
