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
72
from binascii import hexlify
38
73
from cStringIO import StringIO
40
from bzrlib.osutils import (local_time_offset, username,
41
rand_bytes, compact_date, user_email,
75
from bzrlib.atomicfile import AtomicFile
76
from bzrlib.osutils import (local_time_offset,
77
rand_bytes, compact_date,
42
78
kind_marker, is_inside_any, quotefn,
43
sha_string, sha_file, isdir, isfile)
44
from bzrlib.branch import gen_file_id
45
from bzrlib.errors import BzrError, PointlessCommit
46
from bzrlib.revision import Revision, RevisionReference
47
from bzrlib.trace import mutter, note
79
sha_file, isdir, isfile,
82
import bzrlib.errors as errors
83
from bzrlib.errors import (BzrError, PointlessCommit,
88
from bzrlib.revision import Revision
89
from bzrlib.testament import Testament
90
from bzrlib.trace import mutter, note, warning
48
91
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
92
from bzrlib.inventory import Inventory, ROOT_ID, InventoryEntry
93
from bzrlib.symbol_versioning import *
94
from bzrlib.workingtree import WorkingTree
97
@deprecated_function(zero_seven)
56
98
def commit(*args, **kwargs):
57
99
"""Commit a new revision to a branch.
99
169
working inventory.
101
171
def __init__(self,
103
174
if reporter is not None:
104
175
self.reporter = reporter
106
177
self.reporter = NullCommitReporter()
178
if config is not None:
184
branch=DEPRECATED_PARAMETER, message=None,
114
188
specific_files=None,
116
allow_pointless=True):
190
allow_pointless=True,
117
198
"""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.
200
branch -- the deprecated branch to commit to. New callers should pass in
203
message -- the commit message, a mandatory parameter
136
205
timestamp -- if not None, seconds-since-epoch for a
137
206
postdated/predated commit.
140
If true, commit only those files.
208
specific_files -- If true, commit only those files.
143
If set, use this as the new revision id.
210
rev_id -- If set, use this as the new revision id.
144
211
Useful for test or import commands that need to tightly
145
212
control what revisions are assigned. If you duplicate
146
213
a revision id that exists elsewhere it is your own fault.
147
214
If null (default), a time/random revision id is generated.
216
allow_pointless -- If true (default), commit even if nothing
217
has changed and no merges are recorded.
219
strict -- If true, don't allow a commit if the working tree
220
contains unknown files.
222
revprops -- Properties for new revision
223
:param local: Perform a local only commit.
151
self.branch.lock_write()
225
mutter('preparing to commit')
227
if deprecated_passed(branch):
228
warn("Commit.commit (branch, ...): The branch parameter is "
229
"deprecated as of bzr 0.8. Please use working_tree= instead.",
230
DeprecationWarning, stacklevel=2)
232
self.work_tree = self.branch.bzrdir.open_workingtree()
233
elif working_tree is None:
234
raise BzrError("One of branch and working_tree must be passed into commit().")
236
self.work_tree = working_tree
237
self.branch = self.work_tree.branch
239
raise BzrError("The message keyword parameter is required for commit().")
241
self.weave_store = self.branch.repository.weave_store
242
self.bound_branch = None
244
self.master_branch = None
245
self.master_locked = False
152
246
self.rev_id = rev_id
153
247
self.specific_files = specific_files
154
248
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
250
if revprops is not None:
251
self.revprops.update(revprops)
253
if reporter is None and self.reporter is None:
254
self.reporter = NullCommitReporter()
255
elif reporter is not None:
256
self.reporter = reporter
258
self.work_tree.lock_write()
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()
260
# setup the bound branch variables as needed.
261
self._check_bound_branch()
263
# check for out of date working trees
264
# if we are bound, then self.branch is the master branch and this
265
# test is thus all we need.
266
if self.work_tree.last_revision() != self.master_branch.last_revision():
267
raise errors.OutOfDateTree(self.work_tree)
270
# raise an exception as soon as we find a single unknown.
271
for unknown in self.work_tree.unknowns():
272
raise StrictCommitFailed()
274
if timestamp is None:
275
self.timestamp = time.time()
277
self.timestamp = long(timestamp)
279
if self.config is None:
280
self.config = bzrlib.config.BranchConfig(self.branch)
283
self.rev_id = _gen_revision_id(self.config, self.timestamp)
287
if committer is None:
288
self.committer = self.config.username()
290
assert isinstance(committer, basestring), type(committer)
291
self.committer = committer
294
self.timezone = local_time_offset()
296
self.timezone = int(timezone)
298
if isinstance(message, str):
299
message = message.decode(bzrlib.user_encoding)
300
assert isinstance(message, unicode), type(message)
301
self.message = message
302
self._escape_commit_message()
185
304
self.work_inv = self.work_tree.inventory
186
self.basis_tree = self.branch.basis_tree()
305
self.basis_tree = self.work_tree.basis_tree()
187
306
self.basis_inv = self.basis_tree.inventory
189
self.pending_merges = self.branch.pending_merges()
190
if self.pending_merges:
191
raise NotImplementedError("sorry, can't commit merges to the weave format yet")
308
self._gather_parents()
309
if len(self.parents) > 1 and self.specific_files:
310
raise NotImplementedError('selected-file commit of merges is not supported yet: files %r',
312
self._check_parents_present()
193
if self.rev_id is None:
194
self.rev_id = _gen_revision_id(self.branch, time.time())
196
# todo: update hashcache
197
self.delta = compare_trees(self.basis_tree, self.work_tree,
198
specific_files=self.specific_files)
200
if not (self.delta.has_changed()
201
or self.allow_pointless
202
or self.pending_merges):
314
self._remove_deleted()
315
self._populate_new_inv()
316
self._store_snapshot()
317
self._report_deletes()
319
if not (self.allow_pointless
320
or len(self.parents) > 1
321
or self.new_inv != self.basis_inv):
203
322
raise PointlessCommit()
205
self.new_inv = self.basis_inv.copy()
207
self.delta.show(sys.stdout)
209
self._remove_deleted()
212
self.branch._write_inventory(self.work_inv)
213
self._record_inventory()
324
if len(self.work_tree.conflicts())>0:
325
raise ConflictsInTree
327
self.inv_sha1 = self.branch.repository.add_inventory(
215
332
self._make_revision()
216
note('committted r%d {%s}', (self.branch.revno() + 1),
333
# revision data is in the local branch now.
335
# upload revision data to the master.
336
# this will propogate merged revisions too if needed.
337
if self.bound_branch:
338
self.master_branch.repository.fetch(self.branch.repository,
339
revision_id=self.rev_id)
340
# now the master has the revision data
341
# 'commit' to the master first so a timeout here causes the local
342
# branch to be out of date
343
self.master_branch.append_revision(self.rev_id)
345
# and now do the commit locally.
218
346
self.branch.append_revision(self.rev_id)
219
self.branch.set_pending_merges([])
348
self.work_tree.set_pending_merges([])
349
self.work_tree.set_last_revision(self.rev_id)
350
# now the work tree is up to date with the branch
352
self.reporter.completed(self.branch.revno(), self.rev_id)
353
if self.config.post_commit() is not None:
354
hooks = self.config.post_commit().split(' ')
355
# this would be nicer with twisted.python.reflect.namedAny
357
result = eval(hook + '(branch, rev_id)',
358
{'branch':self.branch,
360
'rev_id':self.rev_id})
224
def _record_inventory(self):
226
serializer_v5.write_inventory(self.new_inv, inv_tmp)
227
self.inv_sha1 = sha_string(inv_tmp.getvalue())
229
self.branch.inventory_store.add(inv_tmp, self.rev_id)
362
self._cleanup_bound_branch()
363
self.work_tree.unlock()
365
def _check_bound_branch(self):
366
"""Check to see if the local branch is bound.
368
If it is bound, then most of the commit will actually be
369
done using the remote branch as the target branch.
370
Only at the end will the local branch be updated.
372
if self.local and not self.branch.get_bound_location():
373
raise errors.LocalRequiresBoundBranch()
376
self.master_branch = self.branch.get_master_branch()
378
if not self.master_branch:
379
# make this branch the reference branch for out of date checks.
380
self.master_branch = self.branch
383
# If the master branch is bound, we must fail
384
master_bound_location = self.master_branch.get_bound_location()
385
if master_bound_location:
386
raise errors.CommitToDoubleBoundBranch(self.branch,
387
self.master_branch, master_bound_location)
389
# TODO: jam 20051230 We could automatically push local
390
# commits to the remote branch if they would fit.
391
# But for now, just require remote to be identical
394
# Make sure the local branch is identical to the master
395
master_rh = self.master_branch.revision_history()
396
local_rh = self.branch.revision_history()
397
if local_rh != master_rh:
398
raise errors.BoundBranchOutOfDate(self.branch,
401
# Now things are ready to change the master branch
403
self.bound_branch = self.branch
404
self.master_branch.lock_write()
405
self.master_locked = True
407
#### # Check to see if we have any pending merges. If we do
408
#### # those need to be pushed into the master branch
409
#### pending_merges = self.work_tree.pending_merges()
410
#### if pending_merges:
411
#### for revision_id in pending_merges:
412
#### self.master_branch.repository.fetch(self.bound_branch.repository,
413
#### revision_id=revision_id)
415
def _cleanup_bound_branch(self):
416
"""Executed at the end of a try/finally to cleanup a bound branch.
418
If the branch wasn't bound, this is a no-op.
419
If it was, it resents self.branch to the local branch, instead
422
if not self.bound_branch:
424
if self.master_locked:
425
self.master_branch.unlock()
427
def _escape_commit_message(self):
428
"""Replace xml-incompatible control characters."""
429
# FIXME: RBC 20060419 this should be done by the revision
430
# serialiser not by commit. Then we can also add an unescaper
431
# in the deserializer and start roundtripping revision messages
432
# precisely. See repository_implementations/test_repository.py
434
# Python strings can include characters that can't be
435
# represented in well-formed XML; escape characters that
436
# aren't listed in the XML specification
437
# (http://www.w3.org/TR/REC-xml/#NT-Char).
438
self.message, escape_count = re.subn(
439
u'[^\x09\x0A\x0D\u0020-\uD7FF\uE000-\uFFFD]+',
440
lambda match: match.group(0).encode('unicode_escape'),
443
self.reporter.escaped(escape_count, self.message)
445
def _gather_parents(self):
446
"""Record the parents of a merge for merge detection."""
447
pending_merges = self.work_tree.pending_merges()
449
self.parent_invs = []
450
self.present_parents = []
451
precursor_id = self.branch.last_revision()
453
self.parents.append(precursor_id)
454
self.parents += pending_merges
455
for revision in self.parents:
456
if self.branch.repository.has_revision(revision):
457
inventory = self.branch.repository.get_inventory(revision)
458
self.parent_invs.append(inventory)
459
self.present_parents.append(revision)
461
def _check_parents_present(self):
462
for parent_id in self.parents:
463
mutter('commit parent revision {%s}', parent_id)
464
if not self.branch.repository.has_revision(parent_id):
465
if parent_id == self.branch.last_revision():
466
warning("parent is missing %r", parent_id)
467
raise HistoryMissing(self.branch, 'revision', parent_id)
469
mutter("commit will ghost revision %r", parent_id)
232
471
def _make_revision(self):
233
472
"""Record a new revision object for this commit."""
234
self.rev = Revision(timestamp=self.timestamp,
235
timezone=self.timezone,
236
committer=self.committer,
237
message=self.message,
238
inventory_sha1=self.inv_sha1,
239
revision_id=self.rev_id)
241
self.rev.parents = []
242
precursor_id = self.branch.last_patch()
244
self.rev.parents.append(RevisionReference(precursor_id))
245
for merge_rev in self.pending_merges:
246
rev.parents.append(RevisionReference(merge_rev))
248
rev_tmp = tempfile.TemporaryFile()
249
serializer_v5.write_revision(self.rev, rev_tmp)
251
self.branch.revision_store.add(rev_tmp, self.rev_id)
252
mutter('new revision_id is {%s}', self.rev_id)
473
rev = Revision(timestamp=self.timestamp,
474
timezone=self.timezone,
475
committer=self.committer,
476
message=self.message,
477
inventory_sha1=self.inv_sha1,
478
revision_id=self.rev_id,
479
properties=self.revprops)
480
rev.parent_ids = self.parents
481
self.branch.repository.add_revision(self.rev_id, rev, self.new_inv, self.config)
255
483
def _remove_deleted(self):
256
"""Remove deleted files from the working and stored inventories."""
257
for path, id, kind in self.delta.removed:
258
if self.work_inv.has_id(id):
259
del self.work_inv[id]
260
if self.new_inv.has_id(id):
265
def _store_files(self):
266
"""Store new texts of modified/added files."""
267
# We must make sure that directories are added before anything
268
# inside them is added. the files within the delta report are
269
# sorted by path so we know the directory will come before its
271
for path, file_id, kind in self.delta.added:
273
ie = self.work_inv[file_id].copy()
276
self._store_file_text(file_id)
278
for path, file_id, kind in self.delta.modified:
281
self._store_file_text(file_id)
283
for old_path, new_path, file_id, kind, text_modified in self.delta.renamed:
286
if not text_modified:
288
self._store_file_text(file_id)
291
def _store_file_text(self, file_id):
292
"""Store updated text for one modified or added file."""
293
note('store new text for {%s} in revision {%s}',
294
file_id, self.rev_id)
295
new_lines = self.work_tree.get_file(file_id).readlines()
296
if file_id in self.new_inv: # was in basis inventory
297
ie = self.new_inv[file_id]
298
assert ie.file_id == file_id
299
assert file_id in self.basis_inv
300
assert self.basis_inv[file_id].kind == 'file'
301
old_version = self.basis_inv[file_id].text_version
302
file_parents = [old_version]
303
else: # new in this revision
304
ie = self.work_inv[file_id].copy()
306
assert file_id not in self.basis_inv
308
assert ie.kind == 'file'
309
self._add_text_to_weave(file_id, new_lines, file_parents)
310
# make a new inventory entry for this file, using whatever
311
# it had in the working copy, plus details on the new text
312
ie.text_sha1 = _sha_strings(new_lines)
313
ie.text_size = sum(map(len, new_lines))
314
ie.text_version = self.rev_id
315
ie.entry_version = self.rev_id
318
def _add_text_to_weave(self, file_id, new_lines, parents):
319
weave_fn = self.branch.controlfilename(['weaves', file_id+'.weave'])
320
if os.path.exists(weave_fn):
321
w = read_weave(file(weave_fn, 'rb'))
324
# XXX: Should set the appropriate parents by looking for this file_id
325
# in all revision parents
326
parent_idxs = map(w.lookup, parents)
327
w.add(self.rev_id, parent_idxs, new_lines)
328
af = AtomicFile(weave_fn)
330
write_weave_v5(w, af)
336
def _gen_revision_id(branch, when):
484
"""Remove deleted files from the working inventories.
486
This is done prior to taking the working inventory as the
487
basis for the new committed inventory.
489
This returns true if any files
490
*that existed in the basis inventory* were deleted.
491
Files that were added and deleted
492
in the working copy don't matter.
494
specific = self.specific_files
496
for path, ie in self.work_inv.iter_entries():
497
if specific and not is_inside_any(specific, path):
499
if not self.work_tree.has_filename(path):
500
self.reporter.missing(path)
501
deleted_ids.append((path, ie.file_id))
503
deleted_ids.sort(reverse=True)
504
for path, file_id in deleted_ids:
505
del self.work_inv[file_id]
506
self.work_tree._write_inventory(self.work_inv)
508
def _store_snapshot(self):
509
"""Pass over inventory and record a snapshot.
511
Entries get a new revision when they are modified in
512
any way, which includes a merge with a new set of
513
parents that have the same entry.
515
# XXX: Need to think more here about when the user has
516
# made a specific decision on a particular value -- c.f.
518
for path, ie in self.new_inv.iter_entries():
519
previous_entries = ie.find_previous_heads(
522
self.branch.repository.get_transaction())
523
if ie.revision is None:
524
# we are creating a new revision for ie in the history store
526
ie.snapshot(self.rev_id, path, previous_entries,
527
self.work_tree, self.weave_store,
528
self.branch.repository.get_transaction())
529
# describe the nature of the change that has occured relative to
530
# the basis inventory.
531
if (self.basis_inv.has_id(ie.file_id)):
532
basis_ie = self.basis_inv[ie.file_id]
535
change = ie.describe_change(basis_ie, ie)
536
if change in (InventoryEntry.RENAMED,
537
InventoryEntry.MODIFIED_AND_RENAMED):
538
old_path = self.basis_inv.id2path(ie.file_id)
539
self.reporter.renamed(change, old_path, path)
541
self.reporter.snapshot_change(change, path)
543
def _populate_new_inv(self):
544
"""Build revision inventory.
546
This creates a new empty inventory. Depending on
547
which files are selected for commit, and what is present in the
548
current tree, the new inventory is populated. inventory entries
549
which are candidates for modification have their revision set to
550
None; inventory entries that are carried over untouched have their
551
revision set to their prior value.
553
mutter("Selecting files for commit with filter %s", self.specific_files)
554
self.new_inv = Inventory(revision_id=self.rev_id)
555
for path, new_ie in self.work_inv.iter_entries():
556
file_id = new_ie.file_id
557
mutter('check %s {%s}', path, new_ie.file_id)
558
if self.specific_files:
559
if not is_inside_any(self.specific_files, path):
560
mutter('%s not selected for commit', path)
561
self._carry_entry(file_id)
564
# this is selected, ensure its parents are too.
565
parent_id = new_ie.parent_id
566
while parent_id != ROOT_ID:
567
if not self.new_inv.has_id(parent_id):
568
ie = self._select_entry(self.work_inv[parent_id])
569
mutter('%s selected for commit because of %s',
570
self.new_inv.id2path(parent_id), path)
572
ie = self.new_inv[parent_id]
573
if ie.revision is not None:
575
mutter('%s selected for commit because of %s',
576
self.new_inv.id2path(parent_id), path)
577
parent_id = ie.parent_id
578
mutter('%s selected for commit', path)
579
self._select_entry(new_ie)
581
def _select_entry(self, new_ie):
582
"""Make new_ie be considered for committing."""
588
def _carry_entry(self, file_id):
589
"""Carry the file unchanged from the basis revision."""
590
if self.basis_inv.has_id(file_id):
591
self.new_inv.add(self.basis_inv[file_id].copy())
593
def _report_deletes(self):
594
for path, ie in self.basis_inv.iter_entries():
595
if ie.file_id not in self.new_inv:
596
self.reporter.deleted(path)
598
def _gen_revision_id(config, when):
337
599
"""Return new revision-id."""
338
s = '%s-%s-' % (user_email(branch), compact_date(when))
600
s = '%s-%s-' % (config.user_email(), compact_date(when))
339
601
s += hexlify(rand_bytes(8))
343
def _sha_strings(strings):
344
"""Return the sha-1 of concatenation of strings"""
346
map(s.update, strings)
added
removed
bzrlib/commit.py
