1
# Copyright (C) 2005 Canonical Ltd
3
# This program is free software; you can redistribute it and/or modify
4
# it under the terms of the GNU General Public License as published by
5
# the Free Software Foundation; either version 2 of the License, or
6
# (at your option) any later version.
8
# This program is distributed in the hope that it will be useful,
9
# but WITHOUT ANY WARRANTY; without even the implied warranty of
10
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11
# GNU General Public License for more details.
13
# You should have received a copy of the GNU General Public License
14
# along with this program; if not, write to the Free Software
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
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
# This code requires all merge parents to be present in the branch.
48
# We could relax this but for the sake of simplicity the constraint is
49
# here for now. It's not totally clear to me how we'd know which file
50
# need new text versions if some parents are absent. -- mbp 20050915
52
# TODO: Rather than mashing together the ancestry and storing it back,
53
# perhaps the weave should have single method which does it all in one
54
# go, avoiding a lot of redundant work.
56
# TODO: Perhaps give a warning if one of the revisions marked as
57
# merged is already in the ancestry, and then don't record it as a
60
# TODO: If the file is newly merged but unchanged from the version it
61
# merges from, then it should still be reported as newly added
62
# relative to the basis revision.
71
from binascii import hexlify
72
from cStringIO import StringIO
74
from bzrlib.osutils import (local_time_offset, username,
75
rand_bytes, compact_date, user_email,
76
kind_marker, is_inside_any, quotefn,
77
sha_string, sha_strings, sha_file, isdir, isfile,
79
from bzrlib.branch import gen_file_id
80
from bzrlib.errors import (BzrError, PointlessCommit,
83
from bzrlib.revision import Revision
84
from bzrlib.trace import mutter, note, warning
85
from bzrlib.xml5 import serializer_v5
86
from bzrlib.inventory import Inventory
87
from bzrlib.weave import Weave
88
from bzrlib.weavefile import read_weave, write_weave_v5
89
from bzrlib.atomicfile import AtomicFile
92
def commit(*args, **kwargs):
93
"""Commit a new revision to a branch.
95
Function-style interface for convenience of old callers.
97
New code should use the Commit class instead.
99
## XXX: Remove this in favor of Branch.commit?
100
Commit().commit(*args, **kwargs)
103
class NullCommitReporter(object):
104
"""I report on progress of a commit."""
105
def added(self, path):
108
def removed(self, path):
111
def renamed(self, old_path, new_path):
115
class ReportCommitToLog(NullCommitReporter):
116
def added(self, path):
117
note('added %s', path)
119
def removed(self, path):
120
note('removed %s', path)
122
def renamed(self, old_path, new_path):
123
note('renamed %s => %s', old_path, new_path)
126
class Commit(object):
127
"""Task of committing a new revision.
129
This is a MethodObject: it accumulates state as the commit is
130
prepared, and then it is discarded. It doesn't represent
131
historical revisions, just the act of recording a new one.
134
Modified to hold a list of files that have been deleted from
135
the working directory; these should be removed from the
140
if reporter is not None:
141
self.reporter = reporter
143
self.reporter = NullCommitReporter()
153
allow_pointless=True,
155
"""Commit working copy as a new revision.
157
timestamp -- if not None, seconds-since-epoch for a
158
postdated/predated commit.
160
specific_files -- If true, commit only those files.
162
rev_id -- If set, use this as the new revision id.
163
Useful for test or import commands that need to tightly
164
control what revisions are assigned. If you duplicate
165
a revision id that exists elsewhere it is your own fault.
166
If null (default), a time/random revision id is generated.
168
allow_pointless -- If true (default), commit even if nothing
169
has changed and no merges are recorded.
171
mutter('preparing to commit')
174
self.weave_store = branch.weave_store
176
self.specific_files = specific_files
177
self.allow_pointless = allow_pointless
179
if timestamp is None:
180
self.timestamp = time.time()
182
self.timestamp = long(timestamp)
185
self.rev_id = _gen_revision_id(self.branch, self.timestamp)
189
if committer is None:
190
self.committer = username(self.branch)
192
assert isinstance(committer, basestring), type(committer)
193
self.committer = committer
196
self.timezone = local_time_offset()
198
self.timezone = int(timezone)
200
assert isinstance(message, basestring), type(message)
201
self.message = message
202
self._escape_commit_message()
204
self.branch.lock_write()
206
self.work_tree = self.branch.working_tree()
207
self.work_inv = self.work_tree.inventory
208
self.basis_tree = self.branch.basis_tree()
209
self.basis_inv = self.basis_tree.inventory
211
self._gather_parents()
212
if len(self.parents) > 1 and self.specific_files:
213
raise NotImplementedError('selected-file commit of merges is not supported yet')
214
self._check_parents_present()
216
self._remove_deleted()
217
self.new_inv = Inventory()
218
self._store_entries()
219
self._report_deletes()
220
self._set_name_versions()
222
if not (self.allow_pointless
223
or len(self.parents) > 1
224
or self.new_inv != self.basis_inv):
225
raise PointlessCommit()
227
self._record_inventory()
228
self._record_ancestry()
229
self._make_revision()
230
note('committed r%d {%s}', (self.branch.revno() + 1),
232
self.branch.append_revision(self.rev_id)
233
self.branch.set_pending_merges([])
239
def _record_inventory(self):
240
"""Store the inventory for the new revision."""
241
inv_text = serializer_v5.write_inventory_to_string(self.new_inv)
242
self.inv_sha1 = sha_string(inv_text)
243
s = self.branch.control_weaves
244
s.add_text('inventory', self.rev_id,
245
split_lines(inv_text), self.parents)
247
def _escape_commit_message(self):
248
"""Replace xml-incompatible control characters."""
249
# Python strings can include characters that can't be
250
# represented in well-formed XML; escape characters that
251
# aren't listed in the XML specification
252
# (http://www.w3.org/TR/REC-xml/#NT-Char).
253
if isinstance(self.message, unicode):
254
char_pattern = u'[^\x09\x0A\x0D\u0020-\uD7FF\uE000-\uFFFD]'
256
# Use a regular 'str' as pattern to avoid having re.subn
257
# return 'unicode' results.
258
char_pattern = '[^x09\x0A\x0D\x20-\xFF]'
259
self.message, escape_count = re.subn(
261
lambda match: match.group(0).encode('unicode_escape'),
264
note("replaced %d control characters in message", escape_count)
266
def _record_ancestry(self):
267
"""Append merged revision ancestry to the ancestry file.
269
This should be the merged ancestry of all parents, plus the
271
s = self.branch.control_weaves
272
w = s.get_weave_or_empty('ancestry')
273
lines = self._make_ancestry(w)
274
w.add(self.rev_id, self.parents, lines)
275
s.put_weave('ancestry', w)
278
def _make_ancestry(self, ancestry_weave):
279
"""Return merged ancestry lines.
281
The lines are revision-ids followed by newlines."""
282
parent_ancestries = [ancestry_weave.get(p) for p in self.parents]
283
new_lines = merge_ancestry_lines(self.rev_id, parent_ancestries)
284
mutter('merged ancestry of {%s}:\n%s', self.rev_id, ''.join(new_lines))
288
def _gather_parents(self):
289
pending_merges = self.branch.pending_merges()
291
self.parent_trees = []
292
precursor_id = self.branch.last_revision()
294
self.parents.append(precursor_id)
295
self.parent_trees.append(self.basis_tree)
296
self.parents += pending_merges
297
self.parent_trees.extend(map(self.branch.revision_tree, pending_merges))
300
def _check_parents_present(self):
301
for parent_id in self.parents:
302
mutter('commit parent revision {%s}', parent_id)
303
if not self.branch.has_revision(parent_id):
304
warning("can't commit a merge from an absent parent")
305
raise HistoryMissing(self.branch, 'revision', parent_id)
308
def _make_revision(self):
309
"""Record a new revision object for this commit."""
310
self.rev = Revision(timestamp=self.timestamp,
311
timezone=self.timezone,
312
committer=self.committer,
313
message=self.message,
314
inventory_sha1=self.inv_sha1,
315
revision_id=self.rev_id)
316
self.rev.parent_ids = self.parents
318
serializer_v5.write_revision(self.rev, rev_tmp)
320
self.branch.revision_store.add(rev_tmp, self.rev_id)
321
mutter('new revision_id is {%s}', self.rev_id)
324
def _remove_deleted(self):
325
"""Remove deleted files from the working inventories.
327
This is done prior to taking the working inventory as the
328
basis for the new committed inventory.
330
This returns true if any files
331
*that existed in the basis inventory* were deleted.
332
Files that were added and deleted
333
in the working copy don't matter.
335
specific = self.specific_files
337
for path, ie in self.work_inv.iter_entries():
338
if specific and not is_inside_any(specific, path):
340
if not self.work_tree.has_filename(path):
341
note('missing %s', path)
342
deleted_ids.append(ie.file_id)
344
for file_id in deleted_ids:
345
if file_id in self.work_inv:
346
del self.work_inv[file_id]
347
self.branch._write_inventory(self.work_inv)
350
def _find_file_parents(self, file_id):
351
"""Return the text versions and hashes for all file parents.
353
Returned as a map from text version to inventory entry.
355
This is a set containing the file versions in all parents
356
revisions containing the file. If the file is new, the set
359
for tree in self.parent_trees:
360
if file_id in tree.inventory:
361
ie = tree.inventory[file_id]
362
assert ie.kind == 'file'
363
assert ie.file_id == file_id
364
if ie.text_version in r:
365
assert r[ie.text_version] == ie
367
r[ie.text_version] = ie
371
def _set_name_versions(self):
372
"""Pass over inventory and mark new entry version as needed.
374
Files get a new name version when they are new, have a
375
different parent, or a different name from in the
376
basis inventory, or if the file is in a different place
377
to any of the parents."""
378
# XXX: Need to think more here about when the user has
379
# made a specific decision on a particular value -- c.f.
381
for path, ie in self.new_inv.iter_entries():
384
for parent_tree in self.parent_trees:
385
parent_inv = parent_tree.inventory
386
if file_id not in parent_inv:
388
parent_ie = parent_inv[file_id]
389
if parent_ie.parent_id != ie.parent_id:
392
elif parent_ie.name != ie.name:
395
elif old_version is None:
396
old_version = parent_ie.name_version
397
elif old_version != parent_ie.name_version:
401
pass # so far so good
402
if old_version is None:
403
mutter('new name_version for {%s}', file_id)
404
ie.name_version = self.rev_id
406
mutter('name_version for {%s} inherited as {%s}',
407
file_id, old_version)
408
ie.name_version = old_version
411
def _store_entries(self):
412
"""Build revision inventory and store modified files.
414
This is called with new_inv a new empty inventory. Depending on
415
which files are selected for commit, and which ones have
416
been modified or merged, new inventory entries are built
417
based on the working and parent inventories.
419
As a side-effect this stores new text versions for committed
420
files with text changes or merges.
422
Each entry can have one of several things happen:
424
carry_file -- carried from the previous version (if not
427
commit_nonfile -- no text to worry about
429
commit_old_text -- same text, may have moved
431
commit_file -- new text version
433
for path, new_ie in self.work_inv.iter_entries():
434
file_id = new_ie.file_id
435
mutter('check %s {%s}', path, new_ie.file_id)
436
if self.specific_files:
437
if not is_inside_any(self.specific_files, path):
438
mutter('%s not selected for commit', path)
439
self._carry_file(file_id)
441
if new_ie.kind != 'file':
442
self._commit_nonfile(file_id)
445
file_parents = self._find_file_parents(file_id)
446
mutter('parents of %s are %r', path, file_parents)
447
if len(file_parents) == 1:
448
parent_ie = file_parents.values()[0]
449
wc_sha1 = self.work_tree.get_file_sha1(file_id)
450
if parent_ie.text_sha1 == wc_sha1:
451
# text not changed or merged
452
self._commit_old_text(file_id, parent_ie)
454
# file is either new, or a file merge; need to record
456
if len(file_parents) > 1:
457
note('merged %s', path)
458
elif len(file_parents) == 0:
459
note('added %s', path)
461
note('modified %s', path)
462
self._commit_file(new_ie, file_id, file_parents)
465
def _commit_nonfile(self, file_id):
466
self.new_inv.add(self.work_inv[file_id].copy())
469
def _carry_file(self, file_id):
470
"""Carry the file unchanged from the basis revision."""
471
if self.basis_inv.has_id(file_id):
472
self.new_inv.add(self.basis_inv[file_id].copy())
475
def _commit_old_text(self, file_id, parent_ie):
476
"""Keep the same text as last time, but possibly a different name."""
477
ie = self.work_inv[file_id].copy()
478
ie.text_version = parent_ie.text_version
479
ie.text_size = parent_ie.text_size
480
ie.text_sha1 = parent_ie.text_sha1
484
def _report_deletes(self):
485
for file_id in self.basis_inv:
486
if file_id not in self.new_inv:
487
note('deleted %s', self.basis_inv.id2path(file_id))
490
def _commit_file(self, new_ie, file_id, file_parents):
491
mutter('store new text for {%s} in revision {%s}',
492
file_id, self.rev_id)
493
new_lines = self.work_tree.get_file(file_id).readlines()
494
self._add_text_to_weave(file_id, new_lines, file_parents)
495
new_ie.text_version = self.rev_id
496
new_ie.text_sha1 = sha_strings(new_lines)
497
new_ie.text_size = sum(map(len, new_lines))
498
self.new_inv.add(new_ie)
501
def _add_text_to_weave(self, file_id, new_lines, parents):
502
self.weave_store.add_text(file_id, self.rev_id, new_lines, parents)
505
def _gen_revision_id(branch, when):
506
"""Return new revision-id."""
507
s = '%s-%s-' % (user_email(branch), compact_date(when))
508
s += hexlify(rand_bytes(8))
514
def merge_ancestry_lines(rev_id, ancestries):
515
"""Return merged ancestry lines.
517
rev_id -- id of the new revision
519
ancestries -- a sequence of ancestries for parent revisions,
520
as newline-terminated line lists.
522
if len(ancestries) == 0:
523
return [rev_id + '\n']
524
seen = set(ancestries[0])
525
ancs = ancestries[0][:]
526
for parent_ancestry in ancestries[1:]:
527
for line in parent_ancestry:
528
assert line[-1] == '\n'
added
removed
bzrlib/commit.py
