1
1
# Copyright (C) 2005 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
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.
66
from binascii import hexlify
67
from cStringIO import StringIO
69
from bzrlib.osutils import (local_time_offset,
70
rand_bytes, compact_date,
71
kind_marker, is_inside_any, quotefn,
72
sha_string, sha_strings, sha_file, isdir, isfile,
74
from bzrlib.branch import gen_file_id
76
from bzrlib.errors import (BzrError, PointlessCommit,
80
from bzrlib.revision import Revision
81
from bzrlib.trace import mutter, note, warning
82
from bzrlib.xml5 import serializer_v5
83
from bzrlib.inventory import Inventory, ROOT_ID
84
from bzrlib.weave import Weave
85
from bzrlib.weavefile import read_weave, write_weave_v5
86
from bzrlib.atomicfile import AtomicFile
89
def commit(*args, **kwargs):
90
"""Commit a new revision to a branch.
92
Function-style interface for convenience of old callers.
94
New code should use the Commit class instead.
19
def commit(branch, message,
26
"""Commit working copy as a new revision.
28
The basic approach is to add all the file texts into the
29
store, then the inventory, then make a new revision pointing
30
to that inventory and store that.
32
This is not quite safe if the working copy changes during the
33
commit; for the moment that is simply not allowed. A better
34
approach is to make a temporary copy of the files before
35
computing their hashes, and then add those hashes in turn to
36
the inventory. This should mean at least that there are no
37
broken hash pointers. There is no way we can get a snapshot
38
of the whole directory at an instant. This would also have to
39
be robust against files disappearing, moving, etc. So the
40
whole thing is a bit hard.
42
timestamp -- if not None, seconds-since-epoch for a
43
postdated/predated commit.
46
If true, commit only those files.
49
If set, use this as the new revision id.
50
Useful for test or import commands that need to tightly
51
control what revisions are assigned. If you duplicate
52
a revision id that exists elsewhere it is your own fault.
53
If null (default), a time/random revision id is generated.
96
## XXX: Remove this in favor of Branch.commit?
97
Commit().commit(*args, **kwargs)
100
class NullCommitReporter(object):
101
"""I report on progress of a commit."""
103
def snapshot_change(self, change, path):
106
def completed(self, revno, rev_id):
109
def deleted(self, file_id):
112
def escaped(self, escape_count, message):
115
def missing(self, path):
118
class ReportCommitToLog(NullCommitReporter):
120
def snapshot_change(self, change, path):
121
note("%s %s", change, path)
123
def completed(self, revno, rev_id):
124
note('committed r%d {%s}', revno, rev_id)
56
import os, time, tempfile
58
from inventory import Inventory
59
from osutils import isdir, isfile, sha_string, quotefn, \
60
local_time_offset, username, kind_marker, is_inside_any
126
def deleted(self, file_id):
127
note('deleted %s', file_id)
129
def escaped(self, escape_count, message):
130
note("replaced %d control characters in message", escape_count)
132
def missing(self, path):
133
note('missing %s', path)
135
class Commit(object):
136
"""Task of committing a new revision.
138
This is a MethodObject: it accumulates state as the commit is
139
prepared, and then it is discarded. It doesn't represent
140
historical revisions, just the act of recording a new one.
143
Modified to hold a list of files that have been deleted from
144
the working directory; these should be removed from the
149
if reporter is not None:
150
self.reporter = reporter
152
self.reporter = NullCommitReporter()
162
allow_pointless=True,
164
"""Commit working copy as a new revision.
166
timestamp -- if not None, seconds-since-epoch for a
167
postdated/predated commit.
169
specific_files -- If true, commit only those files.
171
rev_id -- If set, use this as the new revision id.
172
Useful for test or import commands that need to tightly
173
control what revisions are assigned. If you duplicate
174
a revision id that exists elsewhere it is your own fault.
175
If null (default), a time/random revision id is generated.
177
allow_pointless -- If true (default), commit even if nothing
178
has changed and no merges are recorded.
180
mutter('preparing to commit')
183
self.weave_store = branch.weave_store
185
self.specific_files = specific_files
186
self.allow_pointless = allow_pointless
188
if timestamp is None:
189
self.timestamp = time.time()
191
self.timestamp = long(timestamp)
194
self.rev_id = _gen_revision_id(self.branch, self.timestamp)
198
if committer is None:
199
self.committer = bzrlib.config.username(self.branch)
201
assert isinstance(committer, basestring), type(committer)
202
self.committer = committer
205
self.timezone = local_time_offset()
207
self.timezone = int(timezone)
209
assert isinstance(message, basestring), type(message)
210
self.message = message
211
self._escape_commit_message()
213
self.branch.lock_write()
215
self.work_tree = self.branch.working_tree()
216
self.work_inv = self.work_tree.inventory
217
self.basis_tree = self.branch.basis_tree()
218
self.basis_inv = self.basis_tree.inventory
220
self._gather_parents()
221
if len(self.parents) > 1 and self.specific_files:
222
raise NotImplementedError('selected-file commit of merges is not supported yet')
223
self._check_parents_present()
225
self._remove_deleted()
226
self._populate_new_inv()
227
self._store_snapshot()
228
self._report_deletes()
230
if not (self.allow_pointless
231
or len(self.parents) > 1
232
or self.new_inv != self.basis_inv):
233
raise PointlessCommit()
235
if len(list(self.work_tree.iter_conflicts()))>0:
236
raise ConflictsInTree
238
self._record_inventory()
239
self._make_revision()
240
self.reporter.completed(self.branch.revno()+1, self.rev_id)
241
self.branch.append_revision(self.rev_id)
242
self.branch.set_pending_merges([])
246
def _record_inventory(self):
247
"""Store the inventory for the new revision."""
248
inv_text = serializer_v5.write_inventory_to_string(self.new_inv)
249
self.inv_sha1 = sha_string(inv_text)
250
s = self.branch.control_weaves
251
s.add_text('inventory', self.rev_id,
252
split_lines(inv_text), self.present_parents,
253
self.branch.get_transaction())
255
def _escape_commit_message(self):
256
"""Replace xml-incompatible control characters."""
257
# Python strings can include characters that can't be
258
# represented in well-formed XML; escape characters that
259
# aren't listed in the XML specification
260
# (http://www.w3.org/TR/REC-xml/#NT-Char).
261
if isinstance(self.message, unicode):
262
char_pattern = u'[^\x09\x0A\x0D\u0020-\uD7FF\uE000-\uFFFD]'
264
# Use a regular 'str' as pattern to avoid having re.subn
265
# return 'unicode' results.
266
char_pattern = '[^x09\x0A\x0D\x20-\xFF]'
267
self.message, escape_count = re.subn(
269
lambda match: match.group(0).encode('unicode_escape'),
272
self.reporter.escaped(escape_count, self.message)
274
def _gather_parents(self):
275
"""Record the parents of a merge for merge detection."""
276
pending_merges = self.branch.pending_merges()
278
self.parent_invs = []
279
self.present_parents = []
280
precursor_id = self.branch.last_revision()
282
self.parents.append(precursor_id)
283
self.parents += pending_merges
284
for revision in self.parents:
285
if self.branch.has_revision(revision):
286
self.parent_invs.append(self.branch.get_inventory(revision))
287
self.present_parents.append(revision)
289
def _check_parents_present(self):
290
for parent_id in self.parents:
291
mutter('commit parent revision {%s}', parent_id)
292
if not self.branch.has_revision(parent_id):
293
if parent_id == self.branch.last_revision():
294
warning("parent is missing %r", parent_id)
295
raise HistoryMissing(self.branch, 'revision', parent_id)
297
mutter("commit will ghost revision %r", parent_id)
299
def _make_revision(self):
300
"""Record a new revision object for this commit."""
301
self.rev = Revision(timestamp=self.timestamp,
302
timezone=self.timezone,
303
committer=self.committer,
304
message=self.message,
305
inventory_sha1=self.inv_sha1,
306
revision_id=self.rev_id)
307
self.rev.parent_ids = self.parents
309
serializer_v5.write_revision(self.rev, rev_tmp)
311
self.branch.revision_store.add(rev_tmp, self.rev_id)
312
mutter('new revision_id is {%s}', self.rev_id)
314
def _remove_deleted(self):
315
"""Remove deleted files from the working inventories.
317
This is done prior to taking the working inventory as the
318
basis for the new committed inventory.
320
This returns true if any files
321
*that existed in the basis inventory* were deleted.
322
Files that were added and deleted
323
in the working copy don't matter.
325
specific = self.specific_files
327
for path, ie in self.work_inv.iter_entries():
328
if specific and not is_inside_any(specific, path):
330
if not self.work_tree.has_filename(path):
331
self.reporter.missing(path)
332
deleted_ids.append((path, ie.file_id))
334
deleted_ids.sort(reverse=True)
335
for path, file_id in deleted_ids:
336
del self.work_inv[file_id]
337
self.branch._write_inventory(self.work_inv)
339
def _store_snapshot(self):
340
"""Pass over inventory and record a snapshot.
342
Entries get a new revision when they are modified in
343
any way, which includes a merge with a new set of
344
parents that have the same entry.
346
# XXX: Need to think more here about when the user has
347
# made a specific decision on a particular value -- c.f.
349
for path, ie in self.new_inv.iter_entries():
350
previous_entries = ie.find_previous_heads(
352
self.weave_store.get_weave_or_empty(ie.file_id,
353
self.branch.get_transaction()))
354
if ie.revision is None:
355
change = ie.snapshot(self.rev_id, path, previous_entries,
356
self.work_tree, self.weave_store,
357
self.branch.get_transaction())
360
self.reporter.snapshot_change(change, path)
362
def _populate_new_inv(self):
363
"""Build revision inventory.
365
This creates a new empty inventory. Depending on
366
which files are selected for commit, and what is present in the
367
current tree, the new inventory is populated. inventory entries
368
which are candidates for modification have their revision set to
369
None; inventory entries that are carried over untouched have their
370
revision set to their prior value.
372
mutter("Selecting files for commit with filter %s", self.specific_files)
373
self.new_inv = Inventory()
374
for path, new_ie in self.work_inv.iter_entries():
375
file_id = new_ie.file_id
376
mutter('check %s {%s}', path, new_ie.file_id)
377
if self.specific_files:
378
if not is_inside_any(self.specific_files, path):
379
mutter('%s not selected for commit', path)
380
self._carry_entry(file_id)
383
# this is selected, ensure its parents are too.
384
parent_id = new_ie.parent_id
385
while parent_id != ROOT_ID:
386
if not self.new_inv.has_id(parent_id):
387
ie = self._select_entry(self.work_inv[parent_id])
388
mutter('%s selected for commit because of %s',
389
self.new_inv.id2path(parent_id), path)
391
ie = self.new_inv[parent_id]
392
if ie.revision is not None:
394
mutter('%s selected for commit because of %s',
395
self.new_inv.id2path(parent_id), path)
396
parent_id = ie.parent_id
397
mutter('%s selected for commit', path)
398
self._select_entry(new_ie)
400
def _select_entry(self, new_ie):
401
"""Make new_ie be considered for committing."""
407
def _carry_entry(self, file_id):
408
"""Carry the file unchanged from the basis revision."""
409
if self.basis_inv.has_id(file_id):
410
self.new_inv.add(self.basis_inv[file_id].copy())
412
def _report_deletes(self):
413
for file_id in self.basis_inv:
414
if file_id not in self.new_inv:
415
self.reporter.deleted(self.basis_inv.id2path(file_id))
417
def _gen_revision_id(branch, when):
62
from branch import gen_file_id
63
from errors import BzrError
64
from revision import Revision
65
from trace import mutter, note
67
branch._need_writelock()
69
# First walk over the working inventory; and both update that
70
# and also build a new revision inventory. The revision
71
# inventory needs to hold the text-id, sha1 and size of the
72
# actual file versions committed in the revision. (These are
73
# not present in the working inventory.) We also need to
74
# detect missing/deleted files, and remove them from the
77
work_tree = branch.working_tree()
78
work_inv = work_tree.inventory
80
basis = branch.basis_tree()
81
basis_inv = basis.inventory
85
note('looking for changes...')
87
for path, entry in work_inv.iter_entries():
88
## TODO: Check that the file kind has not changed from the previous
89
## revision of this file (if any).
93
p = branch.abspath(path)
94
file_id = entry.file_id
95
mutter('commit prep file %s, id %r ' % (p, file_id))
97
if specific_files and not is_inside_any(specific_files, path):
98
if basis_inv.has_id(file_id):
99
# carry over with previous state
100
inv.add(basis_inv[file_id].copy())
102
# omit this from committed inventory
106
if not work_tree.has_id(file_id):
108
print('deleted %s%s' % (path, kind_marker(entry.kind)))
109
mutter(" file is missing, removing from inventory")
110
missing_ids.append(file_id)
115
if basis_inv.has_id(file_id):
116
old_kind = basis_inv[file_id].kind
117
if old_kind != entry.kind:
118
raise BzrError("entry %r changed kind from %r to %r"
119
% (file_id, old_kind, entry.kind))
121
if entry.kind == 'directory':
123
raise BzrError("%s is entered as directory but not a directory"
125
elif entry.kind == 'file':
127
raise BzrError("%s is entered as file but is not a file" % quotefn(p))
129
new_sha1 = work_tree.get_file_sha1(file_id)
131
old_ie = basis_inv.has_id(file_id) and basis_inv[file_id]
133
and old_ie.text_sha1 == new_sha1):
134
## assert content == basis.get_file(file_id).read()
135
entry.text_id = old_ie.text_id
136
entry.text_sha1 = new_sha1
137
entry.text_size = old_ie.text_size
138
mutter(' unchanged from previous text_id {%s}' %
141
content = file(p, 'rb').read()
143
# calculate the sha again, just in case the file contents
144
# changed since we updated the cache
145
entry.text_sha1 = sha_string(content)
146
entry.text_size = len(content)
148
entry.text_id = gen_file_id(entry.name)
149
branch.text_store.add(content, entry.text_id)
150
mutter(' stored with text_id {%s}' % entry.text_id)
153
print('added %s' % path)
154
elif (old_ie.name == entry.name
155
and old_ie.parent_id == entry.parent_id):
156
print('modified %s' % path)
158
print('renamed %s' % path)
161
for file_id in missing_ids:
162
# Any files that have been deleted are now removed from the
163
# working inventory. Files that were not selected for commit
164
# are left as they were in the working inventory and ommitted
165
# from the revision inventory.
167
# have to do this later so we don't mess up the iterator.
168
# since parents may be removed before their children we
171
# FIXME: There's probably a better way to do this; perhaps
172
# the workingtree should know how to filter itbranch.
173
if work_inv.has_id(file_id):
174
del work_inv[file_id]
178
rev_id = _gen_revision_id(time.time())
181
inv_tmp = tempfile.TemporaryFile()
182
inv.write_xml(inv_tmp)
184
branch.inventory_store.add(inv_tmp, inv_id)
185
mutter('new inventory_id is {%s}' % inv_id)
187
branch._write_inventory(work_inv)
189
if timestamp == None:
190
timestamp = time.time()
192
if committer == None:
193
committer = username()
196
timezone = local_time_offset()
198
mutter("building commit log message")
199
rev = Revision(timestamp=timestamp,
202
precursor = branch.last_patch(),
207
rev_tmp = tempfile.TemporaryFile()
208
rev.write_xml(rev_tmp)
210
branch.revision_store.add(rev_tmp, rev_id)
211
mutter("new revision_id is {%s}" % rev_id)
213
## XXX: Everything up to here can simply be orphaned if we abort
214
## the commit; it will leave junk files behind but that doesn't
217
## TODO: Read back the just-generated changeset, and make sure it
218
## applies and recreates the right state.
220
## TODO: Also calculate and store the inventory SHA1
221
mutter("committing patch r%d" % (branch.revno() + 1))
223
branch.append_revision(rev_id)
226
note("commited r%d" % branch.revno())
230
def _gen_revision_id(when):
418
231
"""Return new revision-id."""
419
s = '%s-%s-' % (bzrlib.config.user_email(branch), compact_date(when))
232
from binascii import hexlify
233
from osutils import rand_bytes, compact_date, user_email
235
s = '%s-%s-' % (user_email(), compact_date(when))
420
236
s += hexlify(rand_bytes(8))
added
removed
bzrlib/commit.py
