1
# Copyright (C) 2005, 2006 by 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
17
# TODO: Check ancestries are correct for every revision: includes
18
# every committed so far, and in a reasonable order.
20
# TODO: Also check non-mainline revisions mentioned as parents.
22
# TODO: Check for extra files in the control directory.
24
# TODO: Check revision, inventory and entry objects have all
27
# TODO: Get every revision in the revision-store even if they're not
28
# referenced by history and make sure they're all valid.
30
# TODO: Perhaps have a way to record errors other than by raising exceptions;
31
# would perhaps be enough to accumulate exception objects in a list without
32
# raising them. If there's more than one exception it'd be good to see them
36
from bzrlib.trace import note, warning
37
from bzrlib.osutils import rename, sha_string, fingerprint_file
38
from bzrlib.trace import mutter
39
from bzrlib.errors import BzrCheckError, NoSuchRevision
45
# The Check object interacts with InventoryEntry.check, etc.
47
def __init__(self, branch):
49
self.repository = branch.repository
50
self.checked_text_cnt = 0
51
self.checked_rev_cnt = 0
53
self.repeated_text_cnt = 0
54
self.missing_parent_links = {}
55
self.missing_inventory_sha_cnt = 0
56
self.missing_revision_cnt = 0
57
# maps (file-id, version) -> sha1; used by InventoryFile._check
58
self.checked_texts = {}
59
self.checked_weaves = {}
62
self.branch.lock_read()
63
self.progress = bzrlib.ui.ui_factory.nested_progress_bar()
65
self.progress.update('retrieving inventory', 0, 0)
66
# do not put in init, as it should be done with progess,
67
# and inside the lock.
68
self.inventory_weave = self.branch.repository.get_inventory_weave()
69
self.history = self.branch.revision_history()
70
if not len(self.history):
76
while revno < len(self.planned_revisions):
77
rev_id = self.planned_revisions[revno]
78
self.progress.update('checking revision', revno,
79
len(self.planned_revisions))
81
self.check_one_rev(rev_id)
83
self.progress.finished()
86
def plan_revisions(self):
87
repository = self.branch.repository
88
self.planned_revisions = set(repository.all_revision_ids())
90
inventoried = set(self.inventory_weave.versions())
91
awol = self.planned_revisions - inventoried
93
raise BzrCheckError('Stored revisions missing from inventory'
94
'{%s}' % ','.join([f for f in awol]))
95
self.planned_revisions = list(self.planned_revisions)
97
def report_results(self, verbose):
98
note('checked branch %s format %s',
102
note('%6d revisions', self.checked_rev_cnt)
103
note('%6d unique file texts', self.checked_text_cnt)
104
note('%6d repeated file texts', self.repeated_text_cnt)
105
note('%6d weaves', len(self.checked_weaves))
106
if self.missing_inventory_sha_cnt:
107
note('%6d revisions are missing inventory_sha1',
108
self.missing_inventory_sha_cnt)
109
if self.missing_revision_cnt:
110
note('%6d revisions are mentioned but not present',
111
self.missing_revision_cnt)
113
note('%6d ghost revisions', len(self.ghosts))
115
for ghost in self.ghosts:
117
if len(self.missing_parent_links):
118
note('%6d revisions missing parents in ancestry',
119
len(self.missing_parent_links))
121
for link, linkers in self.missing_parent_links.items():
122
note(' %s should be in the ancestry for:', link)
123
for linker in linkers:
124
note(' * %s', linker)
126
def check_one_rev(self, rev_id):
127
"""Check one revision.
129
rev_id - the one to check
131
last_rev_id - the previous one on the mainline, if any.
134
# mutter(' revision {%s}', rev_id)
137
rev_history_position = self.history.index(rev_id)
139
rev_history_position = None
141
if rev_history_position:
142
rev = branch.repository.get_revision(rev_id)
143
if rev_history_position > 0:
144
last_rev_id = self.history[rev_history_position - 1]
146
rev = branch.repository.get_revision(rev_id)
148
if rev.revision_id != rev_id:
149
raise BzrCheckError('wrong internal revision id in revision {%s}'
152
# check the previous history entry is a parent of this entry
154
if last_rev_id is not None:
155
for parent_id in rev.parent_ids:
156
if parent_id == last_rev_id:
159
raise BzrCheckError("previous revision {%s} not listed among "
161
% (last_rev_id, rev_id))
162
for parent in rev.parent_ids:
163
if not parent in self.planned_revisions:
164
missing_links = self.missing_parent_links.get(parent, [])
165
missing_links.append(rev_id)
166
self.missing_parent_links[parent] = missing_links
167
# list based so somewhat slow,
168
# TODO have a planned_revisions list and set.
169
if self.branch.repository.has_revision(parent):
170
missing_ancestry = self.repository.get_ancestry(parent)
171
for missing in missing_ancestry:
172
if (missing is not None
173
and missing not in self.planned_revisions):
174
self.planned_revisions.append(missing)
176
self.ghosts.append(rev_id)
178
raise BzrCheckError("revision {%s} has no parents listed "
179
"but preceded by {%s}"
180
% (rev_id, last_rev_id))
182
if rev.inventory_sha1:
183
inv_sha1 = branch.repository.get_inventory_sha1(rev_id)
184
if inv_sha1 != rev.inventory_sha1:
185
raise BzrCheckError('Inventory sha1 hash doesn\'t match'
186
' value in revision {%s}' % rev_id)
188
missing_inventory_sha_cnt += 1
189
mutter("no inventory_sha1 on revision {%s}", rev_id)
190
self._check_revision_tree(rev_id)
191
self.checked_rev_cnt += 1
193
def check_weaves(self):
194
"""Check all the weaves we can get our hands on.
198
if self.branch.repository.weave_store.listable():
199
weave_ids = list(self.branch.repository.weave_store)
200
n_weaves = len(weave_ids)
201
self.progress.update('checking weave', 0, n_weaves)
202
self.inventory_weave.check(progress_bar=self.progress)
203
for i, weave_id in enumerate(weave_ids):
204
self.progress.update('checking weave', i, n_weaves)
205
w = self.branch.repository.weave_store.get_weave(weave_id,
206
self.branch.repository.get_transaction())
207
# No progress here, because it looks ugly.
209
self.checked_weaves[weave_id] = True
211
def _check_revision_tree(self, rev_id):
212
tree = self.branch.repository.revision_tree(rev_id)
216
if file_id in seen_ids:
217
raise BzrCheckError('duplicated file_id {%s} '
218
'in inventory for revision {%s}'
220
seen_ids[file_id] = True
223
ie.check(self, rev_id, inv, tree)
225
for path, ie in inv.iter_entries():
226
if path in seen_names:
227
raise BzrCheckError('duplicated path %s '
228
'in inventory for revision {%s}'
230
seen_names[path] = True
233
def check(branch, verbose):
234
"""Run consistency checks on a branch."""
235
checker = Check(branch)
237
checker.report_results(verbose)
added
removed
bzrlib/check.py
