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
35
from bzrlib.errors import BzrCheckError, NoSuchRevision
36
from bzrlib.symbol_versioning import *
37
from bzrlib.trace import mutter, note, warning
42
"""Check a repository"""
44
# The Check object interacts with InventoryEntry.check, etc.
46
def __init__(self, repository):
47
self.repository = repository
48
self.checked_text_cnt = 0
49
self.checked_rev_cnt = 0
51
self.repeated_text_cnt = 0
52
self.missing_parent_links = {}
53
self.missing_inventory_sha_cnt = 0
54
self.missing_revision_cnt = 0
55
# maps (file-id, version) -> sha1; used by InventoryFile._check
56
self.checked_texts = {}
57
self.checked_weaves = {}
60
self.repository.lock_read()
61
self.progress = bzrlib.ui.ui_factory.nested_progress_bar()
63
self.progress.update('retrieving inventory', 0, 0)
64
# do not put in init, as it should be done with progess,
65
# and inside the lock.
66
self.inventory_weave = self.repository.get_inventory_weave()
70
while revno < len(self.planned_revisions):
71
rev_id = self.planned_revisions[revno]
72
self.progress.update('checking revision', revno,
73
len(self.planned_revisions))
75
self.check_one_rev(rev_id)
77
self.progress.finished()
78
self.repository.unlock()
80
def plan_revisions(self):
81
repository = self.repository
82
self.planned_revisions = set(repository.all_revision_ids())
84
inventoried = set(self.inventory_weave.versions())
85
awol = self.planned_revisions - inventoried
87
raise BzrCheckError('Stored revisions missing from inventory'
88
'{%s}' % ','.join([f for f in awol]))
89
self.planned_revisions = list(self.planned_revisions)
91
def report_results(self, verbose):
92
note('checked repository %s format %s',
93
self.repository.bzrdir.root_transport,
94
self.repository._format)
95
note('%6d revisions', self.checked_rev_cnt)
96
note('%6d unique file texts', self.checked_text_cnt)
97
note('%6d repeated file texts', self.repeated_text_cnt)
98
note('%6d weaves', len(self.checked_weaves))
99
if self.missing_inventory_sha_cnt:
100
note('%6d revisions are missing inventory_sha1',
101
self.missing_inventory_sha_cnt)
102
if self.missing_revision_cnt:
103
note('%6d revisions are mentioned but not present',
104
self.missing_revision_cnt)
106
note('%6d ghost revisions', len(self.ghosts))
108
for ghost in self.ghosts:
110
if len(self.missing_parent_links):
111
note('%6d revisions missing parents in ancestry',
112
len(self.missing_parent_links))
114
for link, linkers in self.missing_parent_links.items():
115
note(' %s should be in the ancestry for:', link)
116
for linker in linkers:
117
note(' * %s', linker)
119
def check_one_rev(self, rev_id):
120
"""Check one revision.
122
rev_id - the one to check
124
rev = self.repository.get_revision(rev_id)
126
if rev.revision_id != rev_id:
127
raise BzrCheckError('wrong internal revision id in revision {%s}'
130
for parent in rev.parent_ids:
131
if not parent in self.planned_revisions:
132
missing_links = self.missing_parent_links.get(parent, [])
133
missing_links.append(rev_id)
134
self.missing_parent_links[parent] = missing_links
135
# list based so somewhat slow,
136
# TODO have a planned_revisions list and set.
137
if self.repository.has_revision(parent):
138
missing_ancestry = self.repository.get_ancestry(parent)
139
for missing in missing_ancestry:
140
if (missing is not None
141
and missing not in self.planned_revisions):
142
self.planned_revisions.append(missing)
144
self.ghosts.append(rev_id)
146
if rev.inventory_sha1:
147
inv_sha1 = self.repository.get_inventory_sha1(rev_id)
148
if inv_sha1 != rev.inventory_sha1:
149
raise BzrCheckError('Inventory sha1 hash doesn\'t match'
150
' value in revision {%s}' % rev_id)
152
self.missing_inventory_sha_cnt += 1
153
mutter("no inventory_sha1 on revision {%s}", rev_id)
154
self._check_revision_tree(rev_id)
155
self.checked_rev_cnt += 1
157
def check_weaves(self):
158
"""Check all the weaves we can get our hands on.
162
if self.repository.weave_store.listable():
163
weave_ids = list(self.repository.weave_store)
164
n_weaves = len(weave_ids)
165
self.progress.update('checking weave', 0, n_weaves)
166
self.inventory_weave.check(progress_bar=self.progress)
167
for i, weave_id in enumerate(weave_ids):
168
self.progress.update('checking weave', i, n_weaves)
169
w = self.repository.weave_store.get_weave(weave_id,
170
self.repository.get_transaction())
171
# No progress here, because it looks ugly.
173
self.checked_weaves[weave_id] = True
175
def _check_revision_tree(self, rev_id):
176
tree = self.repository.revision_tree(rev_id)
180
if file_id in seen_ids:
181
raise BzrCheckError('duplicated file_id {%s} '
182
'in inventory for revision {%s}'
184
seen_ids[file_id] = True
187
ie.check(self, rev_id, inv, tree)
189
for path, ie in inv.iter_entries():
190
if path in seen_names:
191
raise BzrCheckError('duplicated path %s '
192
'in inventory for revision {%s}'
194
seen_names[path] = True
197
def check(branch, verbose):
198
"""Run consistency checks on a branch.
200
Results are reported through logging.
202
:raise BzrCheckError: if there's a consistency error.
206
branch_result = branch.check()
207
repo_result = branch.repository.check([branch.last_revision()])
210
branch_result.report_results(verbose)
211
repo_result.report_results(verbose)
added
removed
bzrlib/check.py
