1
# Copyright (C) 2004, 2005 by Martin Pool
2
# Copyright (C) 2005 by Canonical Ltd
4
# This program is free software; you can redistribute it and/or modify
5
# it under the terms of the GNU General Public License as published by
6
# the Free Software Foundation; either version 2 of the License, or
7
# (at your option) any later version.
9
# This program is distributed in the hope that it will be useful,
10
# but WITHOUT ANY WARRANTY; without even the implied warranty of
11
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12
# GNU General Public License for more details.
14
# You should have received a copy of the GNU General Public License
15
# along with this program; if not, write to the Free Software
16
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
18
# TODO: Check ancestries are correct for every revision: includes
19
# every committed so far, and in a reasonable order.
21
# TODO: Also check non-mainline revisions mentioned as parents.
23
# TODO: Check for extra files in the control directory.
25
# TODO: Check revision, inventory and entry objects have all
28
# TODO: Get every revision in the revision-store even if they're not
29
# referenced by history and make sure they're all valid.
32
from bzrlib.trace import note, warning
33
from bzrlib.osutils import rename, sha_string, fingerprint_file
34
from bzrlib.trace import mutter
35
from bzrlib.errors import BzrCheckError, NoSuchRevision
36
from bzrlib.inventory import ROOT_ID
37
from bzrlib.branch import gen_root_id
43
def __init__(self, branch):
45
self.inventory_weave = branch._get_inventory_weave()
46
self.checked_text_cnt = 0
47
self.checked_rev_cnt = 0
49
self.repeated_text_cnt = 0
50
self.missing_parent_links = {}
51
self.missing_inventory_sha_cnt = 0
52
self.missing_revision_cnt = 0
53
# maps (file-id, version) -> sha1
54
self.checked_texts = {}
57
self.branch.lock_read()
58
self.progress = bzrlib.ui.ui_factory.progress_bar()
60
self.history = self.branch.revision_history()
61
if not len(self.history):
64
if not self.branch.revision_store.listable():
65
raise BzrCheckError("Branch must be local")
66
self.planned_revisions = set(self.branch.revision_store)
67
inventoried = set(self.inventory_weave.names())
68
awol = self.planned_revisions - inventoried
70
raise BzrCheckError('Stored revisions missing from inventory'
71
'{%s}' % ','.join([f for f in awol]))
73
for revno, rev_id in enumerate(self.planned_revisions):
74
self.progress.update('checking revision', revno+1,
75
len(self.planned_revisions))
76
self.check_one_rev(rev_id)
81
def report_results(self, verbose):
82
note('checked branch %s format %d',
84
self.branch._branch_format)
86
note('%6d revisions', self.checked_rev_cnt)
87
note('%6d unique file texts', self.checked_text_cnt)
88
note('%6d repeated file texts', self.repeated_text_cnt)
89
if self.missing_inventory_sha_cnt:
90
note('%6d revisions are missing inventory_sha1',
91
self.missing_inventory_sha_cnt)
92
if self.missing_revision_cnt:
93
note('%6d revisions are mentioned but not present',
94
self.missing_revision_cnt)
96
note('%6d ghost revisions', len(self.ghosts))
98
for ghost in self.ghosts:
100
if len(self.missing_parent_links):
101
note('%6d revisions missing parents in ancestry',
102
len(self.missing_parent_links))
104
for link, linkers in self.missing_parent_links.items():
105
note(' %s should be in the ancestry for:', link)
106
for linker in linkers:
107
note(' * %s', linker)
109
def check_one_rev(self, rev_id):
110
"""Check one revision.
112
rev_id - the one to check
114
last_rev_id - the previous one on the mainline, if any.
117
# mutter(' revision {%s}', rev_id)
120
rev_history_position = self.history.index(rev_id)
122
rev_history_position = None
124
if rev_history_position:
125
rev = branch.get_revision(rev_id)
126
if rev_history_position > 0:
127
last_rev_id = self.history[rev_history_position - 1]
129
rev = branch.get_revision(rev_id)
131
if rev.revision_id != rev_id:
132
raise BzrCheckError('wrong internal revision id in revision {%s}'
135
# check the previous history entry is a parent of this entry
137
if last_rev_id is not None:
138
for parent_id in rev.parent_ids:
139
if parent_id == last_rev_id:
142
raise BzrCheckError("previous revision {%s} not listed among "
144
% (last_rev_id, rev_id))
145
for parent in rev.parent_ids:
146
if not parent in self.planned_revisions:
147
missing_links = self.missing_parent_links.get(parent, [])
148
missing_links.append(rev_id)
149
self.missing_parent_links[parent] = missing_links
150
# list based so slow, TODO have a planned_revisions list and set.
151
if self.branch.has_revision(parent):
152
missing_ancestry = self.branch.get_ancestry(parent)
153
for missing in missing_ancestry:
154
if (missing is not None
155
and missing not in self.planned_revisions):
156
self.planned_revisions.append(missing)
158
self.ghosts.append(rev_id)
160
raise BzrCheckError("revision {%s} has no parents listed "
161
"but preceded by {%s}"
162
% (rev_id, last_rev_id))
164
if rev.inventory_sha1:
165
inv_sha1 = branch.get_inventory_sha1(rev_id)
166
if inv_sha1 != rev.inventory_sha1:
167
raise BzrCheckError('Inventory sha1 hash doesn\'t match'
168
' value in revision {%s}' % rev_id)
170
missing_inventory_sha_cnt += 1
171
mutter("no inventory_sha1 on revision {%s}", rev_id)
172
self._check_revision_tree(rev_id)
173
self.checked_rev_cnt += 1
175
def _check_revision_tree(self, rev_id):
176
tree = self.branch.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."""
199
checker = Check(branch)
201
checker.report_results(verbose)
added
removed
bzrlib/check.py
