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
42
def __init__(self, branch):
44
self.inventory_weave = branch._get_inventory_weave()
45
self.checked_text_cnt = 0
46
self.checked_rev_cnt = 0
48
self.repeated_text_cnt = 0
49
self.missing_parent_links = {}
50
self.missing_inventory_sha_cnt = 0
51
self.missing_revision_cnt = 0
52
# maps (file-id, version) -> sha1
53
self.checked_texts = {}
56
self.branch.lock_read()
57
self.progress = bzrlib.ui.ui_factory.progress_bar()
59
self.history = self.branch.revision_history()
60
if not len(self.history):
63
if not self.branch.revision_store.listable():
64
raise BzrCheckError("Branch must be local")
65
self.planned_revisions = set(self.branch.revision_store)
66
inventoried = set(self.inventory_weave.names())
67
awol = self.planned_revisions - inventoried
69
raise BzrCheckError('Stored revisions missing from inventory'
70
'{%s}' % ','.join([f for f in awol]))
72
for revno, rev_id in enumerate(self.planned_revisions):
73
self.progress.update('checking revision', revno+1,
74
len(self.planned_revisions))
75
self.check_one_rev(rev_id)
80
def report_results(self, verbose):
81
note('checked branch %s format %d',
83
self.branch._branch_format)
85
note('%6d revisions', self.checked_rev_cnt)
86
note('%6d unique file texts', self.checked_text_cnt)
87
note('%6d repeated file texts', self.repeated_text_cnt)
88
if self.missing_inventory_sha_cnt:
89
note('%6d revisions are missing inventory_sha1',
90
self.missing_inventory_sha_cnt)
91
if self.missing_revision_cnt:
92
note('%6d revisions are mentioned but not present',
93
self.missing_revision_cnt)
95
note('%6d ghost revisions', len(self.ghosts))
97
for ghost in self.ghosts:
99
if len(self.missing_parent_links):
100
note('%6d revisions missing parents in ancestry',
101
len(self.missing_parent_links))
103
for link, linkers in self.missing_parent_links.items():
104
note(' %s should be in the ancestry for:', link)
105
for linker in linkers:
106
note(' * %s', linker)
108
def check_one_rev(self, rev_id):
109
"""Check one revision.
111
rev_id - the one to check
113
last_rev_id - the previous one on the mainline, if any.
116
# mutter(' revision {%s}', rev_id)
119
rev_history_position = self.history.index(rev_id)
121
rev_history_position = None
123
if rev_history_position:
124
rev = branch.get_revision(rev_id)
125
if rev_history_position > 0:
126
last_rev_id = self.history[rev_history_position - 1]
128
rev = branch.get_revision(rev_id)
130
if rev.revision_id != rev_id:
131
raise BzrCheckError('wrong internal revision id in revision {%s}'
134
# check the previous history entry is a parent of this entry
136
if last_rev_id is not None:
137
for parent_id in rev.parent_ids:
138
if parent_id == last_rev_id:
141
raise BzrCheckError("previous revision {%s} not listed among "
143
% (last_rev_id, rev_id))
144
for parent in rev.parent_ids:
145
if not parent in self.planned_revisions:
146
missing_links = self.missing_parent_links.get(parent, [])
147
missing_links.append(rev_id)
148
self.missing_parent_links[parent] = missing_links
149
# list based so slow, TODO have a planned_revisions list and set.
150
if self.branch.has_revision(parent):
151
missing_ancestry = self.branch.get_ancestry(parent)
152
for missing in missing_ancestry:
153
if (missing is not None
154
and missing not in self.planned_revisions):
155
self.planned_revisions.append(missing)
157
self.ghosts.append(rev_id)
159
raise BzrCheckError("revision {%s} has no parents listed "
160
"but preceded by {%s}"
161
% (rev_id, last_rev_id))
163
if rev.inventory_sha1:
164
inv_sha1 = branch.get_inventory_sha1(rev_id)
165
if inv_sha1 != rev.inventory_sha1:
166
raise BzrCheckError('Inventory sha1 hash doesn\'t match'
167
' value in revision {%s}' % rev_id)
169
missing_inventory_sha_cnt += 1
170
mutter("no inventory_sha1 on revision {%s}", rev_id)
171
self._check_revision_tree(rev_id)
172
self.checked_rev_cnt += 1
174
def _check_revision_tree(self, rev_id):
175
tree = self.branch.revision_tree(rev_id)
179
if file_id in seen_ids:
180
raise BzrCheckError('duplicated file_id {%s} '
181
'in inventory for revision {%s}'
183
seen_ids[file_id] = True
186
ie.check(self, rev_id, inv, tree)
188
for path, ie in inv.iter_entries():
189
if path in seen_names:
190
raise BzrCheckError('duplicated path %s '
191
'in inventory for revision {%s}'
193
seen_names[path] = True
196
def check(branch, verbose):
197
"""Run consistency checks on a branch."""
198
checker = Check(branch)
200
checker.report_results(verbose)
added
removed
bzrlib/check.py
