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
30
from bzrlib.trace import note, warning
31
from bzrlib.osutils import rename, sha_string, fingerprint_file, sha_strings
32
from bzrlib.trace import mutter
33
from bzrlib.errors import BzrCheckError, NoSuchRevision
34
from bzrlib.inventory import ROOT_ID
35
from bzrlib.branch import gen_root_id
40
def __init__(self, branch):
52
branch.weave_store.enable_cache = True
53
branch.control_weaves.enable_cache = True
55
self.checked_text_cnt = 0
56
self.checked_rev_cnt = 0
57
self.repeated_text_cnt = 0
58
self.missing_inventory_sha_cnt = 0
59
self.missing_revision_cnt = 0
60
# maps (file-id, version) -> sha1
61
self.checked_texts = {}
63
history = branch.revision_history()
65
revcount = len(history)
68
self.progress = bzrlib.ui.ui_factory.progress_bar()
69
for rev_id in history:
70
self.progress.update('checking revision', revno, revcount)
72
self.check_one_rev(rev_id, last_rev_id)
78
def report_results(self):
79
note('checked branch %s format %d',
81
self.branch._branch_format)
83
note('%6d revisions', self.checked_rev_cnt)
84
note('%6d unique file texts', self.checked_text_cnt)
85
note('%6d repeated file texts', self.repeated_text_cnt)
86
if self.missing_inventory_sha_cnt:
87
note('%d revisions are missing inventory_sha1',
88
self.missing_inventory_sha_cnt)
89
if self.missing_revision_cnt:
90
note('%d revisions are mentioned but not present',
91
self.missing_revision_cnt)
94
def check_one_rev(self, rev_id, last_rev_id):
95
"""Check one revision.
97
rev_id - the one to check
99
last_rev_id - the previous one on the mainline, if any.
102
# mutter(' revision {%s}' % rev_id)
104
rev = branch.get_revision(rev_id)
105
if rev.revision_id != rev_id:
106
raise BzrCheckError('wrong internal revision id in revision {%s}'
109
# check the previous history entry is a parent of this entry
111
if last_rev_id is None:
112
raise BzrCheckError("revision {%s} has %d parents, but is the "
113
"start of the branch"
114
% (rev_id, len(rev.parent_ids)))
115
for parent_id in rev.parent_ids:
116
if parent_id == last_rev_id:
119
raise BzrCheckError("previous revision {%s} not listed among "
121
% (last_rev_id, rev_id))
123
raise BzrCheckError("revision {%s} has no parents listed "
124
"but preceded by {%s}"
125
% (rev_id, last_rev_id))
127
if rev.inventory_sha1:
128
inv_sha1 = branch.get_inventory_sha1(rev_id)
129
if inv_sha1 != rev.inventory_sha1:
130
raise BzrCheckError('Inventory sha1 hash doesn\'t match'
131
' value in revision {%s}' % rev_id)
133
missing_inventory_sha_cnt += 1
134
mutter("no inventory_sha1 on revision {%s}" % rev_id)
135
self._check_revision_tree(rev_id)
136
self.checked_rev_cnt += 1
138
def _check_revision_tree(self, rev_id):
139
tree = self.branch.revision_tree(rev_id)
143
if file_id in seen_ids:
144
raise BzrCheckError('duplicated file_id {%s} '
145
'in inventory for revision {%s}'
147
seen_ids[file_id] = True
149
self._check_one_entry(rev_id, inv, tree, file_id)
151
for path, ie in inv.iter_entries():
152
if path in seen_names:
153
raise BzrCheckError('duplicated path %s '
154
'in inventory for revision {%s}'
156
seen_names[path] = True
159
def _check_one_entry(self, rev_id, inv, tree, file_id):
161
if ie.parent_id != None:
162
if not inv.has_id(ie.parent_id):
163
raise BzrCheckError('missing parent {%s} in inventory for revision {%s}'
164
% (ie.parent_id, rev_id))
165
if ie.kind == 'file':
166
text_version = ie.text_version
167
t = (file_id, text_version)
168
if t in self.checked_texts:
169
prev_sha = self.checked_texts[t]
170
if prev_sha != ie.text_sha1:
171
raise BzrCheckError('mismatched sha1 on {%s} in {%s}' %
174
self.repeated_text_cnt += 1
176
mutter('check version {%s} of {%s}', rev_id, file_id)
177
file_lines = tree.get_file_lines(file_id)
178
self.checked_text_cnt += 1
179
if ie.text_size != sum(map(len, file_lines)):
180
raise BzrCheckError('text {%s} wrong size' % ie.text_id)
181
if ie.text_sha1 != sha_strings(file_lines):
182
raise BzrCheckError('text {%s} wrong sha1' % ie.text_id)
183
self.checked_texts[t] = ie.text_sha1
184
elif ie.kind == 'directory':
185
if ie.text_sha1 != None or ie.text_size != None or ie.text_id != None:
186
raise BzrCheckError('directory {%s} has text in revision {%s}'
188
elif ie.kind == 'root_directory':
191
raise BzrCheckError('unknown entry kind %r in revision {%s}' %
196
"""Run consistency checks on a branch."""
added
removed
bzrlib/check.py
