« back to all changes in this revision
Viewing changes to bzrlib/reconcile.py
- browse files at revision 1587.1.1
- compare with another revision
- compare with revision 963
- download diff
- download tarball
- view history from revision 1587.1.1
- Committer: Robert Collins
- Date: 2006-03-03 05:38:04 UTC
- mfrom: (1570.1.14 otavio-fixes)
- mto: (1563.2.25 versioned-file-knit-stores)
- mto: This revision was merged to the branch mainline in revision 1588.
- Revision ID: robertc@robertcollins.net-20060303053804-ef5cbede74123fcc
* New 'reconcile' command will check branch consistency and repair indexes
that can become out of sync in pre 0.8 formats. (Robert Collins,
Aaron Bentley)
that can become out of sync in pre 0.8 formats. (Robert Collins,
Aaron Bentley)
- files added:
- bzrlib/reconcile.py
- files modified:
- NEWS
added
removed
bzrlib/reconcile.py
1
# (C) 2005, 2006 Canonical Limited.
2
#
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.
7
#
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.
12
#
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
16
17
"""Reconcilers are able to fix some potential data errors in a branch."""
18
19
20
__all__ = ['reconcile', 'Reconciler', 'RepoReconciler']
21
22
23
import bzrlib.branch
24
import bzrlib.errors as errors
25
import bzrlib.progress
26
from bzrlib.trace import mutter
27
from bzrlib.tsort import TopoSorter
28
import bzrlib.ui as ui
29
30
31
def reconcile(dir):
32
"""Reconcile the data in dir.
33
34
Currently this is limited to a inventory 'reweave'.
35
36
This is a convenience method, for using a Reconciler object.
37
38
Directly using Reconciler is recommended for library users that
39
desire fine grained control or analysis of the found issues.
40
"""
41
reconciler = Reconciler(dir)
42
reconciler.reconcile()
43
44
45
class Reconciler(object):
46
"""Reconcilers are used to reconcile existing data."""
47
48
def __init__(self, dir):
49
self.bzrdir = dir
50
51
def reconcile(self):
52
"""Perform reconciliation.
53
54
After reconciliation the following attributes document found issues:
55
inconsistent_parents: The number of revisions in the repository whose
56
ancestry was being reported incorrectly.
57
garbage_inventories: The number of inventory objects without revisions
58
that were garbage collected.
59
"""
60
self.pb = ui.ui_factory.progress_bar()
61
self.repo = self.bzrdir.find_repository()
62
self.pb.note('Reconciling repository %s',
63
self.repo.bzrdir.root_transport.base)
64
repo_reconciler = RepoReconciler(self.repo)
65
repo_reconciler.reconcile()
66
self.inconsistent_parents = repo_reconciler.inconsistent_parents
67
self.garbage_inventories = repo_reconciler.garbage_inventories
68
self.pb.note('Reconciliation complete.')
69
70
71
class RepoReconciler(object):
72
"""Reconciler that reconciles a repository.
73
74
Currently this consists of an inventory reweave with revision cross-checks.
75
"""
76
77
def __init__(self, repo):
78
self.repo = repo
79
80
def reconcile(self):
81
"""Perform reconciliation.
82
83
After reconciliation the following attributes document found issues:
84
inconsistent_parents: The number of revisions in the repository whose
85
ancestry was being reported incorrectly.
86
garbage_inventories: The number of inventory objects without revisions
87
that were garbage collected.
88
"""
89
self.pb = ui.ui_factory.progress_bar()
90
self.repo.lock_write()
91
try:
92
self._reweave_inventory()
93
finally:
94
self.repo.unlock()
95
96
def _reweave_inventory(self):
97
"""Regenerate the inventory weave for the repository from scratch."""
98
self.pb.update('Reading inventory data.')
99
self.inventory = self.repo.get_inventory_weave()
100
# the total set of revisions to process
101
self.pending = set([file_id for file_id in self.repo.revision_store])
102
103
# mapping from revision_id to parents
104
self._rev_graph = {}
105
# errors that we detect
106
self.inconsistent_parents = 0
107
# we need the revision id of each revision and its available parents list
108
self._setup_steps(len(self.pending))
109
for rev_id in self.pending:
110
# put a revision into the graph.
111
self._graph_revision(rev_id)
112
# we gc unreferenced inventories too
113
self.garbage_inventories = len(self.inventory.names()) \
114
- len(self._rev_graph)
115
116
if not self.inconsistent_parents and not self.garbage_inventories:
117
self.pb.note('Inventory ok.')
118
return
119
self.pb.update('Backing up inventory...', 0, 0)
120
self.repo.control_weaves.put_weave('inventory.backup',
121
self.inventory,
122
self.repo.get_transaction())
123
self.pb.note('Backup Inventory created.')
124
# asking for '' should never return a non-empty weave
125
new_inventory = self.repo.control_weaves.get_weave_or_empty('',
126
self.repo.get_transaction())
127
128
# we have topological order of revisions and non ghost parents ready.
129
self._setup_steps(len(self._rev_graph))
130
for rev_id in TopoSorter(self._rev_graph.items()).iter_topo_order():
131
parents = self._rev_graph[rev_id]
132
# double check this really is in topological order.
133
unavailable = [p for p in parents if p not in new_inventory]
134
assert len(unavailable) == 0
135
# this entry has all the non ghost parents in the inventory
136
# file already.
137
self._reweave_step('adding inventories')
138
new_inventory.add(rev_id, parents, self.inventory.get(rev_id))
139
140
# if this worked, the set of new_inventory.names should equal
141
# self.pending
142
assert set(new_inventory.names()) == self.pending
143
self.pb.update('Writing weave')
144
self.repo.control_weaves.put_weave('inventory',
145
new_inventory,
146
self.repo.get_transaction())
147
self.inventory = None
148
self.pb.note('Inventory regenerated.')
149
150
def _setup_steps(self, new_total):
151
"""Setup the markers we need to control the progress bar."""
152
self.total = new_total
153
self.count = 0
154
155
def _graph_revision(self, rev_id):
156
"""Load a revision into the revision graph."""
157
# pick a random revision
158
# analyse revision id rev_id and put it in the stack.
159
self._reweave_step('loading revisions')
160
rev = self.repo.get_revision_reconcile(rev_id)
161
assert rev.revision_id == rev_id
162
parents = []
163
for parent in rev.parent_ids:
164
if self._parent_is_available(parent):
165
parents.append(parent)
166
else:
167
mutter('found ghost %s', parent)
168
self._rev_graph[rev_id] = parents
169
if set(self.inventory.parent_names(rev_id)) != set(parents):
170
self.inconsistent_parents += 1
171
172
def _parent_is_available(self, parent):
173
"""True if parent is a fully available revision
174
175
A fully available revision has a inventory and a revision object in the
176
repository.
177
"""
178
return (parent in self._rev_graph or
179
(parent in self.inventory and self.repo.has_revision(parent)))
180
181
def _reweave_step(self, message):
182
"""Mark a single step of regeneration complete."""
183
self.pb.update(message, self.count, self.total)
184
self.count += 1
Loggerhead is a web-based interface for Breezy
Version: 2.0.1