1
# (C) 2005, 2006 Canonical Limited.
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
"""Reconcilers are able to fix some potential data errors in a branch."""
20
__all__ = ['reconcile', 'Reconciler', 'RepoReconciler']
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
32
"""Reconcile the data in dir.
34
Currently this is limited to a inventory 'reweave'.
36
This is a convenience method, for using a Reconciler object.
38
Directly using Reconciler is recommended for library users that
39
desire fine grained control or analysis of the found issues.
41
reconciler = Reconciler(dir)
42
reconciler.reconcile()
45
class Reconciler(object):
46
"""Reconcilers are used to reconcile existing data."""
48
def __init__(self, dir):
52
"""Perform reconciliation.
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.
60
self.pb = ui.ui_factory.nested_progress_bar()
67
"""Helper function for performing reconciliation."""
68
self.repo = self.bzrdir.find_repository()
69
self.pb.note('Reconciling repository %s',
70
self.repo.bzrdir.root_transport.base)
71
repo_reconciler = RepoReconciler(self.repo)
72
repo_reconciler.reconcile()
73
self.inconsistent_parents = repo_reconciler.inconsistent_parents
74
self.garbage_inventories = repo_reconciler.garbage_inventories
75
self.pb.note('Reconciliation complete.')
78
class RepoReconciler(object):
79
"""Reconciler that reconciles a repository.
81
Currently this consists of an inventory reweave with revision cross-checks.
84
def __init__(self, repo):
88
"""Perform reconciliation.
90
After reconciliation the following attributes document found issues:
91
inconsistent_parents: The number of revisions in the repository whose
92
ancestry was being reported incorrectly.
93
garbage_inventories: The number of inventory objects without revisions
94
that were garbage collected.
96
self.repo.lock_write()
98
self.pb = ui.ui_factory.nested_progress_bar()
100
self._reweave_inventory()
106
def _reweave_inventory(self):
107
"""Regenerate the inventory weave for the repository from scratch."""
108
transaction = self.repo.get_transaction()
109
self.pb.update('Reading inventory data.')
110
self.inventory = self.repo.get_inventory_weave()
111
# the total set of revisions to process
112
self.pending = set([rev_id for rev_id in self.repo._revision_store.all_revision_ids(transaction)])
114
# mapping from revision_id to parents
116
# errors that we detect
117
self.inconsistent_parents = 0
118
# we need the revision id of each revision and its available parents list
119
self._setup_steps(len(self.pending))
120
for rev_id in self.pending:
121
# put a revision into the graph.
122
self._graph_revision(rev_id)
123
self._check_garbage_inventories()
124
if not self.inconsistent_parents and not self.garbage_inventories:
125
self.pb.note('Inventory ok.')
127
self.pb.update('Backing up inventory...', 0, 0)
128
self.repo.control_weaves.copy(self.inventory, 'inventory.backup', self.repo.get_transaction())
129
self.pb.note('Backup Inventory created.')
130
# asking for '' should never return a non-empty weave
131
new_inventory = self.repo.control_weaves.get_empty('inventory.new',
132
self.repo.get_transaction())
134
# we have topological order of revisions and non ghost parents ready.
135
self._setup_steps(len(self._rev_graph))
136
for rev_id in TopoSorter(self._rev_graph.items()).iter_topo_order():
137
parents = self._rev_graph[rev_id]
138
# double check this really is in topological order.
139
unavailable = [p for p in parents if p not in new_inventory]
140
assert len(unavailable) == 0
141
# this entry has all the non ghost parents in the inventory
143
self._reweave_step('adding inventories')
144
new_inventory.add_lines(rev_id, parents, self.inventory.get_lines(rev_id))
146
# if this worked, the set of new_inventory.names should equal
148
assert set(new_inventory.versions()) == self.pending
149
self.pb.update('Writing weave')
150
self.repo.control_weaves.copy(new_inventory, 'inventory', self.repo.get_transaction())
151
self.repo.control_weaves.delete('inventory.new', self.repo.get_transaction())
152
self.inventory = None
153
self.pb.note('Inventory regenerated.')
155
def _setup_steps(self, new_total):
156
"""Setup the markers we need to control the progress bar."""
157
self.total = new_total
160
def _graph_revision(self, rev_id):
161
"""Load a revision into the revision graph."""
162
# pick a random revision
163
# analyse revision id rev_id and put it in the stack.
164
self._reweave_step('loading revisions')
165
rev = self.repo.get_revision_reconcile(rev_id)
166
assert rev.revision_id == rev_id
168
for parent in rev.parent_ids:
169
if self._parent_is_available(parent):
170
parents.append(parent)
172
mutter('found ghost %s', parent)
173
self._rev_graph[rev_id] = parents
174
if set(self.inventory.get_parents(rev_id)) != set(parents):
175
self.inconsistent_parents += 1
176
mutter('Inconsistent inventory parents: id {%s} '
177
'inventory claims %r, '
178
'available parents are %r, '
179
'unavailable parents are %r',
181
set(self.inventory.get_parents(rev_id)),
183
set(rev.parent_ids).difference(set(parents)))
185
def _check_garbage_inventories(self):
186
"""Check for garbage inventories which we cannot trust
188
We cant trust them because their pre-requisite file data may not
189
be present - all we know is that their revision was not installed.
191
inventories = set(self.inventory.versions())
192
revisions = set(self._rev_graph.keys())
193
garbage = inventories.difference(revisions)
194
self.garbage_inventories = len(garbage)
195
for revision_id in garbage:
196
mutter('Garbage inventory {%s} found.', revision_id)
198
def _parent_is_available(self, parent):
199
"""True if parent is a fully available revision
201
A fully available revision has a inventory and a revision object in the
204
return (parent in self._rev_graph or
205
(parent in self.inventory and self.repo.has_revision(parent)))
207
def _reweave_step(self, message):
208
"""Mark a single step of regeneration complete."""
209
self.pb.update(message, self.count, self.total)
added
removed
bzrlib/reconcile.py
