1
# Copyright (C) 2005, 2006, 2008, 2009 Canonical Ltd
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., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
18
"""Copying of history from one branch to another.
20
The basic plan is that every branch knows the history of everything
21
that has merged into it. As the first step of a merge, pull, or
22
branch operation we copy history from the source into the destination
33
from bzrlib.revision import NULL_REVISION
34
from bzrlib.tsort import topo_sort
35
from bzrlib.trace import mutter
37
from bzrlib.versionedfile import FulltextContentFactory
40
class RepoFetcher(object):
41
"""Pull revisions and texts from one repository to another.
43
This should not be used directly, it's essential a object to encapsulate
44
the logic in InterRepository.fetch().
47
def __init__(self, to_repository, from_repository, last_revision=None,
48
pb=None, find_ghosts=True, fetch_spec=None):
49
"""Create a repo fetcher.
51
:param last_revision: If set, try to limit to the data this revision
53
:param find_ghosts: If True search the entire history for ghosts.
54
:param _write_group_acquired_callable: Don't use; this parameter only
55
exists to facilitate a hack done in InterPackRepo.fetch. We would
56
like to remove this parameter.
57
:param pb: ProgressBar object to use; deprecated and ignored.
58
This method will just create one on top of the stack.
61
symbol_versioning.warn(
62
symbol_versioning.deprecated_in((1, 14, 0))
63
% "pb parameter to RepoFetcher.__init__")
64
# and for simplicity it is in fact ignored
65
if to_repository.has_same_location(from_repository):
66
# repository.fetch should be taking care of this case.
67
raise errors.BzrError('RepoFetcher run '
68
'between two objects at the same location: '
69
'%r and %r' % (to_repository, from_repository))
70
self.to_repository = to_repository
71
self.from_repository = from_repository
72
self.sink = to_repository._get_sink()
73
# must not mutate self._last_revision as its potentially a shared instance
74
self._last_revision = last_revision
75
self._fetch_spec = fetch_spec
76
self.find_ghosts = find_ghosts
77
self.from_repository.lock_read()
78
mutter("Using fetch logic to copy between %s(%s) and %s(%s)",
79
self.from_repository, self.from_repository._format,
80
self.to_repository, self.to_repository._format)
84
self.from_repository.unlock()
87
"""Primary worker function.
89
This initialises all the needed variables, and then fetches the
90
requested revisions, finally clearing the progress bar.
92
# Roughly this is what we're aiming for fetch to become:
94
# missing = self.sink.insert_stream(self.source.get_stream(search))
96
# missing = self.sink.insert_stream(self.source.get_items(missing))
99
self.file_ids_names = {}
100
pb = bzrlib.ui.ui_factory.nested_progress_bar()
101
pb.show_pct = pb.show_count = False
103
pb.update("Finding revisions", 0, 2)
104
search = self._revids_to_fetch()
107
pb.update("Fetching revisions", 1, 2)
108
self._fetch_everything_for_search(search)
112
def _fetch_everything_for_search(self, search):
113
"""Fetch all data for the given set of revisions."""
114
# The first phase is "file". We pass the progress bar for it directly
115
# into item_keys_introduced_by, which has more information about how
116
# that phase is progressing than we do. Progress updates for the other
117
# phases are taken care of in this function.
118
# XXX: there should be a clear owner of the progress reporting. Perhaps
119
# item_keys_introduced_by should have a richer API than it does at the
120
# moment, so that it can feed the progress information back to this
122
if (self.from_repository._format.rich_root_data and
123
not self.to_repository._format.rich_root_data):
124
raise errors.IncompatibleRepositories(
125
self.from_repository, self.to_repository,
126
"different rich-root support")
127
pb = bzrlib.ui.ui_factory.nested_progress_bar()
129
pb.update("Get stream source")
130
source = self.from_repository._get_source(
131
self.to_repository._format)
132
stream = source.get_stream(search)
133
from_format = self.from_repository._format
134
pb.update("Inserting stream")
135
resume_tokens, missing_keys = self.sink.insert_stream(
136
stream, from_format, [])
137
if self.to_repository._fallback_repositories:
139
self._parent_inventories(search.get_keys()))
141
pb.update("Missing keys")
142
stream = source.get_stream_for_missing_keys(missing_keys)
143
pb.update("Inserting missing keys")
144
resume_tokens, missing_keys = self.sink.insert_stream(
145
stream, from_format, resume_tokens)
147
raise AssertionError(
148
"second push failed to complete a fetch %r." % (
151
raise AssertionError(
152
"second push failed to commit the fetch %r." % (
154
pb.update("Finishing stream")
159
def _revids_to_fetch(self):
160
"""Determines the exact revisions needed from self.from_repository to
161
install self._last_revision in self.to_repository.
163
If no revisions need to be fetched, then this just returns None.
165
if self._fetch_spec is not None:
166
return self._fetch_spec
167
mutter('fetch up to rev {%s}', self._last_revision)
168
if self._last_revision is NULL_REVISION:
169
# explicit limit of no revisions needed
171
if (self._last_revision is not None and
172
self.to_repository.has_revision(self._last_revision)):
174
return self.to_repository.search_missing_revision_ids(
175
self.from_repository, self._last_revision,
176
find_ghosts=self.find_ghosts)
178
def _parent_inventories(self, revision_ids):
179
# Find all the parent revisions referenced by the stream, but
180
# not present in the stream, and make sure we send their
182
parent_maps = self.to_repository.get_parent_map(revision_ids)
184
map(parents.update, parent_maps.itervalues())
185
parents.discard(NULL_REVISION)
186
parents.difference_update(revision_ids)
187
missing_keys = set(('inventories', rev_id) for rev_id in parents)
191
class Inter1and2Helper(object):
192
"""Helper for operations that convert data from model 1 and 2
194
This is for use by fetchers and converters.
197
def __init__(self, source):
200
:param source: The repository data comes from
204
def iter_rev_trees(self, revs):
205
"""Iterate through RevisionTrees efficiently.
207
Additionally, the inventory's revision_id is set if unset.
209
Trees are retrieved in batches of 100, and then yielded in the order
212
:param revs: A list of revision ids
214
# In case that revs is not a list.
217
for tree in self.source.revision_trees(revs[:100]):
218
if tree.inventory.revision_id is None:
219
tree.inventory.revision_id = tree.get_revision_id()
223
def _find_root_ids(self, revs, parent_map, graph):
225
planned_versions = {}
226
for tree in self.iter_rev_trees(revs):
227
revision_id = tree.inventory.root.revision
228
root_id = tree.get_root_id()
229
planned_versions.setdefault(root_id, []).append(revision_id)
230
revision_root[revision_id] = root_id
231
# Find out which parents we don't already know root ids for
233
for revision_parents in parent_map.itervalues():
234
parents.update(revision_parents)
235
parents.difference_update(revision_root.keys() + [NULL_REVISION])
236
# Limit to revisions present in the versionedfile
237
parents = graph.get_parent_map(parents).keys()
238
for tree in self.iter_rev_trees(parents):
239
root_id = tree.get_root_id()
240
revision_root[tree.get_revision_id()] = root_id
241
return revision_root, planned_versions
243
def generate_root_texts(self, revs):
244
"""Generate VersionedFiles for all root ids.
246
:param revs: the revisions to include
248
graph = self.source.get_graph()
249
parent_map = graph.get_parent_map(revs)
250
rev_order = topo_sort(parent_map)
251
rev_id_to_root_id, root_id_to_rev_ids = self._find_root_ids(
252
revs, parent_map, graph)
253
root_id_order = [(rev_id_to_root_id[rev_id], rev_id) for rev_id in
255
# Guaranteed stable, this groups all the file id operations together
256
# retaining topological order within the revisions of a file id.
257
# File id splits and joins would invalidate this, but they don't exist
258
# yet, and are unlikely to in non-rich-root environments anyway.
259
root_id_order.sort(key=operator.itemgetter(0))
260
# Create a record stream containing the roots to create.
262
for key in root_id_order:
263
root_id, rev_id = key
264
rev_parents = parent_map[rev_id]
265
# We drop revision parents with different file-ids, because
266
# that represents a rename of the root to a different location
267
# - its not actually a parent for us. (We could look for that
268
# file id in the revision tree at considerably more expense,
269
# but for now this is sufficient (and reconcile will catch and
270
# correct this anyway).
271
# When a parent revision is a ghost, we guess that its root id
272
# was unchanged (rather than trimming it from the parent list).
273
parent_keys = tuple((root_id, parent) for parent in rev_parents
274
if parent != NULL_REVISION and
275
rev_id_to_root_id.get(parent, root_id) == root_id)
276
yield FulltextContentFactory(key, parent_keys, None, '')
277
return [('texts', yield_roots())]
added
removed
bzrlib/fetch.py
