1
# Copyright (C) 2005-2010 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
28
from bzrlib.lazy_import import lazy_import
29
lazy_import(globals(), """
41
from bzrlib.revision import NULL_REVISION
42
from bzrlib.trace import mutter
45
class RepoFetcher(object):
46
"""Pull revisions and texts from one repository to another.
48
This should not be used directly, it's essential a object to encapsulate
49
the logic in InterRepository.fetch().
52
def __init__(self, to_repository, from_repository, last_revision=None,
53
find_ghosts=True, fetch_spec=None):
54
"""Create a repo fetcher.
56
:param last_revision: If set, try to limit to the data this revision
58
:param find_ghosts: If True search the entire history for ghosts.
60
# repository.fetch has the responsibility for short-circuiting
61
# attempts to copy between a repository and itself.
62
self.to_repository = to_repository
63
self.from_repository = from_repository
64
self.sink = to_repository._get_sink()
65
# must not mutate self._last_revision as its potentially a shared instance
66
self._last_revision = last_revision
67
self._fetch_spec = fetch_spec
68
self.find_ghosts = find_ghosts
69
self.from_repository.lock_read()
70
mutter("Using fetch logic to copy between %s(%s) and %s(%s)",
71
self.from_repository, self.from_repository._format,
72
self.to_repository, self.to_repository._format)
76
self.from_repository.unlock()
79
"""Primary worker function.
81
This initialises all the needed variables, and then fetches the
82
requested revisions, finally clearing the progress bar.
84
# Roughly this is what we're aiming for fetch to become:
86
# missing = self.sink.insert_stream(self.source.get_stream(search))
88
# missing = self.sink.insert_stream(self.source.get_items(missing))
91
self.file_ids_names = {}
92
pb = ui.ui_factory.nested_progress_bar()
93
pb.show_pct = pb.show_count = False
95
pb.update("Finding revisions", 0, 2)
96
search = self._revids_to_fetch()
99
pb.update("Fetching revisions", 1, 2)
100
self._fetch_everything_for_search(search)
104
def _fetch_everything_for_search(self, search):
105
"""Fetch all data for the given set of revisions."""
106
# The first phase is "file". We pass the progress bar for it directly
107
# into item_keys_introduced_by, which has more information about how
108
# that phase is progressing than we do. Progress updates for the other
109
# phases are taken care of in this function.
110
# XXX: there should be a clear owner of the progress reporting. Perhaps
111
# item_keys_introduced_by should have a richer API than it does at the
112
# moment, so that it can feed the progress information back to this
114
if (self.from_repository._format.rich_root_data and
115
not self.to_repository._format.rich_root_data):
116
raise errors.IncompatibleRepositories(
117
self.from_repository, self.to_repository,
118
"different rich-root support")
119
pb = ui.ui_factory.nested_progress_bar()
121
pb.update("Get stream source")
122
source = self.from_repository._get_source(
123
self.to_repository._format)
124
stream = source.get_stream(search)
125
from_format = self.from_repository._format
126
pb.update("Inserting stream")
127
resume_tokens, missing_keys = self.sink.insert_stream(
128
stream, from_format, [])
129
if self.to_repository._fallback_repositories:
131
self._parent_inventories(search.get_keys()))
133
pb.update("Missing keys")
134
stream = source.get_stream_for_missing_keys(missing_keys)
135
pb.update("Inserting missing keys")
136
resume_tokens, missing_keys = self.sink.insert_stream(
137
stream, from_format, resume_tokens)
139
raise AssertionError(
140
"second push failed to complete a fetch %r." % (
143
raise AssertionError(
144
"second push failed to commit the fetch %r." % (
146
pb.update("Finishing stream")
151
def _revids_to_fetch(self):
152
"""Determines the exact revisions needed from self.from_repository to
153
install self._last_revision in self.to_repository.
155
If no revisions need to be fetched, then this just returns None.
157
if self._fetch_spec is not None:
158
return self._fetch_spec
159
mutter('fetch up to rev {%s}', self._last_revision)
160
if self._last_revision is NULL_REVISION:
161
# explicit limit of no revisions needed
163
return self.to_repository.search_missing_revision_ids(
164
self.from_repository, self._last_revision,
165
find_ghosts=self.find_ghosts)
167
def _parent_inventories(self, revision_ids):
168
# Find all the parent revisions referenced by the stream, but
169
# not present in the stream, and make sure we send their
171
parent_maps = self.to_repository.get_parent_map(revision_ids)
173
map(parents.update, parent_maps.itervalues())
174
parents.discard(NULL_REVISION)
175
parents.difference_update(revision_ids)
176
missing_keys = set(('inventories', rev_id) for rev_id in parents)
180
class Inter1and2Helper(object):
181
"""Helper for operations that convert data from model 1 and 2
183
This is for use by fetchers and converters.
186
def __init__(self, source):
189
:param source: The repository data comes from
193
def iter_rev_trees(self, revs):
194
"""Iterate through RevisionTrees efficiently.
196
Additionally, the inventory's revision_id is set if unset.
198
Trees are retrieved in batches of 100, and then yielded in the order
201
:param revs: A list of revision ids
203
# In case that revs is not a list.
206
for tree in self.source.revision_trees(revs[:100]):
207
if tree.inventory.revision_id is None:
208
tree.inventory.revision_id = tree.get_revision_id()
212
def _find_root_ids(self, revs, parent_map, graph):
214
for tree in self.iter_rev_trees(revs):
215
revision_id = tree.inventory.root.revision
216
root_id = tree.get_root_id()
217
revision_root[revision_id] = root_id
218
# Find out which parents we don't already know root ids for
220
for revision_parents in parent_map.itervalues():
221
parents.update(revision_parents)
222
parents.difference_update(revision_root.keys() + [NULL_REVISION])
223
# Limit to revisions present in the versionedfile
224
parents = graph.get_parent_map(parents).keys()
225
for tree in self.iter_rev_trees(parents):
226
root_id = tree.get_root_id()
227
revision_root[tree.get_revision_id()] = root_id
230
def generate_root_texts(self, revs):
231
"""Generate VersionedFiles for all root ids.
233
:param revs: the revisions to include
235
graph = self.source.get_graph()
236
parent_map = graph.get_parent_map(revs)
237
rev_order = tsort.topo_sort(parent_map)
238
rev_id_to_root_id = self._find_root_ids(revs, parent_map, graph)
239
root_id_order = [(rev_id_to_root_id[rev_id], rev_id) for rev_id in
241
# Guaranteed stable, this groups all the file id operations together
242
# retaining topological order within the revisions of a file id.
243
# File id splits and joins would invalidate this, but they don't exist
244
# yet, and are unlikely to in non-rich-root environments anyway.
245
root_id_order.sort(key=operator.itemgetter(0))
246
# Create a record stream containing the roots to create.
248
# XXX: not covered by tests, should have a flag to always run
249
# this. -- mbp 20100129
250
graph = self.source_repo.get_known_graph_ancestry(revs)
251
new_roots_stream = _new_root_data_stream(
252
root_id_order, rev_id_to_root_id, parent_map, self.source, graph)
253
return [('texts', new_roots_stream)]
256
def _get_rich_root_heads_graph(source_repo, revision_ids):
257
"""Get a Graph object suitable for asking heads() for new rich roots."""
261
def _new_root_data_stream(
262
root_keys_to_create, rev_id_to_root_id_map, parent_map, repo, graph=None):
263
"""Generate a texts substream of synthesised root entries.
265
Used in fetches that do rich-root upgrades.
267
:param root_keys_to_create: iterable of (root_id, rev_id) pairs describing
268
the root entries to create.
269
:param rev_id_to_root_id_map: dict of known rev_id -> root_id mappings for
270
calculating the parents. If a parent rev_id is not found here then it
271
will be recalculated.
272
:param parent_map: a parent map for all the revisions in
274
:param graph: a graph to use instead of repo.get_graph().
276
for root_key in root_keys_to_create:
277
root_id, rev_id = root_key
278
parent_keys = _parent_keys_for_root_version(
279
root_id, rev_id, rev_id_to_root_id_map, parent_map, repo, graph)
280
yield versionedfile.FulltextContentFactory(
281
root_key, parent_keys, None, '')
284
def _parent_keys_for_root_version(
285
root_id, rev_id, rev_id_to_root_id_map, parent_map, repo, graph=None):
286
"""Get the parent keys for a given root id.
288
A helper function for _new_root_data_stream.
290
# Include direct parents of the revision, but only if they used the same
291
# root_id and are heads.
292
rev_parents = parent_map[rev_id]
294
for parent_id in rev_parents:
295
if parent_id == NULL_REVISION:
297
if parent_id not in rev_id_to_root_id_map:
298
# We probably didn't read this revision, go spend the extra effort
301
tree = repo.revision_tree(parent_id)
302
except errors.NoSuchRevision:
303
# Ghost, fill out rev_id_to_root_id in case we encounter this
305
# But set parent_root_id to None since we don't really know
306
parent_root_id = None
308
parent_root_id = tree.get_root_id()
309
rev_id_to_root_id_map[parent_id] = None
311
# rev_id_to_root_id_map[parent_id] = parent_root_id
312
# memory consumption maybe?
314
parent_root_id = rev_id_to_root_id_map[parent_id]
315
if root_id == parent_root_id:
316
# With stacking we _might_ want to refer to a non-local revision,
317
# but this code path only applies when we have the full content
318
# available, so ghosts really are ghosts, not just the edge of
320
parent_ids.append(parent_id)
322
# root_id may be in the parent anyway.
324
tree = repo.revision_tree(parent_id)
325
except errors.NoSuchRevision:
326
# ghost, can't refer to it.
330
parent_ids.append(tree.inventory[root_id].revision)
331
except errors.NoSuchId:
334
# Drop non-head parents
336
graph = repo.get_graph()
337
heads = graph.heads(parent_ids)
339
for parent_id in parent_ids:
340
if parent_id in heads and parent_id not in selected_ids:
341
selected_ids.append(parent_id)
342
parent_keys = [(root_id, parent_id) for parent_id in selected_ids]
added
removed
bzrlib/fetch.py
