~bzr-pqm/bzr/bzr.dev

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
# Copyright (C) 2005, 2006, 2008 Canonical Ltd
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA


"""Copying of history from one branch to another.

The basic plan is that every branch knows the history of everything
that has merged into it.  As the first step of a merge, pull, or
branch operation we copy history from the source into the destination
branch.

The copying is done in a slightly complicated order.  We don't want to
add a revision to the store until everything it refers to is also
stored, so that if a revision is present we can totally recreate it.
However, we can't know what files are included in a revision until we
read its inventory.  So we query the inventory store of the source for
the ids we need, and then pull those ids and then return to the inventories.
"""

import operator

import bzrlib
import bzrlib.errors as errors
from bzrlib.errors import InstallFailed
from bzrlib.progress import ProgressPhase
from bzrlib.revision import NULL_REVISION
from bzrlib.tsort import topo_sort
from bzrlib.trace import mutter
import bzrlib.ui
from bzrlib.versionedfile import FulltextContentFactory

# TODO: Avoid repeatedly opening weaves so many times.

# XXX: This doesn't handle ghost (not present in branch) revisions at
# all yet.  I'm not sure they really should be supported.

# NOTE: This doesn't copy revisions which may be present but not
# merged into the last revision.  I'm not sure we want to do that.

# - get a list of revisions that need to be pulled in
# - for each one, pull in that revision file
#   and get the inventory, and store the inventory with right
#   parents.
# - and get the ancestry, and store that with right parents too
# - and keep a note of all file ids and version seen
# - then go through all files; for each one get the weave,
#   and add in all file versions


class RepoFetcher(object):
    """Pull revisions and texts from one repository to another.

    last_revision
        if set, try to limit to the data this revision references.

    This should not be used directly, it's essential a object to encapsulate
    the logic in InterRepository.fetch().
    """

    def __init__(self, to_repository, from_repository, last_revision=None,
        pb=None, find_ghosts=True, fetch_spec=None):
        """Create a repo fetcher.

        :param find_ghosts: If True search the entire history for ghosts.
        :param _write_group_acquired_callable: Don't use; this parameter only
            exists to facilitate a hack done in InterPackRepo.fetch.  We would
            like to remove this parameter.
        """
        if to_repository.has_same_location(from_repository):
            # repository.fetch should be taking care of this case.
            raise errors.BzrError('RepoFetcher run '
                    'between two objects at the same location: '
                    '%r and %r' % (to_repository, from_repository))
        self.to_repository = to_repository
        self.from_repository = from_repository
        self.sink = to_repository._get_sink()
        # must not mutate self._last_revision as its potentially a shared instance
        self._last_revision = last_revision
        self._fetch_spec = fetch_spec
        self.find_ghosts = find_ghosts
        if pb is None:
            self.pb = bzrlib.ui.ui_factory.nested_progress_bar()
            self.nested_pb = self.pb
        else:
            self.pb = pb
            self.nested_pb = None
        self.from_repository.lock_read()
        try:
            try:
                self.__fetch()
            finally:
                if self.nested_pb is not None:
                    self.nested_pb.finished()
        finally:
            self.from_repository.unlock()

    def __fetch(self):
        """Primary worker function.

        This initialises all the needed variables, and then fetches the
        requested revisions, finally clearing the progress bar.
        """
        # Roughly this is what we're aiming for fetch to become:
        #
        # missing = self.sink.insert_stream(self.source.get_stream(search))
        # if missing:
        #     missing = self.sink.insert_stream(self.source.get_items(missing))
        # assert not missing
        self.count_total = 0
        self.file_ids_names = {}
        pp = ProgressPhase('Transferring', 4, self.pb)
        try:
            pp.next_phase()
            search = self._revids_to_fetch()
            if search is None:
                return
            self._fetch_everything_for_search(search, pp)
        finally:
            self.pb.clear()

    def _fetch_everything_for_search(self, search, pp):
        """Fetch all data for the given set of revisions."""
        # The first phase is "file".  We pass the progress bar for it directly
        # into item_keys_introduced_by, which has more information about how
        # that phase is progressing than we do.  Progress updates for the other
        # phases are taken care of in this function.
        # XXX: there should be a clear owner of the progress reporting.  Perhaps
        # item_keys_introduced_by should have a richer API than it does at the
        # moment, so that it can feed the progress information back to this
        # function?
        if (self.from_repository._format.rich_root_data and
            not self.to_repository._format.rich_root_data):
            raise errors.IncompatibleRepositories(
                self.from_repository, self.to_repository,
                "different rich-root support")
        self.pb = bzrlib.ui.ui_factory.nested_progress_bar()
        try:
            source = self.from_repository._get_source(
                self.to_repository._format)
            stream = source.get_stream(search)
            from_format = self.from_repository._format
            resume_tokens, missing_keys = self.sink.insert_stream(
                stream, from_format, [])
            if missing_keys:
                stream = source.get_stream_for_missing_keys(missing_keys)
                resume_tokens, missing_keys = self.sink.insert_stream(
                    stream, from_format, resume_tokens)
            if missing_keys:
                raise AssertionError(
                    "second push failed to complete a fetch %r." % (
                        missing_keys,))
            if resume_tokens:
                raise AssertionError(
                    "second push failed to commit the fetch %r." % (
                        resume_tokens,))
            self.sink.finished()
        finally:
            if self.pb is not None:
                self.pb.finished()

    def _revids_to_fetch(self):
        """Determines the exact revisions needed from self.from_repository to
        install self._last_revision in self.to_repository.

        If no revisions need to be fetched, then this just returns None.
        """
        if self._fetch_spec is not None:
            return self._fetch_spec
        mutter('fetch up to rev {%s}', self._last_revision)
        if self._last_revision is NULL_REVISION:
            # explicit limit of no revisions needed
            return None
        if (self._last_revision is not None and
            self.to_repository.has_revision(self._last_revision)):
            return None
        try:
            return self.to_repository.search_missing_revision_ids(
                self.from_repository, self._last_revision,
                find_ghosts=self.find_ghosts)
        except errors.NoSuchRevision, e:
            raise InstallFailed([self._last_revision])


class Inter1and2Helper(object):
    """Helper for operations that convert data from model 1 and 2

    This is for use by fetchers and converters.
    """

    def __init__(self, source):
        """Constructor.

        :param source: The repository data comes from
        """
        self.source = source

    def iter_rev_trees(self, revs):
        """Iterate through RevisionTrees efficiently.

        Additionally, the inventory's revision_id is set if unset.

        Trees are retrieved in batches of 100, and then yielded in the order
        they were requested.

        :param revs: A list of revision ids
        """
        # In case that revs is not a list.
        revs = list(revs)
        while revs:
            for tree in self.source.revision_trees(revs[:100]):
                if tree.inventory.revision_id is None:
                    tree.inventory.revision_id = tree.get_revision_id()
                yield tree
            revs = revs[100:]

    def _find_root_ids(self, revs, parent_map, graph):
        revision_root = {}
        planned_versions = {}
        for tree in self.iter_rev_trees(revs):
            revision_id = tree.inventory.root.revision
            root_id = tree.get_root_id()
            planned_versions.setdefault(root_id, []).append(revision_id)
            revision_root[revision_id] = root_id
        # Find out which parents we don't already know root ids for
        parents = set()
        for revision_parents in parent_map.itervalues():
            parents.update(revision_parents)
        parents.difference_update(revision_root.keys() + [NULL_REVISION])
        # Limit to revisions present in the versionedfile
        parents = graph.get_parent_map(parents).keys()
        for tree in self.iter_rev_trees(parents):
            root_id = tree.get_root_id()
            revision_root[tree.get_revision_id()] = root_id
        return revision_root, planned_versions

    def generate_root_texts(self, revs):
        """Generate VersionedFiles for all root ids.

        :param revs: the revisions to include
        """
        graph = self.source.get_graph()
        parent_map = graph.get_parent_map(revs)
        rev_order = topo_sort(parent_map)
        rev_id_to_root_id, root_id_to_rev_ids = self._find_root_ids(
            revs, parent_map, graph)
        root_id_order = [(rev_id_to_root_id[rev_id], rev_id) for rev_id in
            rev_order]
        # Guaranteed stable, this groups all the file id operations together
        # retaining topological order within the revisions of a file id.
        # File id splits and joins would invalidate this, but they don't exist
        # yet, and are unlikely to in non-rich-root environments anyway.
        root_id_order.sort(key=operator.itemgetter(0))
        # Create a record stream containing the roots to create.
        def yield_roots():
            for key in root_id_order:
                root_id, rev_id = key
                rev_parents = parent_map[rev_id]
                # We drop revision parents with different file-ids, because
                # that represents a rename of the root to a different location
                # - its not actually a parent for us. (We could look for that
                # file id in the revision tree at considerably more expense,
                # but for now this is sufficient (and reconcile will catch and
                # correct this anyway).
                # When a parent revision is a ghost, we guess that its root id
                # was unchanged (rather than trimming it from the parent list).
                parent_keys = tuple((root_id, parent) for parent in rev_parents
                    if parent != NULL_REVISION and
                        rev_id_to_root_id.get(parent, root_id) == root_id)
                yield FulltextContentFactory(key, parent_keys, None, '')
        return [('texts', yield_roots())]