13
13
# You should have received a copy of the GNU General Public License
14
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
15
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
17
17
"""Tree classes, representing directory at point in time.
22
from bzrlib.lazy_import import lazy_import
23
lazy_import(globals(), """
21
from collections import deque
22
from cStringIO import StringIO
26
25
from bzrlib import (
27
26
conflicts as _mod_conflicts,
34
29
revision as _mod_revision,
38
from bzrlib.i18n import gettext
41
33
from bzrlib.decorators import needs_read_lock
34
from bzrlib.errors import BzrError, BzrCheckError
35
from bzrlib import errors
36
from bzrlib.inventory import Inventory, InventoryFile
42
37
from bzrlib.inter import InterObject
43
from bzrlib.symbol_versioning import (
38
from bzrlib.osutils import fingerprint_file
39
import bzrlib.revision
40
from bzrlib.trace import mutter, note
49
43
class Tree(object):
50
44
"""Abstract file tree.
52
46
There are several subclasses:
54
48
* `WorkingTree` exists as files on disk editable by the user.
56
50
* `RevisionTree` is a tree as recorded at some point in the past.
52
Trees contain an `Inventory` object, and also know how to retrieve
53
file texts mentioned in the inventory, either from a working
54
directory or from a store.
56
It is possible for trees to contain files that are not described
57
in their inventory or vice versa; for this use `filenames()`.
58
59
Trees can be compared, etc, regardless of whether they are working
59
60
trees or versioned trees.
62
def has_versioned_directories(self):
63
"""Whether this tree can contain explicitly versioned directories.
65
This defaults to True, but some implementations may want to override
70
63
def changes_from(self, other, want_unchanged=False, specific_files=None,
71
64
extra_trees=None, require_versioned=False, include_root=False,
72
65
want_unversioned=False):
186
198
The yield order (ignoring root) would be::
188
199
a, f, a/b, a/d, a/b/c, a/d/e, f/g
190
:param yield_parents: If True, yield the parents from the root leading
191
down to specific_file_ids that have been requested. This has no
192
impact if specific_file_ids is None.
194
raise NotImplementedError(self.iter_entries_by_dir)
196
def list_files(self, include_root=False, from_dir=None, recursive=True):
197
"""List all files in this tree.
199
:param include_root: Whether to include the entry for the tree root
200
:param from_dir: Directory under which to list files
201
:param recursive: Whether to list files recursively
202
:return: iterator over tuples of (path, versioned, kind, file_id,
205
raise NotImplementedError(self.list_files)
201
return self.inventory.iter_entries_by_dir(
202
specific_file_ids=specific_file_ids)
207
204
def iter_references(self):
208
if self.supports_tree_reference():
209
for path, entry in self.iter_entries_by_dir():
210
if entry.kind == 'tree-reference':
211
yield path, entry.file_id
205
for path, entry in self.iter_entries_by_dir():
206
if entry.kind == 'tree-reference':
207
yield path, entry.file_id
213
209
def kind(self, file_id):
214
210
raise NotImplementedError("Tree subclass %s must implement kind"
259
251
def _file_size(self, entry, stat_value):
260
252
raise NotImplementedError(self._file_size)
254
def _get_inventory(self):
255
return self._inventory
262
257
def get_file(self, file_id, path=None):
263
258
"""Return a file object for the file file_id in the tree.
265
260
If both file_id and path are defined, it is implementation defined as
266
261
to which one is used.
268
263
raise NotImplementedError(self.get_file)
270
def get_file_with_stat(self, file_id, path=None):
271
"""Get a file handle and stat object for file_id.
273
The default implementation returns (self.get_file, None) for backwards
276
:param file_id: The file id to read.
277
:param path: The path of the file, if it is known.
278
:return: A tuple (file_handle, stat_value_or_None). If the tree has
279
no stat facility, or need for a stat cache feedback during commit,
280
it may return None for the second element of the tuple.
282
return (self.get_file(file_id, path), None)
284
def get_file_text(self, file_id, path=None):
285
"""Return the byte content of a file.
287
:param file_id: The file_id of the file.
288
:param path: The path of the file.
290
If both file_id and path are supplied, an implementation may use
293
:returns: A single byte string for the whole file.
295
my_file = self.get_file(file_id, path)
297
return my_file.read()
301
def get_file_lines(self, file_id, path=None):
302
"""Return the content of a file, as lines.
304
:param file_id: The file_id of the file.
305
:param path: The path of the file.
307
If both file_id and path are supplied, an implementation may use
310
return osutils.split_lines(self.get_file_text(file_id, path))
312
def get_file_verifier(self, file_id, path=None, stat_value=None):
313
"""Return a verifier for a file.
315
The default implementation returns a sha1.
317
:param file_id: The handle for this file.
318
:param path: The path that this file can be found at.
319
These must point to the same object.
320
:param stat_value: Optional stat value for the object
321
:return: Tuple with verifier name and verifier data
323
return ("SHA1", self.get_file_sha1(file_id, path=path,
324
stat_value=stat_value))
326
def get_file_sha1(self, file_id, path=None, stat_value=None):
327
"""Return the SHA1 file for a file.
329
:note: callers should use get_file_verifier instead
330
where possible, as the underlying repository implementation may
331
have quicker access to a non-sha1 verifier.
333
:param file_id: The handle for this file.
334
:param path: The path that this file can be found at.
335
These must point to the same object.
336
:param stat_value: Optional stat value for the object
338
raise NotImplementedError(self.get_file_sha1)
340
265
def get_file_mtime(self, file_id, path=None):
341
266
"""Return the modification time for a file.
469
382
except errors.NoSuchRevisionInTree:
470
383
yield self.repository.revision_tree(revision_id)
386
def _file_revision(revision_tree, file_id):
387
"""Determine the revision associated with a file in a given tree."""
388
revision_tree.lock_read()
390
return revision_tree.inventory[file_id].revision
392
revision_tree.unlock()
472
394
def _get_file_revision(self, file_id, vf, tree_revision):
473
395
"""Ensure that file_id, tree_revision is in vf to plan the merge."""
475
397
if getattr(self, '_repository', None) is None:
476
398
last_revision = tree_revision
477
parent_keys = [(file_id, t.get_file_revision(file_id)) for t in
399
parent_keys = [(file_id, self._file_revision(t, file_id)) for t in
478
400
self._iter_parent_trees()]
479
401
vf.add_lines((file_id, last_revision), parent_keys,
480
self.get_file_lines(file_id))
402
self.get_file(file_id).readlines())
481
403
repo = self.branch.repository
482
404
base_vf = repo.texts
484
last_revision = self.get_file_revision(file_id)
406
last_revision = self._file_revision(self, file_id)
485
407
base_vf = self._repository.texts
486
408
if base_vf not in vf.fallback_versionedfiles:
487
409
vf.fallback_versionedfiles.append(base_vf)
488
410
return last_revision
412
inventory = property(_get_inventory,
413
doc="Inventory of this Tree")
490
415
def _check_retrieved(self, ie, f):
491
416
if not __debug__:
493
fp = osutils.fingerprint_file(f)
418
fp = fingerprint_file(f)
496
421
if ie.text_size is not None:
497
422
if ie.text_size != fp['size']:
498
raise errors.BzrError(
499
"mismatched size for file %r in %r" %
500
(ie.file_id, self._store),
423
raise BzrError("mismatched size for file %r in %r" % (ie.file_id, self._store),
501
424
["inventory expects %d bytes" % ie.text_size,
502
425
"file is actually %d bytes" % fp['size'],
503
426
"store is probably damaged/corrupt"])
505
428
if ie.text_sha1 != fp['sha1']:
506
raise errors.BzrError("wrong SHA-1 for file %r in %r" %
507
(ie.file_id, self._store),
429
raise BzrError("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
508
430
["inventory expects %s" % ie.text_sha1,
509
431
"file is actually %s" % fp['sha1'],
510
432
"store is probably damaged/corrupt"])
512
435
def path2id(self, path):
513
436
"""Return the id for path in this tree."""
514
raise NotImplementedError(self.path2id)
437
return self._inventory.path2id(path)
516
439
def paths2ids(self, paths, trees=[], require_versioned=True):
517
440
"""Return all the ids that can be reached by walking from paths.
519
442
Each path is looked up in this tree and any extras provided in
520
443
trees, and this is repeated recursively: the children in an extra tree
521
444
of a directory that has been renamed under a provided path in this tree
610
534
raise NotImplementedError(self.walkdirs)
612
def supports_content_filtering(self):
615
def _content_filter_stack(self, path=None, file_id=None):
616
"""The stack of content filters for a path if filtering is supported.
618
Readers will be applied in first-to-last order.
619
Writers will be applied in last-to-first order.
620
Either the path or the file-id needs to be provided.
622
:param path: path relative to the root of the tree
624
:param file_id: file_id or None if unknown
625
:return: the list of filters - [] if there are none
627
filter_pref_names = filters._get_registered_names()
628
if len(filter_pref_names) == 0:
631
path = self.id2path(file_id)
632
prefs = self.iter_search_rules([path], filter_pref_names).next()
633
stk = filters._get_filter_stack_for(prefs)
634
if 'filters' in debug.debug_flags:
635
trace.note(gettext("*** {0} content-filter: {1} => {2!r}").format(path,prefs,stk))
638
def _content_filter_stack_provider(self):
639
"""A function that returns a stack of ContentFilters.
641
The function takes a path (relative to the top of the tree) and a
642
file-id as parameters.
644
:return: None if content filtering is not supported by this tree.
646
if self.supports_content_filtering():
647
return lambda path, file_id: \
648
self._content_filter_stack(path, file_id)
652
536
def iter_search_rules(self, path_names, pref_names=None,
653
_default_searcher=None):
537
_default_searcher=rules._per_user_searcher):
654
538
"""Find the preferences for filenames in a tree.
656
540
:param path_names: an iterable of paths to find attributes for.
671
553
for path in path_names:
672
554
yield searcher.get_items(path)
674
557
def _get_rules_searcher(self, default_searcher):
675
558
"""Get the RulesSearcher for this tree given the default one."""
676
559
searcher = default_searcher
680
class InventoryTree(Tree):
681
"""A tree that relies on an inventory for its metadata.
683
Trees contain an `Inventory` object, and also know how to retrieve
684
file texts mentioned in the inventory, either from a working
685
directory or from a store.
687
It is possible for trees to contain files that are not described
688
in their inventory or vice versa; for this use `filenames()`.
690
Subclasses should set the _inventory attribute, which is considered
691
private to external API users.
563
class EmptyTree(Tree):
566
self._inventory = Inventory(root_id=None)
567
symbol_versioning.warn('EmptyTree is deprecated as of bzr 0.9 please'
568
' use repository.revision_tree instead.',
569
DeprecationWarning, stacklevel=2)
571
def get_parent_ids(self):
574
def get_symlink_target(self, file_id):
577
def has_filename(self, filename):
580
def kind(self, file_id):
583
def list_files(self, include_root=False):
586
def __contains__(self, file_id):
587
return (file_id in self._inventory)
589
def get_file_sha1(self, file_id, path=None, stat_value=None):
593
######################################################################
596
# TODO: Merge these two functions into a single one that can operate
597
# on either a whole tree or a set of files.
599
# TODO: Return the diff in order by filename, not by category or in
600
# random order. Can probably be done by lock-stepping through the
601
# filenames from both trees.
604
def file_status(filename, old_tree, new_tree):
605
"""Return single-letter status, old and new names for a file.
607
The complexity here is in deciding how to represent renames;
608
many complex cases are possible.
694
def get_canonical_inventory_paths(self, paths):
695
"""Like get_canonical_inventory_path() but works on multiple items.
697
:param paths: A sequence of paths relative to the root of the tree.
698
:return: A list of paths, with each item the corresponding input path
699
adjusted to account for existing elements that match case
702
return list(self._yield_canonical_inventory_paths(paths))
704
def get_canonical_inventory_path(self, path):
705
"""Returns the first inventory item that case-insensitively matches path.
707
If a path matches exactly, it is returned. If no path matches exactly
708
but more than one path matches case-insensitively, it is implementation
709
defined which is returned.
711
If no path matches case-insensitively, the input path is returned, but
712
with as many path entries that do exist changed to their canonical
715
If you need to resolve many names from the same tree, you should
716
use get_canonical_inventory_paths() to avoid O(N) behaviour.
718
:param path: A paths relative to the root of the tree.
719
:return: The input path adjusted to account for existing elements
720
that match case insensitively.
722
return self._yield_canonical_inventory_paths([path]).next()
724
def _yield_canonical_inventory_paths(self, paths):
726
# First, if the path as specified exists exactly, just use it.
727
if self.path2id(path) is not None:
731
cur_id = self.get_root_id()
733
bit_iter = iter(path.split("/"))
737
for child in self.iter_children(cur_id):
739
# XXX: it seem like if the child is known to be in the
740
# tree, we shouldn't need to go from its id back to
741
# its path -- mbp 2010-02-11
743
# XXX: it seems like we could be more efficient
744
# by just directly looking up the original name and
745
# only then searching all children; also by not
746
# chopping paths so much. -- mbp 2010-02-11
747
child_base = os.path.basename(self.id2path(child))
748
if (child_base == elt):
749
# if we found an exact match, we can stop now; if
750
# we found an approximate match we need to keep
751
# searching because there might be an exact match
754
new_path = osutils.pathjoin(cur_path, child_base)
756
elif child_base.lower() == lelt:
758
new_path = osutils.pathjoin(cur_path, child_base)
759
except errors.NoSuchId:
760
# before a change is committed we can see this error...
765
# got to the end of this directory and no entries matched.
766
# Return what matched so far, plus the rest as specified.
767
cur_path = osutils.pathjoin(cur_path, elt, *list(bit_iter))
772
def _get_inventory(self):
773
return self._inventory
775
inventory = property(_get_inventory,
776
doc="Inventory of this Tree")
779
def path2id(self, path):
780
"""Return the id for path in this tree."""
781
return self._inventory.path2id(path)
783
def id2path(self, file_id):
784
"""Return the path for a file id.
788
return self.inventory.id2path(file_id)
790
def has_id(self, file_id):
791
return self.inventory.has_id(file_id)
793
def has_or_had_id(self, file_id):
794
return self.inventory.has_id(file_id)
796
def all_file_ids(self):
797
return set(self.inventory)
799
@deprecated_method(deprecated_in((2, 4, 0)))
801
return iter(self.inventory)
803
def filter_unversioned_files(self, paths):
804
"""Filter out paths that are versioned.
806
:return: set of paths.
808
# NB: we specifically *don't* call self.has_filename, because for
809
# WorkingTrees that can indicate files that exist on disk but that
811
pred = self.inventory.has_filename
812
return set((p for p in paths if not pred(p)))
815
def iter_entries_by_dir(self, specific_file_ids=None, yield_parents=False):
816
"""Walk the tree in 'by_dir' order.
818
This will yield each entry in the tree as a (path, entry) tuple.
819
The order that they are yielded is:
821
See Tree.iter_entries_by_dir for details.
823
:param yield_parents: If True, yield the parents from the root leading
824
down to specific_file_ids that have been requested. This has no
825
impact if specific_file_ids is None.
827
return self.inventory.iter_entries_by_dir(
828
specific_file_ids=specific_file_ids, yield_parents=yield_parents)
830
def get_file_by_path(self, path):
831
return self.get_file(self._inventory.path2id(path), path)
610
old_inv = old_tree.inventory
611
new_inv = new_tree.inventory
612
new_id = new_inv.path2id(filename)
613
old_id = old_inv.path2id(filename)
615
if not new_id and not old_id:
616
# easy: doesn't exist in either; not versioned at all
617
if new_tree.is_ignored(filename):
618
return 'I', None, None
620
return '?', None, None
622
# There is now a file of this name, great.
625
# There is no longer a file of this name, but we can describe
626
# what happened to the file that used to have
627
# this name. There are two possibilities: either it was
628
# deleted entirely, or renamed.
629
if new_inv.has_id(old_id):
630
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
632
return 'D', old_inv.id2path(old_id), None
634
# if the file_id is new in this revision, it is added
635
if new_id and not old_inv.has_id(new_id):
638
# if there used to be a file of this name, but that ID has now
639
# disappeared, it is deleted
640
if old_id and not new_inv.has_id(old_id):
647
def find_renames(old_inv, new_inv):
648
for file_id in old_inv:
649
if file_id not in new_inv:
651
old_name = old_inv.id2path(file_id)
652
new_name = new_inv.id2path(file_id)
653
if old_name != new_name:
654
yield (old_name, new_name)
834
657
def find_ids_across_trees(filenames, trees, require_versioned=True):
835
658
"""Find the ids corresponding to specified filenames.
837
660
All matches in all trees will be used, and all children of matched
838
661
directories will be used.
915
739
will pass through to InterTree as appropriate.
918
# Formats that will be used to test this InterTree. If both are
919
# None, this InterTree will not be tested (e.g. because a complex
921
_matching_from_tree_format = None
922
_matching_to_tree_format = None
927
def is_compatible(kls, source, target):
928
# The default implementation is naive and uses the public API, so
929
# it works for all trees.
932
def _changes_from_entries(self, source_entry, target_entry,
933
source_path=None, target_path=None):
934
"""Generate a iter_changes tuple between source_entry and target_entry.
936
:param source_entry: An inventory entry from self.source, or None.
937
:param target_entry: An inventory entry from self.target, or None.
938
:param source_path: The path of source_entry, if known. If not known
939
it will be looked up.
940
:param target_path: The path of target_entry, if known. If not known
941
it will be looked up.
942
:return: A tuple, item 0 of which is an iter_changes result tuple, and
943
item 1 is True if there are any changes in the result tuple.
945
if source_entry is None:
946
if target_entry is None:
948
file_id = target_entry.file_id
950
file_id = source_entry.file_id
951
if source_entry is not None:
952
source_versioned = True
953
source_name = source_entry.name
954
source_parent = source_entry.parent_id
955
if source_path is None:
956
source_path = self.source.id2path(file_id)
957
source_kind, source_executable, source_stat = \
958
self.source._comparison_data(source_entry, source_path)
960
source_versioned = False
964
source_executable = None
965
if target_entry is not None:
966
target_versioned = True
967
target_name = target_entry.name
968
target_parent = target_entry.parent_id
969
if target_path is None:
970
target_path = self.target.id2path(file_id)
971
target_kind, target_executable, target_stat = \
972
self.target._comparison_data(target_entry, target_path)
974
target_versioned = False
978
target_executable = None
979
versioned = (source_versioned, target_versioned)
980
kind = (source_kind, target_kind)
981
changed_content = False
982
if source_kind != target_kind:
983
changed_content = True
984
elif source_kind == 'file':
985
if not self.file_content_matches(file_id, file_id, source_path,
986
target_path, source_stat, target_stat):
987
changed_content = True
988
elif source_kind == 'symlink':
989
if (self.source.get_symlink_target(file_id) !=
990
self.target.get_symlink_target(file_id)):
991
changed_content = True
992
# XXX: Yes, the indentation below is wrong. But fixing it broke
993
# test_merge.TestMergerEntriesLCAOnDisk.
994
# test_nested_tree_subtree_renamed_and_modified. We'll wait for
995
# the fix from bzr.dev -- vila 2009026
996
elif source_kind == 'tree-reference':
997
if (self.source.get_reference_revision(file_id, source_path)
998
!= self.target.get_reference_revision(file_id, target_path)):
999
changed_content = True
1000
parent = (source_parent, target_parent)
1001
name = (source_name, target_name)
1002
executable = (source_executable, target_executable)
1003
if (changed_content is not False or versioned[0] != versioned[1]
1004
or parent[0] != parent[1] or name[0] != name[1] or
1005
executable[0] != executable[1]):
1009
return (file_id, (source_path, target_path), changed_content,
1010
versioned, parent, name, kind, executable), changes
1012
744
@needs_read_lock
1013
745
def compare(self, want_unchanged=False, specific_files=None,
1014
746
extra_trees=None, require_versioned=False, include_root=False,
1075
809
:param require_versioned: Raise errors.PathsNotVersionedError if a
1076
810
path in the specific_files list is not versioned in one of
1077
811
source, target or extra_trees.
1078
:param specific_files: An optional list of file paths to restrict the
1079
comparison to. When mapping filenames to ids, all matches in all
1080
trees (including optional extra_trees) are used, and all children
1081
of matched directories are included. The parents in the target tree
1082
of the specific files up to and including the root of the tree are
1083
always evaluated for changes too.
1084
812
:param want_unversioned: Should unversioned files be returned in the
1085
813
output. An unversioned file is defined as one with (False, False)
1086
814
for the versioned pair.
1088
817
lookup_trees = [self.source]
1090
819
lookup_trees.extend(extra_trees)
1091
# The ids of items we need to examine to insure delta consistency.
1092
precise_file_ids = set()
1093
changed_file_ids = []
1094
820
if specific_files == []:
1095
821
specific_file_ids = []
1097
823
specific_file_ids = self.target.paths2ids(specific_files,
1098
824
lookup_trees, require_versioned=require_versioned)
1099
if specific_files is not None:
1100
# reparented or added entries must have their parents included
1101
# so that valid deltas can be created. The seen_parents set
1102
# tracks the parents that we need to have.
1103
# The seen_dirs set tracks directory entries we've yielded.
1104
# After outputting version object in to_entries we set difference
1105
# the two seen sets and start checking parents.
1106
seen_parents = set()
1108
825
if want_unversioned:
1109
826
all_unversioned = sorted([(p.split('/'), p) for p in
1110
827
self.target.extras()
1111
828
if specific_files is None or
1112
829
osutils.is_inside_any(specific_files, p)])
1113
all_unversioned = collections.deque(all_unversioned)
830
all_unversioned = deque(all_unversioned)
1115
all_unversioned = collections.deque()
832
all_unversioned = deque()
1117
from_entries_by_dir = list(self.source.iter_entries_by_dir(
834
from_entries_by_dir = list(self.source.inventory.iter_entries_by_dir(
1118
835
specific_file_ids=specific_file_ids))
1119
836
from_data = dict((e.file_id, (p, e)) for p, e in from_entries_by_dir)
1120
to_entries_by_dir = list(self.target.iter_entries_by_dir(
837
to_entries_by_dir = list(self.target.inventory.iter_entries_by_dir(
1121
838
specific_file_ids=specific_file_ids))
1122
839
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
1124
# the unversioned path lookup only occurs on real trees - where there
841
# the unversioned path lookup only occurs on real trees - where there
1125
842
# can be extras. So the fake_entry is solely used to look up
1126
843
# executable it values when execute is not supported.
1127
fake_entry = inventory.InventoryFile('unused', 'unused', 'unused')
1128
for target_path, target_entry in to_entries_by_dir:
1129
while (all_unversioned and
1130
all_unversioned[0][0] < target_path.split('/')):
844
fake_entry = InventoryFile('unused', 'unused', 'unused')
845
for to_path, to_entry in to_entries_by_dir:
846
while all_unversioned and all_unversioned[0][0] < to_path.split('/'):
1131
847
unversioned_path = all_unversioned.popleft()
1132
target_kind, target_executable, target_stat = \
848
to_kind, to_executable, to_stat = \
1133
849
self.target._comparison_data(fake_entry, unversioned_path[1])
1134
850
yield (None, (None, unversioned_path[1]), True, (False, False),
1136
852
(None, unversioned_path[0][-1]),
1137
(None, target_kind),
1138
(None, target_executable))
1139
source_path, source_entry = from_data.get(target_entry.file_id,
1141
result, changes = self._changes_from_entries(source_entry,
1142
target_entry, source_path=source_path, target_path=target_path)
1143
to_paths[result[0]] = result[1][1]
854
(None, to_executable))
855
file_id = to_entry.file_id
856
to_paths[file_id] = to_path
1144
857
entry_count += 1
858
changed_content = False
859
from_path, from_entry = from_data.get(file_id, (None, None))
860
from_versioned = (from_entry is not None)
861
if from_entry is not None:
862
from_versioned = True
863
from_name = from_entry.name
864
from_parent = from_entry.parent_id
865
from_kind, from_executable, from_stat = \
866
self.source._comparison_data(from_entry, from_path)
1146
867
entry_count += 1
869
from_versioned = False
873
from_executable = None
874
versioned = (from_versioned, True)
875
to_kind, to_executable, to_stat = \
876
self.target._comparison_data(to_entry, to_path)
877
kind = (from_kind, to_kind)
878
if kind[0] != kind[1]:
879
changed_content = True
880
elif from_kind == 'file':
881
from_size = self.source._file_size(from_entry, from_stat)
882
to_size = self.target._file_size(to_entry, to_stat)
883
if from_size != to_size:
884
changed_content = True
885
elif (self.source.get_file_sha1(file_id, from_path, from_stat) !=
886
self.target.get_file_sha1(file_id, to_path, to_stat)):
887
changed_content = True
888
elif from_kind == 'symlink':
889
if (self.source.get_symlink_target(file_id) !=
890
self.target.get_symlink_target(file_id)):
891
changed_content = True
892
elif from_kind == 'tree-reference':
893
if (self.source.get_reference_revision(file_id, from_path)
894
!= self.target.get_reference_revision(file_id, to_path)):
895
changed_content = True
896
parent = (from_parent, to_entry.parent_id)
897
name = (from_name, to_entry.name)
898
executable = (from_executable, to_executable)
1147
899
if pb is not None:
1148
900
pb.update('comparing files', entry_count, num_entries)
1149
if changes or include_unchanged:
1150
if specific_file_ids is not None:
1151
new_parent_id = result[4][1]
1152
precise_file_ids.add(new_parent_id)
1153
changed_file_ids.append(result[0])
1155
# Ensure correct behaviour for reparented/added specific files.
1156
if specific_files is not None:
1157
# Record output dirs
1158
if result[6][1] == 'directory':
1159
seen_dirs.add(result[0])
1160
# Record parents of reparented/added entries.
1161
versioned = result[3]
1163
if not versioned[0] or parents[0] != parents[1]:
1164
seen_parents.add(parents[1])
901
if (changed_content is not False or versioned[0] != versioned[1]
902
or parent[0] != parent[1] or name[0] != name[1] or
903
executable[0] != executable[1] or include_unchanged):
904
yield (file_id, (from_path, to_path), changed_content,
905
versioned, parent, name, kind, executable)
1165
907
while all_unversioned:
1166
908
# yield any trailing unversioned paths
1167
909
unversioned_path = all_unversioned.popleft()
1196
950
executable = (from_executable, None)
1197
951
changed_content = from_kind is not None
1198
952
# the parent's path is necessarily known at this point.
1199
changed_file_ids.append(file_id)
1200
953
yield(file_id, (path, to_path), changed_content, versioned, parent,
1201
954
name, kind, executable)
1202
changed_file_ids = set(changed_file_ids)
1203
if specific_file_ids is not None:
1204
for result in self._handle_precise_ids(precise_file_ids,
1208
def _get_entry(self, tree, file_id):
1209
"""Get an inventory entry from a tree, with missing entries as None.
1211
If the tree raises NotImplementedError on accessing .inventory, then
1212
this is worked around using iter_entries_by_dir on just the file id
1215
:param tree: The tree to lookup the entry in.
1216
:param file_id: The file_id to lookup.
1219
inventory = tree.inventory
1220
except NotImplementedError:
1221
# No inventory available.
1223
iterator = tree.iter_entries_by_dir(specific_file_ids=[file_id])
1224
return iterator.next()[1]
1225
except StopIteration:
1229
return inventory[file_id]
1230
except errors.NoSuchId:
1233
def _handle_precise_ids(self, precise_file_ids, changed_file_ids,
1234
discarded_changes=None):
1235
"""Fill out a partial iter_changes to be consistent.
1237
:param precise_file_ids: The file ids of parents that were seen during
1239
:param changed_file_ids: The file ids of already emitted items.
1240
:param discarded_changes: An optional dict of precalculated
1241
iter_changes items which the partial iter_changes had not output
1243
:return: A generator of iter_changes items to output.
1245
# process parents of things that had changed under the users
1246
# requested paths to prevent incorrect paths or parent ids which
1247
# aren't in the tree.
1248
while precise_file_ids:
1249
precise_file_ids.discard(None)
1250
# Don't emit file_ids twice
1251
precise_file_ids.difference_update(changed_file_ids)
1252
if not precise_file_ids:
1254
# If the there was something at a given output path in source, we
1255
# have to include the entry from source in the delta, or we would
1256
# be putting this entry into a used path.
1258
for parent_id in precise_file_ids:
1260
paths.append(self.target.id2path(parent_id))
1261
except errors.NoSuchId:
1262
# This id has been dragged in from the source by delta
1263
# expansion and isn't present in target at all: we don't
1264
# need to check for path collisions on it.
1267
old_id = self.source.path2id(path)
1268
precise_file_ids.add(old_id)
1269
precise_file_ids.discard(None)
1270
current_ids = precise_file_ids
1271
precise_file_ids = set()
1272
# We have to emit all of precise_file_ids that have been altered.
1273
# We may have to output the children of some of those ids if any
1274
# directories have stopped being directories.
1275
for file_id in current_ids:
1277
if discarded_changes:
1278
result = discarded_changes.get(file_id)
1283
old_entry = self._get_entry(self.source, file_id)
1284
new_entry = self._get_entry(self.target, file_id)
1285
result, changes = self._changes_from_entries(
1286
old_entry, new_entry)
1289
# Get this parents parent to examine.
1290
new_parent_id = result[4][1]
1291
precise_file_ids.add(new_parent_id)
1293
if (result[6][0] == 'directory' and
1294
result[6][1] != 'directory'):
1295
# This stopped being a directory, the old children have
1297
if old_entry is None:
1298
# Reusing a discarded change.
1299
old_entry = self._get_entry(self.source, file_id)
1300
for child in old_entry.children.values():
1301
precise_file_ids.add(child.file_id)
1302
changed_file_ids.add(result[0])
1306
def file_content_matches(self, source_file_id, target_file_id,
1307
source_path=None, target_path=None, source_stat=None, target_stat=None):
1308
"""Check if two files are the same in the source and target trees.
1310
This only checks that the contents of the files are the same,
1311
it does not touch anything else.
1313
:param source_file_id: File id of the file in the source tree
1314
:param target_file_id: File id of the file in the target tree
1315
:param source_path: Path of the file in the source tree
1316
:param target_path: Path of the file in the target tree
1317
:param source_stat: Optional stat value of the file in the source tree
1318
:param target_stat: Optional stat value of the file in the target tree
1319
:return: Boolean indicating whether the files have the same contents
1321
source_verifier_kind, source_verifier_data = self.source.get_file_verifier(
1322
source_file_id, source_path, source_stat)
1323
target_verifier_kind, target_verifier_data = self.target.get_file_verifier(
1324
target_file_id, target_path, target_stat)
1325
if source_verifier_kind == target_verifier_kind:
1326
return (source_verifier_data == target_verifier_data)
1327
# Fall back to SHA1 for now
1328
if source_verifier_kind != "SHA1":
1329
source_sha1 = self.source.get_file_sha1(source_file_id,
1330
source_path, source_stat)
1332
source_sha1 = source_verifier_data
1333
if target_verifier_kind != "SHA1":
1334
target_sha1 = self.target.get_file_sha1(target_file_id,
1335
target_path, target_stat)
1337
target_sha1 = target_verifier_data
1338
return (source_sha1 == target_sha1)
1340
InterTree.register_optimiser(InterTree)
1343
957
class MultiWalker(object):
added
removed
bzrlib/tree.py
