1
# Copyright (C) 2005 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., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
17
"""Tree classes, representing directory at point in time.
21
from collections import deque
22
from cStringIO import StringIO
26
conflicts as _mod_conflicts,
29
revision as _mod_revision,
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
37
from bzrlib.inter import InterObject
38
from bzrlib.osutils import fingerprint_file
39
import bzrlib.revision
40
from bzrlib.trace import mutter, note
44
"""Abstract file tree.
46
There are several subclasses:
48
* `WorkingTree` exists as files on disk editable by the user.
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()`.
59
Trees can be compared, etc, regardless of whether they are working
60
trees or versioned trees.
63
def changes_from(self, other, want_unchanged=False, specific_files=None,
64
extra_trees=None, require_versioned=False, include_root=False,
65
want_unversioned=False):
66
"""Return a TreeDelta of the changes from other to this tree.
68
:param other: A tree to compare with.
69
:param specific_files: An optional list of file paths to restrict the
70
comparison to. When mapping filenames to ids, all matches in all
71
trees (including optional extra_trees) are used, and all children of
72
matched directories are included.
73
:param want_unchanged: An optional boolean requesting the inclusion of
74
unchanged entries in the result.
75
:param extra_trees: An optional list of additional trees to use when
76
mapping the contents of specific_files (paths) to file_ids.
77
:param require_versioned: An optional boolean (defaults to False). When
78
supplied and True all the 'specific_files' must be versioned, or
79
a PathsNotVersionedError will be thrown.
80
:param want_unversioned: Scan for unversioned paths.
82
The comparison will be performed by an InterTree object looked up on
85
# Martin observes that Tree.changes_from returns a TreeDelta and this
86
# may confuse people, because the class name of the returned object is
87
# a synonym of the object referenced in the method name.
88
return InterTree.get(other, self).compare(
89
want_unchanged=want_unchanged,
90
specific_files=specific_files,
91
extra_trees=extra_trees,
92
require_versioned=require_versioned,
93
include_root=include_root,
94
want_unversioned=want_unversioned,
97
@symbol_versioning.deprecated_method(symbol_versioning.one_three)
98
def _iter_changes(self, *args, **kwargs):
99
return self.iter_changes(*args, **kwargs)
101
def iter_changes(self, from_tree, include_unchanged=False,
102
specific_files=None, pb=None, extra_trees=None,
103
require_versioned=True, want_unversioned=False):
104
intertree = InterTree.get(from_tree, self)
105
return intertree.iter_changes(include_unchanged, specific_files, pb,
106
extra_trees, require_versioned, want_unversioned=want_unversioned)
109
"""Get a list of the conflicts in the tree.
111
Each conflict is an instance of bzrlib.conflicts.Conflict.
113
return _mod_conflicts.ConflictList()
116
"""For trees that can have unversioned files, return all such paths."""
119
def get_parent_ids(self):
120
"""Get the parent ids for this tree.
122
:return: a list of parent ids. [] is returned to indicate
123
a tree with no parents.
124
:raises: BzrError if the parents are not known.
126
raise NotImplementedError(self.get_parent_ids)
128
def has_filename(self, filename):
129
"""True if the tree has given filename."""
130
raise NotImplementedError(self.has_filename)
132
def has_id(self, file_id):
133
return self.inventory.has_id(file_id)
135
__contains__ = has_id
137
def has_or_had_id(self, file_id):
138
if file_id == self.inventory.root.file_id:
140
return self.inventory.has_id(file_id)
142
def is_ignored(self, filename):
143
"""Check whether the filename is ignored by this tree.
145
:param filename: The relative filename within the tree.
146
:return: True if the filename is ignored.
151
return iter(self.inventory)
153
def all_file_ids(self):
154
"""Iterate through all file ids, including ids for missing files."""
155
return set(self.inventory)
157
def id2path(self, file_id):
158
"""Return the path for a file id.
162
return self.inventory.id2path(file_id)
164
def is_control_filename(self, filename):
165
"""True if filename is the name of a control file in this tree.
167
:param filename: A filename within the tree. This is a relative path
168
from the root of this tree.
170
This is true IF and ONLY IF the filename is part of the meta data
171
that bzr controls in this tree. I.E. a random .bzr directory placed
172
on disk will not be a control file for this tree.
174
return self.bzrdir.is_control_filename(filename)
177
def iter_entries_by_dir(self, specific_file_ids=None):
178
"""Walk the tree in 'by_dir' order.
180
This will yield each entry in the tree as a (path, entry) tuple. The
181
order that they are yielded is: the contents of a directory are
182
preceeded by the parent of a directory, and all the contents of a
183
directory are grouped together.
185
return self.inventory.iter_entries_by_dir(
186
specific_file_ids=specific_file_ids)
188
def iter_references(self):
189
for path, entry in self.iter_entries_by_dir():
190
if entry.kind == 'tree-reference':
191
yield path, entry.file_id
193
def kind(self, file_id):
194
raise NotImplementedError("Tree subclass %s must implement kind"
195
% self.__class__.__name__)
197
def stored_kind(self, file_id):
198
"""File kind stored for this file_id.
200
May not match kind on disk for working trees. Always available
201
for versioned files, even when the file itself is missing.
203
return self.kind(file_id)
205
def path_content_summary(self, path):
206
"""Get a summary of the information about path.
208
:param path: A relative path within the tree.
209
:return: A tuple containing kind, size, exec, sha1-or-link.
210
Kind is always present (see tree.kind()).
211
size is present if kind is file, None otherwise.
212
exec is None unless kind is file and the platform supports the 'x'
214
sha1-or-link is the link target if kind is symlink, or the sha1 if
215
it can be obtained without reading the file.
217
raise NotImplementedError(self.path_content_summary)
219
def get_reference_revision(self, file_id, path=None):
220
raise NotImplementedError("Tree subclass %s must implement "
221
"get_reference_revision"
222
% self.__class__.__name__)
224
def _comparison_data(self, entry, path):
225
"""Return a tuple of kind, executable, stat_value for a file.
227
entry may be None if there is no inventory entry for the file, but
228
path must always be supplied.
230
kind is None if there is no file present (even if an inventory id is
231
present). executable is False for non-file entries.
233
raise NotImplementedError(self._comparison_data)
235
def _file_size(self, entry, stat_value):
236
raise NotImplementedError(self._file_size)
238
def _get_inventory(self):
239
return self._inventory
241
def get_file(self, file_id, path=None):
242
"""Return a file object for the file file_id in the tree.
244
If both file_id and path are defined, it is implementation defined as
245
to which one is used.
247
raise NotImplementedError(self.get_file)
249
def get_file_mtime(self, file_id, path=None):
250
"""Return the modification time for a file.
252
:param file_id: The handle for this file.
253
:param path: The path that this file can be found at.
254
These must point to the same object.
256
raise NotImplementedError(self.get_file_mtime)
258
def get_file_size(self, file_id):
259
"""Return the size of a file in bytes.
261
This applies only to regular files. If invoked on directories or
262
symlinks, it will return None.
263
:param file_id: The file-id of the file
265
raise NotImplementedError(self.get_file_size)
267
def get_file_by_path(self, path):
268
return self.get_file(self._inventory.path2id(path), path)
270
def iter_files_bytes(self, desired_files):
271
"""Iterate through file contents.
273
Files will not necessarily be returned in the order they occur in
274
desired_files. No specific order is guaranteed.
276
Yields pairs of identifier, bytes_iterator. identifier is an opaque
277
value supplied by the caller as part of desired_files. It should
278
uniquely identify the file version in the caller's context. (Examples:
279
an index number or a TreeTransform trans_id.)
281
bytes_iterator is an iterable of bytestrings for the file. The
282
kind of iterable and length of the bytestrings are unspecified, but for
283
this implementation, it is a tuple containing a single bytestring with
284
the complete text of the file.
286
:param desired_files: a list of (file_id, identifier) pairs
288
for file_id, identifier in desired_files:
289
# We wrap the string in a tuple so that we can return an iterable
290
# of bytestrings. (Technically, a bytestring is also an iterable
291
# of bytestrings, but iterating through each character is not
293
cur_file = (self.get_file_text(file_id),)
294
yield identifier, cur_file
296
def get_symlink_target(self, file_id):
297
"""Get the target for a given file_id.
299
It is assumed that the caller already knows that file_id is referencing
301
:param file_id: Handle for the symlink entry.
302
:return: The path the symlink points to.
304
raise NotImplementedError(self.get_symlink_target)
306
def get_root_id(self):
307
"""Return the file_id for the root of this tree."""
308
raise NotImplementedError(self.get_root_id)
310
def annotate_iter(self, file_id,
311
default_revision=_mod_revision.CURRENT_REVISION):
312
"""Return an iterator of revision_id, line tuples.
314
For working trees (and mutable trees in general), the special
315
revision_id 'current:' will be used for lines that are new in this
316
tree, e.g. uncommitted changes.
317
:param file_id: The file to produce an annotated version from
318
:param default_revision: For lines that don't match a basis, mark them
319
with this revision id. Not all implementations will make use of
322
raise NotImplementedError(self.annotate_iter)
324
def _get_plan_merge_data(self, file_id, other, base):
325
from bzrlib import merge, versionedfile
326
vf = versionedfile._PlanMergeVersionedFile(file_id)
327
last_revision_a = self._get_file_revision(file_id, vf, 'this:')
328
last_revision_b = other._get_file_revision(file_id, vf, 'other:')
330
last_revision_base = None
332
last_revision_base = base._get_file_revision(file_id, vf, 'base:')
333
return vf, last_revision_a, last_revision_b, last_revision_base
335
def plan_file_merge(self, file_id, other, base=None):
336
"""Generate a merge plan based on annotations.
338
If the file contains uncommitted changes in this tree, they will be
339
attributed to the 'current:' pseudo-revision. If the file contains
340
uncommitted changes in the other tree, they will be assigned to the
341
'other:' pseudo-revision.
343
data = self._get_plan_merge_data(file_id, other, base)
344
vf, last_revision_a, last_revision_b, last_revision_base = data
345
return vf.plan_merge(last_revision_a, last_revision_b,
348
def plan_file_lca_merge(self, file_id, other, base=None):
349
"""Generate a merge plan based lca-newness.
351
If the file contains uncommitted changes in this tree, they will be
352
attributed to the 'current:' pseudo-revision. If the file contains
353
uncommitted changes in the other tree, they will be assigned to the
354
'other:' pseudo-revision.
356
data = self._get_plan_merge_data(file_id, other, base)
357
vf, last_revision_a, last_revision_b, last_revision_base = data
358
return vf.plan_lca_merge(last_revision_a, last_revision_b,
361
def _get_file_revision(self, file_id, vf, tree_revision):
362
"""Ensure that file_id, tree_revision is in vf to plan the merge."""
363
def file_revision(revision_tree):
364
revision_tree.lock_read()
366
return revision_tree.inventory[file_id].revision
368
revision_tree.unlock()
370
def iter_parent_trees():
371
for revision_id in self.get_parent_ids():
373
yield self.revision_tree(revision_id)
375
yield self.repository.revision_tree(revision_id)
377
if getattr(self, '_repository', None) is None:
378
last_revision = tree_revision
379
parent_keys = [(file_id, file_revision(t)) for t in
381
vf.add_lines((file_id, last_revision), parent_keys,
382
self.get_file(file_id).readlines())
383
repo = self.branch.repository
386
last_revision = file_revision(self)
387
base_vf = self._repository.texts
388
if base_vf not in vf.fallback_versionedfiles:
389
vf.fallback_versionedfiles.append(base_vf)
392
inventory = property(_get_inventory,
393
doc="Inventory of this Tree")
395
def _check_retrieved(self, ie, f):
398
fp = fingerprint_file(f)
401
if ie.text_size is not None:
402
if ie.text_size != fp['size']:
403
raise BzrError("mismatched size for file %r in %r" % (ie.file_id, self._store),
404
["inventory expects %d bytes" % ie.text_size,
405
"file is actually %d bytes" % fp['size'],
406
"store is probably damaged/corrupt"])
408
if ie.text_sha1 != fp['sha1']:
409
raise BzrError("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
410
["inventory expects %s" % ie.text_sha1,
411
"file is actually %s" % fp['sha1'],
412
"store is probably damaged/corrupt"])
415
def path2id(self, path):
416
"""Return the id for path in this tree."""
417
return self._inventory.path2id(path)
419
def paths2ids(self, paths, trees=[], require_versioned=True):
420
"""Return all the ids that can be reached by walking from paths.
422
Each path is looked up in this tree and any extras provided in
423
trees, and this is repeated recursively: the children in an extra tree
424
of a directory that has been renamed under a provided path in this tree
425
are all returned, even if none exist under a provided path in this
426
tree, and vice versa.
428
:param paths: An iterable of paths to start converting to ids from.
429
Alternatively, if paths is None, no ids should be calculated and None
430
will be returned. This is offered to make calling the api unconditional
431
for code that *might* take a list of files.
432
:param trees: Additional trees to consider.
433
:param require_versioned: If False, do not raise NotVersionedError if
434
an element of paths is not versioned in this tree and all of trees.
436
return find_ids_across_trees(paths, [self] + list(trees), require_versioned)
438
@symbol_versioning.deprecated_method(symbol_versioning.one_six)
439
def print_file(self, file_id):
440
"""Print file with id `file_id` to stdout."""
442
sys.stdout.write(self.get_file_text(file_id))
447
def revision_tree(self, revision_id):
448
"""Obtain a revision tree for the revision revision_id.
450
The intention of this method is to allow access to possibly cached
451
tree data. Implementors of this method should raise NoSuchRevision if
452
the tree is not locally available, even if they could obtain the
453
tree via a repository or some other means. Callers are responsible
454
for finding the ultimate source for a revision tree.
456
:param revision_id: The revision_id of the requested tree.
458
:raises: NoSuchRevision if the tree cannot be obtained.
460
raise errors.NoSuchRevisionInTree(self, revision_id)
463
"""What files are present in this tree and unknown.
465
:return: an iterator over the unknown files.
472
def filter_unversioned_files(self, paths):
473
"""Filter out paths that are versioned.
475
:return: set of paths.
477
# NB: we specifically *don't* call self.has_filename, because for
478
# WorkingTrees that can indicate files that exist on disk but that
480
pred = self.inventory.has_filename
481
return set((p for p in paths if not pred(p)))
483
def walkdirs(self, prefix=""):
484
"""Walk the contents of this tree from path down.
486
This yields all the data about the contents of a directory at a time.
487
After each directory has been yielded, if the caller has mutated the
488
list to exclude some directories, they are then not descended into.
490
The data yielded is of the form:
491
((directory-relpath, directory-path-from-root, directory-fileid),
492
[(relpath, basename, kind, lstat, path_from_tree_root, file_id,
493
versioned_kind), ...]),
494
- directory-relpath is the containing dirs relpath from prefix
495
- directory-path-from-root is the containing dirs path from /
496
- directory-fileid is the id of the directory if it is versioned.
497
- relpath is the relative path within the subtree being walked.
498
- basename is the basename
499
- kind is the kind of the file now. If unknonwn then the file is not
500
present within the tree - but it may be recorded as versioned. See
502
- lstat is the stat data *if* the file was statted.
503
- path_from_tree_root is the path from the root of the tree.
504
- file_id is the file_id if the entry is versioned.
505
- versioned_kind is the kind of the file as last recorded in the
506
versioning system. If 'unknown' the file is not versioned.
507
One of 'kind' and 'versioned_kind' must not be 'unknown'.
509
:param prefix: Start walking from prefix within the tree rather than
510
at the root. This allows one to walk a subtree but get paths that are
511
relative to a tree rooted higher up.
512
:return: an iterator over the directory data.
514
raise NotImplementedError(self.walkdirs)
516
def iter_search_rules(self, path_names, pref_names=None,
517
_default_searcher=rules._per_user_searcher):
518
"""Find the preferences for filenames in a tree.
520
:param path_names: an iterable of paths to find attributes for.
521
Paths are given relative to the root of the tree.
522
:param pref_names: the list of preferences to lookup - None for all
523
:param _default_searcher: private parameter to assist testing - don't use
524
:return: an iterator of tuple sequences, one per path-name.
525
See _RulesSearcher.get_items for details on the tuple sequence.
527
searcher = self._get_rules_searcher(_default_searcher)
528
if searcher is not None:
529
if pref_names is not None:
530
for path in path_names:
531
yield searcher.get_selected_items(path, pref_names)
533
for path in path_names:
534
yield searcher.get_items(path)
537
def _get_rules_searcher(self, default_searcher):
538
"""Get the RulesSearcher for this tree given the default one."""
539
searcher = default_searcher
540
file_id = self.path2id(rules.RULES_TREE_FILENAME)
541
if file_id is not None:
542
ini_file = self.get_file(file_id)
543
searcher = rules._StackedRulesSearcher(
544
[rules._IniBasedRulesSearcher(ini_file), default_searcher])
548
class EmptyTree(Tree):
551
self._inventory = Inventory(root_id=None)
552
symbol_versioning.warn('EmptyTree is deprecated as of bzr 0.9 please'
553
' use repository.revision_tree instead.',
554
DeprecationWarning, stacklevel=2)
556
def get_parent_ids(self):
559
def get_symlink_target(self, file_id):
562
def has_filename(self, filename):
565
def kind(self, file_id):
568
def list_files(self, include_root=False):
571
def __contains__(self, file_id):
572
return (file_id in self._inventory)
574
def get_file_sha1(self, file_id, path=None, stat_value=None):
578
######################################################################
581
# TODO: Merge these two functions into a single one that can operate
582
# on either a whole tree or a set of files.
584
# TODO: Return the diff in order by filename, not by category or in
585
# random order. Can probably be done by lock-stepping through the
586
# filenames from both trees.
589
def file_status(filename, old_tree, new_tree):
590
"""Return single-letter status, old and new names for a file.
592
The complexity here is in deciding how to represent renames;
593
many complex cases are possible.
595
old_inv = old_tree.inventory
596
new_inv = new_tree.inventory
597
new_id = new_inv.path2id(filename)
598
old_id = old_inv.path2id(filename)
600
if not new_id and not old_id:
601
# easy: doesn't exist in either; not versioned at all
602
if new_tree.is_ignored(filename):
603
return 'I', None, None
605
return '?', None, None
607
# There is now a file of this name, great.
610
# There is no longer a file of this name, but we can describe
611
# what happened to the file that used to have
612
# this name. There are two possibilities: either it was
613
# deleted entirely, or renamed.
614
if new_inv.has_id(old_id):
615
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
617
return 'D', old_inv.id2path(old_id), None
619
# if the file_id is new in this revision, it is added
620
if new_id and not old_inv.has_id(new_id):
623
# if there used to be a file of this name, but that ID has now
624
# disappeared, it is deleted
625
if old_id and not new_inv.has_id(old_id):
632
def find_renames(old_inv, new_inv):
633
for file_id in old_inv:
634
if file_id not in new_inv:
636
old_name = old_inv.id2path(file_id)
637
new_name = new_inv.id2path(file_id)
638
if old_name != new_name:
639
yield (old_name, new_name)
642
def find_ids_across_trees(filenames, trees, require_versioned=True):
643
"""Find the ids corresponding to specified filenames.
645
All matches in all trees will be used, and all children of matched
646
directories will be used.
648
:param filenames: The filenames to find file_ids for (if None, returns
650
:param trees: The trees to find file_ids within
651
:param require_versioned: if true, all specified filenames must occur in
653
:return: a set of file ids for the specified filenames and their children.
657
specified_path_ids = _find_ids_across_trees(filenames, trees,
659
return _find_children_across_trees(specified_path_ids, trees)
662
def _find_ids_across_trees(filenames, trees, require_versioned):
663
"""Find the ids corresponding to specified filenames.
665
All matches in all trees will be used, but subdirectories are not scanned.
667
:param filenames: The filenames to find file_ids for
668
:param trees: The trees to find file_ids within
669
:param require_versioned: if true, all specified filenames must occur in
671
:return: a set of file ids for the specified filenames
674
interesting_ids = set()
675
for tree_path in filenames:
678
file_id = tree.path2id(tree_path)
679
if file_id is not None:
680
interesting_ids.add(file_id)
683
not_versioned.append(tree_path)
684
if len(not_versioned) > 0 and require_versioned:
685
raise errors.PathsNotVersionedError(not_versioned)
686
return interesting_ids
689
def _find_children_across_trees(specified_ids, trees):
690
"""Return a set including specified ids and their children.
692
All matches in all trees will be used.
694
:param trees: The trees to find file_ids within
695
:return: a set containing all specified ids and their children
697
interesting_ids = set(specified_ids)
698
pending = interesting_ids
699
# now handle children of interesting ids
700
# we loop so that we handle all children of each id in both trees
701
while len(pending) > 0:
703
for file_id in pending:
705
if not tree.has_id(file_id):
707
entry = tree.inventory[file_id]
708
for child in getattr(entry, 'children', {}).itervalues():
709
if child.file_id not in interesting_ids:
710
new_pending.add(child.file_id)
711
interesting_ids.update(new_pending)
712
pending = new_pending
713
return interesting_ids
716
class InterTree(InterObject):
717
"""This class represents operations taking place between two Trees.
719
Its instances have methods like 'compare' and contain references to the
720
source and target trees these operations are to be carried out on.
722
Clients of bzrlib should not need to use InterTree directly, rather they
723
should use the convenience methods on Tree such as 'Tree.compare()' which
724
will pass through to InterTree as appropriate.
730
def compare(self, want_unchanged=False, specific_files=None,
731
extra_trees=None, require_versioned=False, include_root=False,
732
want_unversioned=False):
733
"""Return the changes from source to target.
735
:return: A TreeDelta.
736
:param specific_files: An optional list of file paths to restrict the
737
comparison to. When mapping filenames to ids, all matches in all
738
trees (including optional extra_trees) are used, and all children of
739
matched directories are included.
740
:param want_unchanged: An optional boolean requesting the inclusion of
741
unchanged entries in the result.
742
:param extra_trees: An optional list of additional trees to use when
743
mapping the contents of specific_files (paths) to file_ids.
744
:param require_versioned: An optional boolean (defaults to False). When
745
supplied and True all the 'specific_files' must be versioned, or
746
a PathsNotVersionedError will be thrown.
747
:param want_unversioned: Scan for unversioned paths.
749
# NB: show_status depends on being able to pass in non-versioned files
750
# and report them as unknown
751
trees = (self.source,)
752
if extra_trees is not None:
753
trees = trees + tuple(extra_trees)
754
# target is usually the newer tree:
755
specific_file_ids = self.target.paths2ids(specific_files, trees,
756
require_versioned=require_versioned)
757
if specific_files and not specific_file_ids:
758
# All files are unversioned, so just return an empty delta
759
# _compare_trees would think we want a complete delta
760
result = delta.TreeDelta()
761
fake_entry = InventoryFile('unused', 'unused', 'unused')
762
result.unversioned = [(path, None,
763
self.target._comparison_data(fake_entry, path)[0]) for path in
766
return delta._compare_trees(self.source, self.target, want_unchanged,
767
specific_files, include_root, extra_trees=extra_trees,
768
require_versioned=require_versioned,
769
want_unversioned=want_unversioned)
771
def iter_changes(self, include_unchanged=False,
772
specific_files=None, pb=None, extra_trees=[],
773
require_versioned=True, want_unversioned=False):
774
"""Generate an iterator of changes between trees.
777
(file_id, (path_in_source, path_in_target),
778
changed_content, versioned, parent, name, kind,
781
Changed_content is True if the file's content has changed. This
782
includes changes to its kind, and to a symlink's target.
784
versioned, parent, name, kind, executable are tuples of (from, to).
785
If a file is missing in a tree, its kind is None.
787
Iteration is done in parent-to-child order, relative to the target
790
There is no guarantee that all paths are in sorted order: the
791
requirement to expand the search due to renames may result in children
792
that should be found early being found late in the search, after
793
lexically later results have been returned.
794
:param require_versioned: Raise errors.PathsNotVersionedError if a
795
path in the specific_files list is not versioned in one of
796
source, target or extra_trees.
797
:param want_unversioned: Should unversioned files be returned in the
798
output. An unversioned file is defined as one with (False, False)
799
for the versioned pair.
802
lookup_trees = [self.source]
804
lookup_trees.extend(extra_trees)
805
if specific_files == []:
806
specific_file_ids = []
808
specific_file_ids = self.target.paths2ids(specific_files,
809
lookup_trees, require_versioned=require_versioned)
811
all_unversioned = sorted([(p.split('/'), p) for p in
813
if specific_files is None or
814
osutils.is_inside_any(specific_files, p)])
815
all_unversioned = deque(all_unversioned)
817
all_unversioned = deque()
819
from_entries_by_dir = list(self.source.inventory.iter_entries_by_dir(
820
specific_file_ids=specific_file_ids))
821
from_data = dict((e.file_id, (p, e)) for p, e in from_entries_by_dir)
822
to_entries_by_dir = list(self.target.inventory.iter_entries_by_dir(
823
specific_file_ids=specific_file_ids))
824
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
826
# the unversioned path lookup only occurs on real trees - where there
827
# can be extras. So the fake_entry is solely used to look up
828
# executable it values when execute is not supported.
829
fake_entry = InventoryFile('unused', 'unused', 'unused')
830
for to_path, to_entry in to_entries_by_dir:
831
while all_unversioned and all_unversioned[0][0] < to_path.split('/'):
832
unversioned_path = all_unversioned.popleft()
833
to_kind, to_executable, to_stat = \
834
self.target._comparison_data(fake_entry, unversioned_path[1])
835
yield (None, (None, unversioned_path[1]), True, (False, False),
837
(None, unversioned_path[0][-1]),
839
(None, to_executable))
840
file_id = to_entry.file_id
841
to_paths[file_id] = to_path
843
changed_content = False
844
from_path, from_entry = from_data.get(file_id, (None, None))
845
from_versioned = (from_entry is not None)
846
if from_entry is not None:
847
from_versioned = True
848
from_name = from_entry.name
849
from_parent = from_entry.parent_id
850
from_kind, from_executable, from_stat = \
851
self.source._comparison_data(from_entry, from_path)
854
from_versioned = False
858
from_executable = None
859
versioned = (from_versioned, True)
860
to_kind, to_executable, to_stat = \
861
self.target._comparison_data(to_entry, to_path)
862
kind = (from_kind, to_kind)
863
if kind[0] != kind[1]:
864
changed_content = True
865
elif from_kind == 'file':
866
from_size = self.source._file_size(from_entry, from_stat)
867
to_size = self.target._file_size(to_entry, to_stat)
868
if from_size != to_size:
869
changed_content = True
870
elif (self.source.get_file_sha1(file_id, from_path, from_stat) !=
871
self.target.get_file_sha1(file_id, to_path, to_stat)):
872
changed_content = True
873
elif from_kind == 'symlink':
874
if (self.source.get_symlink_target(file_id) !=
875
self.target.get_symlink_target(file_id)):
876
changed_content = True
877
elif from_kind == 'tree-reference':
878
if (self.source.get_reference_revision(file_id, from_path)
879
!= self.target.get_reference_revision(file_id, to_path)):
880
changed_content = True
881
parent = (from_parent, to_entry.parent_id)
882
name = (from_name, to_entry.name)
883
executable = (from_executable, to_executable)
885
pb.update('comparing files', entry_count, num_entries)
886
if (changed_content is not False or versioned[0] != versioned[1]
887
or parent[0] != parent[1] or name[0] != name[1] or
888
executable[0] != executable[1] or include_unchanged):
889
yield (file_id, (from_path, to_path), changed_content,
890
versioned, parent, name, kind, executable)
892
while all_unversioned:
893
# yield any trailing unversioned paths
894
unversioned_path = all_unversioned.popleft()
895
to_kind, to_executable, to_stat = \
896
self.target._comparison_data(fake_entry, unversioned_path[1])
897
yield (None, (None, unversioned_path[1]), True, (False, False),
899
(None, unversioned_path[0][-1]),
901
(None, to_executable))
903
def get_to_path(to_entry):
904
if to_entry.parent_id is None:
905
to_path = '' # the root
907
if to_entry.parent_id not in to_paths:
909
return get_to_path(self.target.inventory[to_entry.parent_id])
910
to_path = osutils.pathjoin(to_paths[to_entry.parent_id],
912
to_paths[to_entry.file_id] = to_path
915
for path, from_entry in from_entries_by_dir:
916
file_id = from_entry.file_id
917
if file_id in to_paths:
920
if not file_id in self.target.inventory:
921
# common case - paths we have not emitted are not present in
925
to_path = get_to_path(self.target.inventory[file_id])
928
pb.update('comparing files', entry_count, num_entries)
929
versioned = (True, False)
930
parent = (from_entry.parent_id, None)
931
name = (from_entry.name, None)
932
from_kind, from_executable, stat_value = \
933
self.source._comparison_data(from_entry, path)
934
kind = (from_kind, None)
935
executable = (from_executable, None)
936
changed_content = True
937
# the parent's path is necessarily known at this point.
938
yield(file_id, (path, to_path), changed_content, versioned, parent,
939
name, kind, executable)
added
removed
bzrlib/tree.py
