1
# Copyright (C) 2005-2011 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
17
"""Tree classes, representing directory at point in time.
20
from __future__ import absolute_import
23
from collections.abc import deque
24
except ImportError: # python < 3.7
25
from collections import deque
27
from .lazy_import import lazy_import
28
lazy_import(globals(), """
31
conflicts as _mod_conflicts,
35
revision as _mod_revision,
39
from breezy.i18n import gettext
47
from .inter import InterObject
54
class FileTimestampUnavailable(errors.BzrError):
56
_fmt = "The filestamp for %(path)s is not available."
60
def __init__(self, path):
64
class TreeEntry(object):
65
"""An entry that implements the minimum interface used by commands.
68
def __eq__(self, other):
69
# yes, this is ugly, TODO: best practice __eq__ style.
70
return (isinstance(other, TreeEntry)
71
and other.__class__ == self.__class__)
75
def kind_character(self):
79
class TreeDirectory(TreeEntry):
80
"""See TreeEntry. This is a directory in a working tree."""
84
def kind_character(self):
88
class TreeFile(TreeEntry):
89
"""See TreeEntry. This is a regular file in a working tree."""
93
def kind_character(self):
97
class TreeLink(TreeEntry):
98
"""See TreeEntry. This is a symlink in a working tree."""
102
def kind_character(self):
106
class TreeReference(TreeEntry):
107
"""See TreeEntry. This is a reference to a nested tree in a working tree."""
109
kind = 'tree-reference'
111
def kind_character(self):
116
"""Abstract file tree.
118
There are several subclasses:
120
* `WorkingTree` exists as files on disk editable by the user.
122
* `RevisionTree` is a tree as recorded at some point in the past.
124
Trees can be compared, etc, regardless of whether they are working
125
trees or versioned trees.
128
def supports_rename_tracking(self):
129
"""Whether this tree supports rename tracking.
131
This defaults to True, but some implementations may want to override
136
def has_versioned_directories(self):
137
"""Whether this tree can contain explicitly versioned directories.
139
This defaults to True, but some implementations may want to override
144
def changes_from(self, other, want_unchanged=False, specific_files=None,
145
extra_trees=None, require_versioned=False, include_root=False,
146
want_unversioned=False):
147
"""Return a TreeDelta of the changes from other to this tree.
149
:param other: A tree to compare with.
150
:param specific_files: An optional list of file paths to restrict the
151
comparison to. When mapping filenames to ids, all matches in all
152
trees (including optional extra_trees) are used, and all children of
153
matched directories are included.
154
:param want_unchanged: An optional boolean requesting the inclusion of
155
unchanged entries in the result.
156
:param extra_trees: An optional list of additional trees to use when
157
mapping the contents of specific_files (paths) to their identities.
158
:param require_versioned: An optional boolean (defaults to False). When
159
supplied and True all the 'specific_files' must be versioned, or
160
a PathsNotVersionedError will be thrown.
161
:param want_unversioned: Scan for unversioned paths.
163
The comparison will be performed by an InterTree object looked up on
166
# Martin observes that Tree.changes_from returns a TreeDelta and this
167
# may confuse people, because the class name of the returned object is
168
# a synonym of the object referenced in the method name.
169
return InterTree.get(other, self).compare(
170
want_unchanged=want_unchanged,
171
specific_files=specific_files,
172
extra_trees=extra_trees,
173
require_versioned=require_versioned,
174
include_root=include_root,
175
want_unversioned=want_unversioned,
178
def iter_changes(self, from_tree, include_unchanged=False,
179
specific_files=None, pb=None, extra_trees=None,
180
require_versioned=True, want_unversioned=False):
181
"""See InterTree.iter_changes"""
182
intertree = InterTree.get(from_tree, self)
183
return intertree.iter_changes(include_unchanged, specific_files, pb,
184
extra_trees, require_versioned, want_unversioned=want_unversioned)
187
"""Get a list of the conflicts in the tree.
189
Each conflict is an instance of breezy.conflicts.Conflict.
191
return _mod_conflicts.ConflictList()
194
"""For trees that can have unversioned files, return all such paths."""
197
def get_parent_ids(self):
198
"""Get the parent ids for this tree.
200
:return: a list of parent ids. [] is returned to indicate
201
a tree with no parents.
202
:raises: BzrError if the parents are not known.
204
raise NotImplementedError(self.get_parent_ids)
206
def has_filename(self, filename):
207
"""True if the tree has given filename."""
208
raise NotImplementedError(self.has_filename)
210
def is_ignored(self, filename):
211
"""Check whether the filename is ignored by this tree.
213
:param filename: The relative filename within the tree.
214
:return: True if the filename is ignored.
218
def all_file_ids(self):
219
"""Iterate through all file ids, including ids for missing files."""
220
raise NotImplementedError(self.all_file_ids)
222
def all_versioned_paths(self):
223
"""Iterate through all paths, including paths for missing files."""
224
raise NotImplementedError(self.all_versioned_paths)
226
def id2path(self, file_id):
227
"""Return the path for a file id.
231
raise NotImplementedError(self.id2path)
233
def iter_entries_by_dir(self, specific_files=None):
234
"""Walk the tree in 'by_dir' order.
236
This will yield each entry in the tree as a (path, entry) tuple.
237
The order that they are yielded is:
239
Directories are walked in a depth-first lexicographical order,
240
however, whenever a directory is reached, all of its direct child
241
nodes are yielded in lexicographical order before yielding the
244
For example, in the tree::
254
The yield order (ignoring root) would be::
256
a, f, a/b, a/d, a/b/c, a/d/e, f/g
258
raise NotImplementedError(self.iter_entries_by_dir)
260
def iter_child_entries(self, path):
261
"""Iterate over the children of a directory or tree reference.
263
:param path: Path of the directory
264
:raise NoSuchFile: When the path does not exist
265
:return: Iterator over entries in the directory
267
raise NotImplementedError(self.iter_child_entries)
269
def list_files(self, include_root=False, from_dir=None, recursive=True):
270
"""List all files in this tree.
272
:param include_root: Whether to include the entry for the tree root
273
:param from_dir: Directory under which to list files
274
:param recursive: Whether to list files recursively
275
:return: iterator over tuples of
276
(path, versioned, kind, inventory entry)
278
raise NotImplementedError(self.list_files)
280
def iter_references(self):
281
if self.supports_tree_reference():
282
for path, entry in self.iter_entries_by_dir():
283
if entry.kind == 'tree-reference':
284
yield path, entry.file_id
286
def kind(self, path):
287
raise NotImplementedError("Tree subclass %s must implement kind"
288
% self.__class__.__name__)
290
def stored_kind(self, path):
291
"""File kind stored for this path.
293
May not match kind on disk for working trees. Always available
294
for versioned files, even when the file itself is missing.
296
return self.kind(path)
298
def path_content_summary(self, path):
299
"""Get a summary of the information about path.
301
All the attributes returned are for the canonical form, not the
302
convenient form (if content filters are in use.)
304
:param path: A relative path within the tree.
305
:return: A tuple containing kind, size, exec, sha1-or-link.
306
Kind is always present (see tree.kind()).
307
size is present if kind is file and the size of the
308
canonical form can be cheaply determined, None otherwise.
309
exec is None unless kind is file and the platform supports the 'x'
311
sha1-or-link is the link target if kind is symlink, or the sha1 if
312
it can be obtained without reading the file.
314
raise NotImplementedError(self.path_content_summary)
316
def get_reference_revision(self, path):
317
raise NotImplementedError("Tree subclass %s must implement "
318
"get_reference_revision"
319
% self.__class__.__name__)
321
def _comparison_data(self, entry, path):
322
"""Return a tuple of kind, executable, stat_value for a file.
324
entry may be None if there is no inventory entry for the file, but
325
path must always be supplied.
327
kind is None if there is no file present (even if an inventory id is
328
present). executable is False for non-file entries.
330
raise NotImplementedError(self._comparison_data)
332
def get_file(self, path):
333
"""Return a file object for the file path in the tree.
335
raise NotImplementedError(self.get_file)
337
def get_file_with_stat(self, path):
338
"""Get a file handle and stat object for path.
340
The default implementation returns (self.get_file, None) for backwards
343
:param path: The path of the file.
344
:return: A tuple (file_handle, stat_value_or_None). If the tree has
345
no stat facility, or need for a stat cache feedback during commit,
346
it may return None for the second element of the tuple.
348
return (self.get_file(path), None)
350
def get_file_text(self, path):
351
"""Return the byte content of a file.
353
:param path: The path of the file.
355
:returns: A single byte string for the whole file.
357
with self.get_file(path) as my_file:
358
return my_file.read()
360
def get_file_lines(self, path):
361
"""Return the content of a file, as lines.
363
:param path: The path of the file.
365
return osutils.split_lines(self.get_file_text(path))
367
def get_file_verifier(self, path, stat_value=None):
368
"""Return a verifier for a file.
370
The default implementation returns a sha1.
372
:param path: The path that this file can be found at.
373
These must point to the same object.
374
:param stat_value: Optional stat value for the object
375
:return: Tuple with verifier name and verifier data
377
return ("SHA1", self.get_file_sha1(path, stat_value=stat_value))
379
def get_file_sha1(self, path, stat_value=None):
380
"""Return the SHA1 file for a file.
382
:note: callers should use get_file_verifier instead
383
where possible, as the underlying repository implementation may
384
have quicker access to a non-sha1 verifier.
386
:param path: The path that this file can be found at.
387
:param stat_value: Optional stat value for the object
389
raise NotImplementedError(self.get_file_sha1)
391
def get_file_mtime(self, path):
392
"""Return the modification time for a file.
394
:param path: The path that this file can be found at.
396
raise NotImplementedError(self.get_file_mtime)
398
def get_file_size(self, path):
399
"""Return the size of a file in bytes.
401
This applies only to regular files. If invoked on directories or
402
symlinks, it will return None.
404
raise NotImplementedError(self.get_file_size)
406
def is_executable(self, path):
407
"""Check if a file is executable.
409
:param path: The path that this file can be found at.
411
raise NotImplementedError(self.is_executable)
413
def iter_files_bytes(self, desired_files):
414
"""Iterate through file contents.
416
Files will not necessarily be returned in the order they occur in
417
desired_files. No specific order is guaranteed.
419
Yields pairs of identifier, bytes_iterator. identifier is an opaque
420
value supplied by the caller as part of desired_files. It should
421
uniquely identify the file version in the caller's context. (Examples:
422
an index number or a TreeTransform trans_id.)
424
bytes_iterator is an iterable of bytestrings for the file. The
425
kind of iterable and length of the bytestrings are unspecified, but for
426
this implementation, it is a tuple containing a single bytestring with
427
the complete text of the file.
429
:param desired_files: a list of (path, identifier) pairs
431
for path, identifier in desired_files:
432
# We wrap the string in a tuple so that we can return an iterable
433
# of bytestrings. (Technically, a bytestring is also an iterable
434
# of bytestrings, but iterating through each character is not
436
cur_file = (self.get_file_text(path),)
437
yield identifier, cur_file
439
def get_symlink_target(self, path):
440
"""Get the target for a given path.
442
It is assumed that the caller already knows that path is referencing
444
:param path: The path of the file.
445
:return: The path the symlink points to.
447
raise NotImplementedError(self.get_symlink_target)
449
def get_root_id(self):
450
"""Return the file_id for the root of this tree."""
451
raise NotImplementedError(self.get_root_id)
453
def annotate_iter(self, path,
454
default_revision=_mod_revision.CURRENT_REVISION):
455
"""Return an iterator of revision_id, line tuples.
457
For working trees (and mutable trees in general), the special
458
revision_id 'current:' will be used for lines that are new in this
459
tree, e.g. uncommitted changes.
460
:param path: The file to produce an annotated version from
461
:param default_revision: For lines that don't match a basis, mark them
462
with this revision id. Not all implementations will make use of
465
raise NotImplementedError(self.annotate_iter)
467
def _iter_parent_trees(self):
468
"""Iterate through parent trees, defaulting to Tree.revision_tree."""
469
for revision_id in self.get_parent_ids():
471
yield self.revision_tree(revision_id)
472
except errors.NoSuchRevisionInTree:
473
yield self.repository.revision_tree(revision_id)
475
def path2id(self, path):
476
"""Return the id for path in this tree."""
477
raise NotImplementedError(self.path2id)
479
def is_versioned(self, path):
480
"""Check whether path is versioned.
482
:param path: Path to check
485
return self.path2id(path) is not None
487
def find_related_paths_across_trees(self, paths, trees=[],
488
require_versioned=True):
489
"""Find related paths in tree corresponding to specified filenames in any
492
All matches in all trees will be used, and all children of matched
493
directories will be used.
495
:param paths: The filenames to find related paths for (if None, returns
497
:param trees: The trees to find file_ids within
498
:param require_versioned: if true, all specified filenames must occur in
500
:return: a set of paths for the specified filenames and their children
503
raise NotImplementedError(self.find_related_paths_across_trees)
506
"""Lock this tree for multiple read only operations.
508
:return: A breezy.lock.LogicalLockResult.
510
return lock.LogicalLockResult(self.unlock)
512
def revision_tree(self, revision_id):
513
"""Obtain a revision tree for the revision revision_id.
515
The intention of this method is to allow access to possibly cached
516
tree data. Implementors of this method should raise NoSuchRevision if
517
the tree is not locally available, even if they could obtain the
518
tree via a repository or some other means. Callers are responsible
519
for finding the ultimate source for a revision tree.
521
:param revision_id: The revision_id of the requested tree.
523
:raises: NoSuchRevision if the tree cannot be obtained.
525
raise errors.NoSuchRevisionInTree(self, revision_id)
528
"""What files are present in this tree and unknown.
530
:return: an iterator over the unknown files.
537
def filter_unversioned_files(self, paths):
538
"""Filter out paths that are versioned.
540
:return: set of paths.
542
# NB: we specifically *don't* call self.has_filename, because for
543
# WorkingTrees that can indicate files that exist on disk but that
545
return set(p for p in paths if not self.is_versioned(p))
547
def walkdirs(self, prefix=""):
548
"""Walk the contents of this tree from path down.
550
This yields all the data about the contents of a directory at a time.
551
After each directory has been yielded, if the caller has mutated the
552
list to exclude some directories, they are then not descended into.
554
The data yielded is of the form:
555
((directory-relpath, directory-path-from-root, directory-fileid),
556
[(relpath, basename, kind, lstat, path_from_tree_root, file_id,
557
versioned_kind), ...]),
558
- directory-relpath is the containing dirs relpath from prefix
559
- directory-path-from-root is the containing dirs path from /
560
- directory-fileid is the id of the directory if it is versioned.
561
- relpath is the relative path within the subtree being walked.
562
- basename is the basename
563
- kind is the kind of the file now. If unknonwn then the file is not
564
present within the tree - but it may be recorded as versioned. See
566
- lstat is the stat data *if* the file was statted.
567
- path_from_tree_root is the path from the root of the tree.
568
- file_id is the file_id if the entry is versioned.
569
- versioned_kind is the kind of the file as last recorded in the
570
versioning system. If 'unknown' the file is not versioned.
571
One of 'kind' and 'versioned_kind' must not be 'unknown'.
573
:param prefix: Start walking from prefix within the tree rather than
574
at the root. This allows one to walk a subtree but get paths that are
575
relative to a tree rooted higher up.
576
:return: an iterator over the directory data.
578
raise NotImplementedError(self.walkdirs)
580
def supports_content_filtering(self):
583
def _content_filter_stack(self, path=None):
584
"""The stack of content filters for a path if filtering is supported.
586
Readers will be applied in first-to-last order.
587
Writers will be applied in last-to-first order.
588
Either the path or the file-id needs to be provided.
590
:param path: path relative to the root of the tree
592
:return: the list of filters - [] if there are none
594
filter_pref_names = filters._get_registered_names()
595
if len(filter_pref_names) == 0:
597
prefs = next(self.iter_search_rules([path], filter_pref_names))
598
stk = filters._get_filter_stack_for(prefs)
599
if 'filters' in debug.debug_flags:
601
gettext("*** {0} content-filter: {1} => {2!r}").format(path, prefs, stk))
604
def _content_filter_stack_provider(self):
605
"""A function that returns a stack of ContentFilters.
607
The function takes a path (relative to the top of the tree) and a
608
file-id as parameters.
610
:return: None if content filtering is not supported by this tree.
612
if self.supports_content_filtering():
613
return lambda path, file_id: \
614
self._content_filter_stack(path)
618
def iter_search_rules(self, path_names, pref_names=None,
619
_default_searcher=None):
620
"""Find the preferences for filenames in a tree.
622
:param path_names: an iterable of paths to find attributes for.
623
Paths are given relative to the root of the tree.
624
:param pref_names: the list of preferences to lookup - None for all
625
:param _default_searcher: private parameter to assist testing - don't use
626
:return: an iterator of tuple sequences, one per path-name.
627
See _RulesSearcher.get_items for details on the tuple sequence.
629
if _default_searcher is None:
630
_default_searcher = rules._per_user_searcher
631
searcher = self._get_rules_searcher(_default_searcher)
632
if searcher is not None:
633
if pref_names is not None:
634
for path in path_names:
635
yield searcher.get_selected_items(path, pref_names)
637
for path in path_names:
638
yield searcher.get_items(path)
640
def _get_rules_searcher(self, default_searcher):
641
"""Get the RulesSearcher for this tree given the default one."""
642
searcher = default_searcher
645
def archive(self, format, name, root='', subdir=None,
647
"""Create an archive of this tree.
649
:param format: Format name (e.g. 'tar')
650
:param name: target file name
651
:param root: Root directory name (or None)
652
:param subdir: Subdirectory to export (or None)
653
:return: Iterator over archive chunks
655
from .archive import create_archive
656
with self.lock_read():
657
return create_archive(format, self, name, root,
658
subdir, force_mtime=force_mtime)
661
def versionable_kind(cls, kind):
662
"""Check if this tree support versioning a specific file kind."""
663
return (kind in ('file', 'directory', 'symlink', 'tree-reference'))
666
class InterTree(InterObject):
667
"""This class represents operations taking place between two Trees.
669
Its instances have methods like 'compare' and contain references to the
670
source and target trees these operations are to be carried out on.
672
Clients of breezy should not need to use InterTree directly, rather they
673
should use the convenience methods on Tree such as 'Tree.compare()' which
674
will pass through to InterTree as appropriate.
677
# Formats that will be used to test this InterTree. If both are
678
# None, this InterTree will not be tested (e.g. because a complex
680
_matching_from_tree_format = None
681
_matching_to_tree_format = None
686
def is_compatible(kls, source, target):
687
# The default implementation is naive and uses the public API, so
688
# it works for all trees.
691
def _changes_from_entries(self, source_entry, target_entry, source_path,
693
"""Generate a iter_changes tuple between source_entry and target_entry.
695
:param source_entry: An inventory entry from self.source, or None.
696
:param target_entry: An inventory entry from self.target, or None.
697
:param source_path: The path of source_entry.
698
:param target_path: The path of target_entry.
699
:return: A tuple, item 0 of which is an iter_changes result tuple, and
700
item 1 is True if there are any changes in the result tuple.
702
if source_entry is None:
703
if target_entry is None:
705
file_id = target_entry.file_id
707
file_id = source_entry.file_id
708
if source_entry is not None:
709
source_versioned = True
710
source_name = source_entry.name
711
source_parent = source_entry.parent_id
712
source_kind, source_executable, source_stat = \
713
self.source._comparison_data(source_entry, source_path)
715
source_versioned = False
719
source_executable = None
720
if target_entry is not None:
721
target_versioned = True
722
target_name = target_entry.name
723
target_parent = target_entry.parent_id
724
target_kind, target_executable, target_stat = \
725
self.target._comparison_data(target_entry, target_path)
727
target_versioned = False
731
target_executable = None
732
versioned = (source_versioned, target_versioned)
733
kind = (source_kind, target_kind)
734
changed_content = False
735
if source_kind != target_kind:
736
changed_content = True
737
elif source_kind == 'file':
738
if not self.file_content_matches(
739
source_path, target_path,
740
source_stat, target_stat):
741
changed_content = True
742
elif source_kind == 'symlink':
743
if (self.source.get_symlink_target(source_path) !=
744
self.target.get_symlink_target(target_path)):
745
changed_content = True
746
elif source_kind == 'tree-reference':
747
if (self.source.get_reference_revision(source_path)
748
!= self.target.get_reference_revision(target_path)):
749
changed_content = True
750
parent = (source_parent, target_parent)
751
name = (source_name, target_name)
752
executable = (source_executable, target_executable)
753
if (changed_content is not False or versioned[0] != versioned[1] or
754
parent[0] != parent[1] or name[0] != name[1] or
755
executable[0] != executable[1]):
759
return (file_id, (source_path, target_path), changed_content,
760
versioned, parent, name, kind, executable), changes
762
def compare(self, want_unchanged=False, specific_files=None,
763
extra_trees=None, require_versioned=False, include_root=False,
764
want_unversioned=False):
765
"""Return the changes from source to target.
767
:return: A TreeDelta.
768
:param specific_files: An optional list of file paths to restrict the
769
comparison to. When mapping filenames to ids, all matches in all
770
trees (including optional extra_trees) are used, and all children of
771
matched directories are included.
772
:param want_unchanged: An optional boolean requesting the inclusion of
773
unchanged entries in the result.
774
:param extra_trees: An optional list of additional trees to use when
775
mapping the contents of specific_files (paths) to file_ids.
776
:param require_versioned: An optional boolean (defaults to False). When
777
supplied and True all the 'specific_files' must be versioned, or
778
a PathsNotVersionedError will be thrown.
779
:param want_unversioned: Scan for unversioned paths.
781
trees = (self.source,)
782
if extra_trees is not None:
783
trees = trees + tuple(extra_trees)
784
with self.lock_read():
785
return delta._compare_trees(self.source, self.target, want_unchanged,
786
specific_files, include_root, extra_trees=extra_trees,
787
require_versioned=require_versioned,
788
want_unversioned=want_unversioned)
790
def iter_changes(self, include_unchanged=False,
791
specific_files=None, pb=None, extra_trees=[],
792
require_versioned=True, want_unversioned=False):
793
"""Generate an iterator of changes between trees.
796
(file_id, (path_in_source, path_in_target),
797
changed_content, versioned, parent, name, kind,
800
Changed_content is True if the file's content has changed. This
801
includes changes to its kind, and to a symlink's target.
803
versioned, parent, name, kind, executable are tuples of (from, to).
804
If a file is missing in a tree, its kind is None.
806
Iteration is done in parent-to-child order, relative to the target
809
There is no guarantee that all paths are in sorted order: the
810
requirement to expand the search due to renames may result in children
811
that should be found early being found late in the search, after
812
lexically later results have been returned.
813
:param require_versioned: Raise errors.PathsNotVersionedError if a
814
path in the specific_files list is not versioned in one of
815
source, target or extra_trees.
816
:param specific_files: An optional list of file paths to restrict the
817
comparison to. When mapping filenames to ids, all matches in all
818
trees (including optional extra_trees) are used, and all children
819
of matched directories are included. The parents in the target tree
820
of the specific files up to and including the root of the tree are
821
always evaluated for changes too.
822
:param want_unversioned: Should unversioned files be returned in the
823
output. An unversioned file is defined as one with (False, False)
824
for the versioned pair.
829
extra_trees = list(extra_trees)
830
# The ids of items we need to examine to insure delta consistency.
831
precise_file_ids = set()
832
changed_file_ids = []
833
if specific_files == []:
834
target_specific_files = []
835
source_specific_files = []
837
target_specific_files = self.target.find_related_paths_across_trees(
838
specific_files, [self.source] + extra_trees,
839
require_versioned=require_versioned)
840
source_specific_files = self.source.find_related_paths_across_trees(
841
specific_files, [self.target] + extra_trees,
842
require_versioned=require_versioned)
843
if specific_files is not None:
844
# reparented or added entries must have their parents included
845
# so that valid deltas can be created. The seen_parents set
846
# tracks the parents that we need to have.
847
# The seen_dirs set tracks directory entries we've yielded.
848
# After outputting version object in to_entries we set difference
849
# the two seen sets and start checking parents.
853
all_unversioned = sorted([(p.split('/'), p) for p in
855
if specific_files is None or
856
osutils.is_inside_any(specific_files, p)])
857
all_unversioned = deque(all_unversioned)
859
all_unversioned = deque()
861
from_entries_by_dir = list(self.source.iter_entries_by_dir(
862
specific_files=source_specific_files))
863
from_data = dict((e.file_id, (p, e)) for p, e in from_entries_by_dir)
864
to_entries_by_dir = list(self.target.iter_entries_by_dir(
865
specific_files=target_specific_files))
866
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
868
# the unversioned path lookup only occurs on real trees - where there
869
# can be extras. So the fake_entry is solely used to look up
870
# executable it values when execute is not supported.
871
fake_entry = TreeFile()
872
for target_path, target_entry in to_entries_by_dir:
873
while (all_unversioned and
874
all_unversioned[0][0] < target_path.split('/')):
875
unversioned_path = all_unversioned.popleft()
876
target_kind, target_executable, target_stat = \
877
self.target._comparison_data(
878
fake_entry, unversioned_path[1])
879
yield (None, (None, unversioned_path[1]), True, (False, False),
881
(None, unversioned_path[0][-1]),
883
(None, target_executable))
884
source_path, source_entry = from_data.get(target_entry.file_id,
886
result, changes = self._changes_from_entries(source_entry,
887
target_entry, source_path=source_path, target_path=target_path)
888
to_paths[result[0]] = result[1][1]
893
pb.update('comparing files', entry_count, num_entries)
894
if changes or include_unchanged:
895
if specific_files is not None:
896
new_parent_id = result[4][1]
897
precise_file_ids.add(new_parent_id)
898
changed_file_ids.append(result[0])
900
# Ensure correct behaviour for reparented/added specific files.
901
if specific_files is not None:
903
if result[6][1] == 'directory':
904
seen_dirs.add(result[0])
905
# Record parents of reparented/added entries.
906
versioned = result[3]
908
if not versioned[0] or parents[0] != parents[1]:
909
seen_parents.add(parents[1])
910
while all_unversioned:
911
# yield any trailing unversioned paths
912
unversioned_path = all_unversioned.popleft()
913
to_kind, to_executable, to_stat = \
914
self.target._comparison_data(fake_entry, unversioned_path[1])
915
yield (None, (None, unversioned_path[1]), True, (False, False),
917
(None, unversioned_path[0][-1]),
919
(None, to_executable))
920
# Yield all remaining source paths
921
for path, from_entry in from_entries_by_dir:
922
file_id = from_entry.file_id
923
if file_id in to_paths:
926
to_path = find_previous_path(self.source, self.target, path)
929
pb.update('comparing files', entry_count, num_entries)
930
versioned = (True, False)
931
parent = (from_entry.parent_id, None)
932
name = (from_entry.name, None)
933
from_kind, from_executable, stat_value = \
934
self.source._comparison_data(from_entry, path)
935
kind = (from_kind, None)
936
executable = (from_executable, None)
937
changed_content = from_kind is not None
938
# the parent's path is necessarily known at this point.
939
changed_file_ids.append(file_id)
940
yield(file_id, (path, to_path), changed_content, versioned, parent,
941
name, kind, executable)
942
changed_file_ids = set(changed_file_ids)
943
if specific_files is not None:
944
for result in self._handle_precise_ids(precise_file_ids,
948
def _get_entry(self, tree, path):
949
"""Get an inventory entry from a tree, with missing entries as None.
951
If the tree raises NotImplementedError on accessing .inventory, then
952
this is worked around using iter_entries_by_dir on just the file id
955
:param tree: The tree to lookup the entry in.
956
:param path: The path to look up
958
# No inventory available.
960
iterator = tree.iter_entries_by_dir(specific_files=[path])
961
return next(iterator)[1]
962
except StopIteration:
965
def _handle_precise_ids(self, precise_file_ids, changed_file_ids,
966
discarded_changes=None):
967
"""Fill out a partial iter_changes to be consistent.
969
:param precise_file_ids: The file ids of parents that were seen during
971
:param changed_file_ids: The file ids of already emitted items.
972
:param discarded_changes: An optional dict of precalculated
973
iter_changes items which the partial iter_changes had not output
975
:return: A generator of iter_changes items to output.
977
# process parents of things that had changed under the users
978
# requested paths to prevent incorrect paths or parent ids which
979
# aren't in the tree.
980
while precise_file_ids:
981
precise_file_ids.discard(None)
982
# Don't emit file_ids twice
983
precise_file_ids.difference_update(changed_file_ids)
984
if not precise_file_ids:
986
# If the there was something at a given output path in source, we
987
# have to include the entry from source in the delta, or we would
988
# be putting this entry into a used path.
990
for parent_id in precise_file_ids:
992
paths.append(self.target.id2path(parent_id))
993
except errors.NoSuchId:
994
# This id has been dragged in from the source by delta
995
# expansion and isn't present in target at all: we don't
996
# need to check for path collisions on it.
999
old_id = self.source.path2id(path)
1000
precise_file_ids.add(old_id)
1001
precise_file_ids.discard(None)
1002
current_ids = precise_file_ids
1003
precise_file_ids = set()
1004
# We have to emit all of precise_file_ids that have been altered.
1005
# We may have to output the children of some of those ids if any
1006
# directories have stopped being directories.
1007
for file_id in current_ids:
1009
if discarded_changes:
1010
result = discarded_changes.get(file_id)
1016
source_path = self.source.id2path(file_id)
1017
except errors.NoSuchId:
1021
source_entry = self._get_entry(
1022
self.source, source_path)
1024
target_path = self.target.id2path(file_id)
1025
except errors.NoSuchId:
1029
target_entry = self._get_entry(
1030
self.target, target_path)
1031
result, changes = self._changes_from_entries(
1032
source_entry, target_entry, source_path, target_path)
1035
# Get this parents parent to examine.
1036
new_parent_id = result[4][1]
1037
precise_file_ids.add(new_parent_id)
1039
if (result[6][0] == 'directory' and
1040
result[6][1] != 'directory'):
1041
# This stopped being a directory, the old children have
1043
if source_entry is None:
1044
# Reusing a discarded change.
1045
source_entry = self._get_entry(
1046
self.source, result[1][0])
1047
precise_file_ids.update(
1049
for child in self.source.iter_child_entries(result[1][0]))
1050
changed_file_ids.add(result[0])
1053
def file_content_matches(
1054
self, source_path, target_path,
1055
source_stat=None, target_stat=None):
1056
"""Check if two files are the same in the source and target trees.
1058
This only checks that the contents of the files are the same,
1059
it does not touch anything else.
1061
:param source_path: Path of the file in the source tree
1062
:param target_path: Path of the file in the target tree
1063
:param source_file_id: Optional file id of the file in the source tree
1064
:param target_file_id: Optional file id of the file in the target tree
1065
:param source_stat: Optional stat value of the file in the source tree
1066
:param target_stat: Optional stat value of the file in the target tree
1067
:return: Boolean indicating whether the files have the same contents
1069
with self.lock_read():
1070
source_verifier_kind, source_verifier_data = (
1071
self.source.get_file_verifier(source_path, source_stat))
1072
target_verifier_kind, target_verifier_data = (
1073
self.target.get_file_verifier(
1074
target_path, target_stat))
1075
if source_verifier_kind == target_verifier_kind:
1076
return (source_verifier_data == target_verifier_data)
1077
# Fall back to SHA1 for now
1078
if source_verifier_kind != "SHA1":
1079
source_sha1 = self.source.get_file_sha1(
1080
source_path, source_file_id, source_stat)
1082
source_sha1 = source_verifier_data
1083
if target_verifier_kind != "SHA1":
1084
target_sha1 = self.target.get_file_sha1(
1085
target_path, target_file_id, target_stat)
1087
target_sha1 = target_verifier_data
1088
return (source_sha1 == target_sha1)
1091
InterTree.register_optimiser(InterTree)
1094
class MultiWalker(object):
1095
"""Walk multiple trees simultaneously, getting combined results."""
1097
# Note: This could be written to not assume you can do out-of-order
1098
# lookups. Instead any nodes that don't match in all trees could be
1099
# marked as 'deferred', and then returned in the final cleanup loop.
1100
# For now, I think it is "nicer" to return things as close to the
1101
# "master_tree" order as we can.
1103
def __init__(self, master_tree, other_trees):
1104
"""Create a new MultiWalker.
1106
All trees being walked must implement "iter_entries_by_dir()", such
1107
that they yield (path, object) tuples, where that object will have a
1108
'.file_id' member, that can be used to check equality.
1110
:param master_tree: All trees will be 'slaved' to the master_tree such
1111
that nodes in master_tree will be used as 'first-pass' sync points.
1112
Any nodes that aren't in master_tree will be merged in a second
1114
:param other_trees: A list of other trees to walk simultaneously.
1116
self._master_tree = master_tree
1117
self._other_trees = other_trees
1119
# Keep track of any nodes that were properly processed just out of
1120
# order, that way we don't return them at the end, we don't have to
1121
# track *all* processed file_ids, just the out-of-order ones
1122
self._out_of_order_processed = set()
1125
def _step_one(iterator):
1126
"""Step an iter_entries_by_dir iterator.
1128
:return: (has_more, path, ie)
1129
If has_more is False, path and ie will be None.
1132
path, ie = next(iterator)
1133
except StopIteration:
1134
return False, None, None
1136
return True, path, ie
1139
def _lt_path_by_dirblock(path1, path2):
1140
"""Compare two paths based on what directory they are in.
1142
This generates a sort order, such that all children of a directory are
1143
sorted together, and grandchildren are in the same order as the
1144
children appear. But all grandchildren come after all children.
1146
:param path1: first path
1147
:param path2: the second path
1148
:return: negative number if ``path1`` comes first,
1149
0 if paths are equal
1150
and a positive number if ``path2`` sorts first
1152
# Shortcut this special case
1155
# This is stolen from _dirstate_helpers_py.py, only switching it to
1156
# Unicode objects. Consider using encode_utf8() and then using the
1157
# optimized versions, or maybe writing optimized unicode versions.
1158
if not isinstance(path1, text_type):
1159
raise TypeError("'path1' must be a unicode string, not %s: %r"
1160
% (type(path1), path1))
1161
if not isinstance(path2, text_type):
1162
raise TypeError("'path2' must be a unicode string, not %s: %r"
1163
% (type(path2), path2))
1164
return (MultiWalker._path_to_key(path1) <
1165
MultiWalker._path_to_key(path2))
1168
def _path_to_key(path):
1169
dirname, basename = osutils.split(path)
1170
return (dirname.split(u'/'), basename)
1172
def _lookup_by_file_id(self, extra_entries, other_tree, file_id):
1173
"""Lookup an inventory entry by file_id.
1175
This is called when an entry is missing in the normal order.
1176
Generally this is because a file was either renamed, or it was
1177
deleted/added. If the entry was found in the inventory and not in
1178
extra_entries, it will be added to self._out_of_order_processed
1180
:param extra_entries: A dictionary of {file_id: (path, ie)}. This
1181
should be filled with entries that were found before they were
1182
used. If file_id is present, it will be removed from the
1184
:param other_tree: The Tree to search, in case we didn't find the entry
1186
:param file_id: The file_id to look for
1187
:return: (path, ie) if found or (None, None) if not present.
1189
if file_id in extra_entries:
1190
return extra_entries.pop(file_id)
1191
# TODO: Is id2path better as the first call, or is
1192
# inventory[file_id] better as a first check?
1194
cur_path = other_tree.id2path(file_id)
1195
except errors.NoSuchId:
1197
if cur_path is None:
1200
self._out_of_order_processed.add(file_id)
1201
cur_ie = other_tree.root_inventory.get_entry(file_id)
1202
return (cur_path, cur_ie)
1205
"""Match up the values in the different trees."""
1206
for result in self._walk_master_tree():
1208
self._finish_others()
1209
for result in self._walk_others():
1212
def _walk_master_tree(self):
1213
"""First pass, walk all trees in lock-step.
1215
When we are done, all nodes in the master_tree will have been
1216
processed. _other_walkers, _other_entries, and _others_extra will be
1217
set on 'self' for future processing.
1219
# This iterator has the most "inlining" done, because it tends to touch
1220
# every file in the tree, while the others only hit nodes that don't
1222
master_iterator = self._master_tree.iter_entries_by_dir()
1224
other_walkers = [other.iter_entries_by_dir()
1225
for other in self._other_trees]
1226
other_entries = [self._step_one(walker) for walker in other_walkers]
1227
# Track extra nodes in the other trees
1228
others_extra = [{} for _ in range(len(self._other_trees))]
1230
master_has_more = True
1231
step_one = self._step_one
1232
lookup_by_file_id = self._lookup_by_file_id
1233
out_of_order_processed = self._out_of_order_processed
1235
while master_has_more:
1236
(master_has_more, path, master_ie) = step_one(master_iterator)
1237
if not master_has_more:
1240
file_id = master_ie.file_id
1242
other_values_append = other_values.append
1243
next_other_entries = []
1244
next_other_entries_append = next_other_entries.append
1245
for idx, (other_has_more, other_path, other_ie) in enumerate(other_entries):
1246
if not other_has_more:
1247
other_values_append(lookup_by_file_id(
1248
others_extra[idx], self._other_trees[idx], file_id))
1249
next_other_entries_append((False, None, None))
1250
elif file_id == other_ie.file_id:
1251
# This is the critical code path, as most of the entries
1252
# should match between most trees.
1253
other_values_append((other_path, other_ie))
1254
next_other_entries_append(step_one(other_walkers[idx]))
1256
# This walker did not match, step it until it either
1257
# matches, or we know we are past the current walker.
1258
other_walker = other_walkers[idx]
1259
other_extra = others_extra[idx]
1260
while (other_has_more and
1261
self._lt_path_by_dirblock(other_path, path)):
1262
other_file_id = other_ie.file_id
1263
if other_file_id not in out_of_order_processed:
1264
other_extra[other_file_id] = (other_path, other_ie)
1265
other_has_more, other_path, other_ie = \
1266
step_one(other_walker)
1267
if other_has_more and other_ie.file_id == file_id:
1268
# We ended up walking to this point, match and step
1270
other_values_append((other_path, other_ie))
1271
other_has_more, other_path, other_ie = \
1272
step_one(other_walker)
1274
# This record isn't in the normal order, see if it
1276
other_values_append(lookup_by_file_id(
1277
other_extra, self._other_trees[idx], file_id))
1278
next_other_entries_append((other_has_more, other_path,
1280
other_entries = next_other_entries
1282
# We've matched all the walkers, yield this datapoint
1283
yield path, file_id, master_ie, other_values
1284
self._other_walkers = other_walkers
1285
self._other_entries = other_entries
1286
self._others_extra = others_extra
1288
def _finish_others(self):
1289
"""Finish walking the other iterators, so we get all entries."""
1290
for idx, info in enumerate(self._other_entries):
1291
other_extra = self._others_extra[idx]
1292
(other_has_more, other_path, other_ie) = info
1293
while other_has_more:
1294
other_file_id = other_ie.file_id
1295
if other_file_id not in self._out_of_order_processed:
1296
other_extra[other_file_id] = (other_path, other_ie)
1297
other_has_more, other_path, other_ie = \
1298
self._step_one(self._other_walkers[idx])
1299
del self._other_entries
1301
def _walk_others(self):
1302
"""Finish up by walking all the 'deferred' nodes."""
1303
# TODO: One alternative would be to grab all possible unprocessed
1304
# file_ids, and then sort by path, and then yield them. That
1305
# might ensure better ordering, in case a caller strictly
1306
# requires parents before children.
1307
for idx, other_extra in enumerate(self._others_extra):
1308
others = sorted(viewvalues(other_extra),
1309
key=lambda x: self._path_to_key(x[0]))
1310
for other_path, other_ie in others:
1311
file_id = other_ie.file_id
1312
# We don't need to check out_of_order_processed here, because
1313
# the lookup_by_file_id will be removing anything processed
1314
# from the extras cache
1315
other_extra.pop(file_id)
1316
other_values = [(None, None)] * idx
1317
other_values.append((other_path, other_ie))
1318
for alt_idx, alt_extra in enumerate(self._others_extra[idx + 1:]):
1319
alt_idx = alt_idx + idx + 1
1320
alt_extra = self._others_extra[alt_idx]
1321
alt_tree = self._other_trees[alt_idx]
1322
other_values.append(self._lookup_by_file_id(
1323
alt_extra, alt_tree, file_id))
1324
yield other_path, file_id, None, other_values
1327
def find_previous_paths(from_tree, to_tree, paths):
1328
"""Find previous tree paths.
1330
:param from_tree: From tree
1331
:param to_tree: To tree
1332
:param paths: Iterable over paths to search for
1333
:return: Dictionary mapping from from_tree paths to paths in to_tree, or
1334
None if there is no equivalent path.
1338
ret[path] = find_previous_path(from_tree, to_tree, path)
1342
def find_previous_path(from_tree, to_tree, path, file_id=None):
1343
"""Find previous tree path.
1345
:param from_tree: From tree
1346
:param to_tree: To tree
1347
:param path: Path to search for
1348
:return: path in to_tree, or None if there is no equivalent path.
1351
file_id = from_tree.path2id(path)
1353
raise errors.NoSuchFile(path)
1355
return to_tree.id2path(file_id)
1356
except errors.NoSuchId:
1360
def get_canonical_path(tree, path, normalize):
1361
"""Find the canonical path of an item, ignoring case.
1363
:param tree: Tree to traverse
1364
:param path: Case-insensitive path to look up
1365
:param normalize: Function to normalize a filename for comparison
1366
:return: The canonical path
1370
bit_iter = iter(path.split("/"))
1371
for elt in bit_iter:
1372
lelt = normalize(elt)
1375
for child in tree.iter_child_entries(cur_path):
1377
if child.name == elt:
1378
# if we found an exact match, we can stop now; if
1379
# we found an approximate match we need to keep
1380
# searching because there might be an exact match
1382
new_path = osutils.pathjoin(cur_path, child.name)
1384
elif normalize(child.name) == lelt:
1385
new_path = osutils.pathjoin(cur_path, child.name)
1386
except errors.NoSuchId:
1387
# before a change is committed we can see this error...
1389
except errors.NotADirectory:
1394
# got to the end of this directory and no entries matched.
1395
# Return what matched so far, plus the rest as specified.
1396
cur_path = osutils.pathjoin(cur_path, elt, *list(bit_iter))
added
removed
breezy/tree.py
