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.
21
from collections.abc import deque
22
except ImportError: # python < 3.7
23
from collections import deque
25
from .lazy_import import lazy_import
26
lazy_import(globals(), """
29
conflicts as _mod_conflicts,
33
revision as _mod_revision,
37
from breezy.i18n import gettext
45
from .inter import InterObject
48
class FileTimestampUnavailable(errors.BzrError):
50
_fmt = "The filestamp for %(path)s is not available."
54
def __init__(self, path):
58
class MissingNestedTree(errors.BzrError):
60
_fmt = "The nested tree for %(path)s can not be resolved."""
62
def __init__(self, path):
66
class TreeEntry(object):
67
"""An entry that implements the minimum interface used by commands.
72
def __eq__(self, other):
73
# yes, this is ugly, TODO: best practice __eq__ style.
74
return (isinstance(other, TreeEntry)
75
and other.__class__ == self.__class__)
79
def kind_character(self):
82
def is_unmodified(self, other):
83
"""Does this entry reference the same entry?
85
This is mostly the same as __eq__, but returns False
86
for entries without enough information (i.e. revision is None)
91
class TreeDirectory(TreeEntry):
92
"""See TreeEntry. This is a directory in a working tree."""
98
def kind_character(self):
102
class TreeFile(TreeEntry):
103
"""See TreeEntry. This is a regular file in a working tree."""
109
def kind_character(self):
113
class TreeLink(TreeEntry):
114
"""See TreeEntry. This is a symlink in a working tree."""
120
def kind_character(self):
124
class TreeReference(TreeEntry):
125
"""See TreeEntry. This is a reference to a nested tree in a working tree."""
129
kind = 'tree-reference'
131
def kind_character(self):
135
class TreeChange(object):
136
"""Describes the changes between the same item in two different trees."""
138
__slots__ = ['file_id', 'path', 'changed_content', 'versioned', 'parent_id',
139
'name', 'kind', 'executable', 'copied']
141
def __init__(self, file_id, path, changed_content, versioned, parent_id,
142
name, kind, executable, copied=False):
143
self.file_id = file_id
145
self.changed_content = changed_content
146
self.versioned = versioned
147
self.parent_id = parent_id
150
self.executable = executable
154
return "%s%r" % (self.__class__.__name__, self._as_tuple())
157
return len(self.__slots__)
160
return (self.file_id, self.path, self.changed_content, self.versioned,
161
self.parent_id, self.name, self.kind, self.executable, self.copied)
163
def __eq__(self, other):
164
if isinstance(other, TreeChange):
165
return self._as_tuple() == other._as_tuple()
166
if isinstance(other, tuple):
167
return self._as_tuple() == other
170
def __lt__(self, other):
171
return self._as_tuple() < other._as_tuple()
173
def meta_modified(self):
174
if self.versioned == (True, True):
175
return (self.executable[0] != self.executable[1])
178
def is_reparented(self):
179
return self.parent_id[0] != self.parent_id[1]
181
def discard_new(self):
182
return self.__class__(
183
self.file_id, (self.path[0], None), self.changed_content,
184
(self.versioned[0], None), (self.parent_id[0], None),
185
(self.name[0], None), (self.kind[0], None),
186
(self.executable[0], None),
191
"""Abstract file tree.
193
There are several subclasses:
195
* `WorkingTree` exists as files on disk editable by the user.
197
* `RevisionTree` is a tree as recorded at some point in the past.
199
Trees can be compared, etc, regardless of whether they are working
200
trees or versioned trees.
203
def supports_rename_tracking(self):
204
"""Whether this tree supports rename tracking.
206
This defaults to True, but some implementations may want to override
211
def has_versioned_directories(self):
212
"""Whether this tree can contain explicitly versioned directories.
214
This defaults to True, but some implementations may want to override
219
def supports_symlinks(self):
220
"""Does this tree support symbolic links?
222
return osutils.has_symlinks()
224
def changes_from(self, other, want_unchanged=False, specific_files=None,
225
extra_trees=None, require_versioned=False, include_root=False,
226
want_unversioned=False):
227
"""Return a TreeDelta of the changes from other to this tree.
229
:param other: A tree to compare with.
230
:param specific_files: An optional list of file paths to restrict the
231
comparison to. When mapping filenames to ids, all matches in all
232
trees (including optional extra_trees) are used, and all children of
233
matched directories are included.
234
:param want_unchanged: An optional boolean requesting the inclusion of
235
unchanged entries in the result.
236
:param extra_trees: An optional list of additional trees to use when
237
mapping the contents of specific_files (paths) to their identities.
238
:param require_versioned: An optional boolean (defaults to False). When
239
supplied and True all the 'specific_files' must be versioned, or
240
a PathsNotVersionedError will be thrown.
241
:param want_unversioned: Scan for unversioned paths.
243
The comparison will be performed by an InterTree object looked up on
246
# Martin observes that Tree.changes_from returns a TreeDelta and this
247
# may confuse people, because the class name of the returned object is
248
# a synonym of the object referenced in the method name.
249
return InterTree.get(other, self).compare(
250
want_unchanged=want_unchanged,
251
specific_files=specific_files,
252
extra_trees=extra_trees,
253
require_versioned=require_versioned,
254
include_root=include_root,
255
want_unversioned=want_unversioned,
258
def iter_changes(self, from_tree, include_unchanged=False,
259
specific_files=None, pb=None, extra_trees=None,
260
require_versioned=True, want_unversioned=False):
261
"""See InterTree.iter_changes"""
262
intertree = InterTree.get(from_tree, self)
263
return intertree.iter_changes(include_unchanged, specific_files, pb,
264
extra_trees, require_versioned,
265
want_unversioned=want_unversioned)
268
"""Get a list of the conflicts in the tree.
270
Each conflict is an instance of breezy.conflicts.Conflict.
272
return _mod_conflicts.ConflictList()
275
"""For trees that can have unversioned files, return all such paths."""
278
def get_parent_ids(self):
279
"""Get the parent ids for this tree.
281
:return: a list of parent ids. [] is returned to indicate
282
a tree with no parents.
283
:raises: BzrError if the parents are not known.
285
raise NotImplementedError(self.get_parent_ids)
287
def has_filename(self, filename):
288
"""True if the tree has given filename."""
289
raise NotImplementedError(self.has_filename)
291
def is_ignored(self, filename):
292
"""Check whether the filename is ignored by this tree.
294
:param filename: The relative filename within the tree.
295
:return: True if the filename is ignored.
299
def all_file_ids(self):
300
"""Iterate through all file ids, including ids for missing files."""
301
raise NotImplementedError(self.all_file_ids)
303
def all_versioned_paths(self):
304
"""Iterate through all paths, including paths for missing files."""
305
raise NotImplementedError(self.all_versioned_paths)
307
def id2path(self, file_id, recurse='down'):
308
"""Return the path for a file id.
312
raise NotImplementedError(self.id2path)
314
def iter_entries_by_dir(self, specific_files=None, recurse_nested=False):
315
"""Walk the tree in 'by_dir' order.
317
This will yield each entry in the tree as a (path, entry) tuple.
318
The order that they are yielded is:
320
Directories are walked in a depth-first lexicographical order,
321
however, whenever a directory is reached, all of its direct child
322
nodes are yielded in lexicographical order before yielding the
325
For example, in the tree::
335
The yield order (ignoring root) would be::
337
a, f, a/b, a/d, a/b/c, a/d/e, f/g
339
If recurse_nested is enabled then nested trees are included as if
340
they were a part of the tree. If is disabled then TreeReference
341
objects (without any children) are yielded.
343
raise NotImplementedError(self.iter_entries_by_dir)
345
def iter_child_entries(self, path):
346
"""Iterate over the children of a directory or tree reference.
348
:param path: Path of the directory
349
:raise NoSuchFile: When the path does not exist
350
:return: Iterator over entries in the directory
352
raise NotImplementedError(self.iter_child_entries)
354
def list_files(self, include_root=False, from_dir=None, recursive=True,
355
recurse_nested=False):
356
"""List all files in this tree.
358
:param include_root: Whether to include the entry for the tree root
359
:param from_dir: Directory under which to list files
360
:param recursive: Whether to list files recursively
361
:param recurse_nested: enter nested trees
362
:return: iterator over tuples of
363
(path, versioned, kind, inventory entry)
365
raise NotImplementedError(self.list_files)
367
def iter_references(self):
368
if self.supports_tree_reference():
369
for path, entry in self.iter_entries_by_dir():
370
if entry.kind == 'tree-reference':
373
def get_containing_nested_tree(self, path):
374
"""Find the nested tree that contains a path.
376
:return: tuple with (nested tree and path inside the nested tree)
378
for nested_path in self.iter_references():
380
if path.startswith(nested_path):
381
nested_tree = self.get_nested_tree(nested_path)
382
return nested_tree, path[len(nested_path):]
386
def get_nested_tree(self, path):
387
"""Open the nested tree at the specified path.
389
:param path: Path from which to resolve tree reference.
390
:return: A Tree object for the nested tree
391
:raise MissingNestedTree: If the nested tree can not be resolved
393
raise NotImplementedError(self.get_nested_tree)
395
def kind(self, path):
396
raise NotImplementedError("Tree subclass %s must implement kind"
397
% self.__class__.__name__)
399
def stored_kind(self, path):
400
"""File kind stored for this path.
402
May not match kind on disk for working trees. Always available
403
for versioned files, even when the file itself is missing.
405
return self.kind(path)
407
def path_content_summary(self, path):
408
"""Get a summary of the information about path.
410
All the attributes returned are for the canonical form, not the
411
convenient form (if content filters are in use.)
413
:param path: A relative path within the tree.
414
:return: A tuple containing kind, size, exec, sha1-or-link.
415
Kind is always present (see tree.kind()).
416
size is present if kind is file and the size of the
417
canonical form can be cheaply determined, None otherwise.
418
exec is None unless kind is file and the platform supports the 'x'
420
sha1-or-link is the link target if kind is symlink, or the sha1 if
421
it can be obtained without reading the file.
423
raise NotImplementedError(self.path_content_summary)
425
def get_reference_revision(self, path, branch=None):
426
raise NotImplementedError("Tree subclass %s must implement "
427
"get_reference_revision"
428
% self.__class__.__name__)
430
def _comparison_data(self, entry, path):
431
"""Return a tuple of kind, executable, stat_value for a file.
433
entry may be None if there is no inventory entry for the file, but
434
path must always be supplied.
436
kind is None if there is no file present (even if an inventory id is
437
present). executable is False for non-file entries.
439
raise NotImplementedError(self._comparison_data)
441
def get_file(self, path):
442
"""Return a file object for the file path in the tree.
444
raise NotImplementedError(self.get_file)
446
def get_file_with_stat(self, path):
447
"""Get a file handle and stat object for path.
449
The default implementation returns (self.get_file, None) for backwards
452
:param path: The path of the file.
453
:return: A tuple (file_handle, stat_value_or_None). If the tree has
454
no stat facility, or need for a stat cache feedback during commit,
455
it may return None for the second element of the tuple.
457
return (self.get_file(path), None)
459
def get_file_text(self, path):
460
"""Return the byte content of a file.
462
:param path: The path of the file.
464
:returns: A single byte string for the whole file.
466
with self.get_file(path) as my_file:
467
return my_file.read()
469
def get_file_lines(self, path):
470
"""Return the content of a file, as lines.
472
:param path: The path of the file.
474
return osutils.split_lines(self.get_file_text(path))
476
def get_file_verifier(self, path, stat_value=None):
477
"""Return a verifier for a file.
479
The default implementation returns a sha1.
481
:param path: The path that this file can be found at.
482
These must point to the same object.
483
:param stat_value: Optional stat value for the object
484
:return: Tuple with verifier name and verifier data
486
return ("SHA1", self.get_file_sha1(path, stat_value=stat_value))
488
def get_file_sha1(self, path, stat_value=None):
489
"""Return the SHA1 file for a file.
491
:note: callers should use get_file_verifier instead
492
where possible, as the underlying repository implementation may
493
have quicker access to a non-sha1 verifier.
495
:param path: The path that this file can be found at.
496
:param stat_value: Optional stat value for the object
498
raise NotImplementedError(self.get_file_sha1)
500
def get_file_mtime(self, path):
501
"""Return the modification time for a file.
503
:param path: The path that this file can be found at.
505
raise NotImplementedError(self.get_file_mtime)
507
def get_file_size(self, path):
508
"""Return the size of a file in bytes.
510
This applies only to regular files. If invoked on directories or
511
symlinks, it will return None.
513
raise NotImplementedError(self.get_file_size)
515
def is_executable(self, path):
516
"""Check if a file is executable.
518
:param path: The path that this file can be found at.
520
raise NotImplementedError(self.is_executable)
522
def iter_files_bytes(self, desired_files):
523
"""Iterate through file contents.
525
Files will not necessarily be returned in the order they occur in
526
desired_files. No specific order is guaranteed.
528
Yields pairs of identifier, bytes_iterator. identifier is an opaque
529
value supplied by the caller as part of desired_files. It should
530
uniquely identify the file version in the caller's context. (Examples:
531
an index number or a TreeTransform trans_id.)
533
bytes_iterator is an iterable of bytestrings for the file. The
534
kind of iterable and length of the bytestrings are unspecified, but for
535
this implementation, it is a tuple containing a single bytestring with
536
the complete text of the file.
538
:param desired_files: a list of (path, identifier) pairs
540
for path, identifier in desired_files:
541
# We wrap the string in a tuple so that we can return an iterable
542
# of bytestrings. (Technically, a bytestring is also an iterable
543
# of bytestrings, but iterating through each character is not
545
cur_file = (self.get_file_text(path),)
546
yield identifier, cur_file
548
def get_symlink_target(self, path):
549
"""Get the target for a given path.
551
It is assumed that the caller already knows that path is referencing
553
:param path: The path of the file.
554
:return: The path the symlink points to.
556
raise NotImplementedError(self.get_symlink_target)
558
def annotate_iter(self, path,
559
default_revision=_mod_revision.CURRENT_REVISION):
560
"""Return an iterator of revision_id, line tuples.
562
For working trees (and mutable trees in general), the special
563
revision_id 'current:' will be used for lines that are new in this
564
tree, e.g. uncommitted changes.
565
:param path: The file to produce an annotated version from
566
:param default_revision: For lines that don't match a basis, mark them
567
with this revision id. Not all implementations will make use of
570
raise NotImplementedError(self.annotate_iter)
572
def path2id(self, path):
573
"""Return the id for path in this tree."""
574
raise NotImplementedError(self.path2id)
576
def is_versioned(self, path):
577
"""Check whether path is versioned.
579
:param path: Path to check
582
return self.path2id(path) is not None
584
def find_related_paths_across_trees(self, paths, trees=[],
585
require_versioned=True):
586
"""Find related paths in tree corresponding to specified filenames in any
589
All matches in all trees will be used, and all children of matched
590
directories will be used.
592
:param paths: The filenames to find related paths for (if None, returns
594
:param trees: The trees to find file_ids within
595
:param require_versioned: if true, all specified filenames must occur in
597
:return: a set of paths for the specified filenames and their children
600
raise NotImplementedError(self.find_related_paths_across_trees)
603
"""Lock this tree for multiple read only operations.
605
:return: A breezy.lock.LogicalLockResult.
607
return lock.LogicalLockResult(self.unlock)
609
def revision_tree(self, revision_id):
610
"""Obtain a revision tree for the revision revision_id.
612
The intention of this method is to allow access to possibly cached
613
tree data. Implementors of this method should raise NoSuchRevision if
614
the tree is not locally available, even if they could obtain the
615
tree via a repository or some other means. Callers are responsible
616
for finding the ultimate source for a revision tree.
618
:param revision_id: The revision_id of the requested tree.
620
:raises: NoSuchRevision if the tree cannot be obtained.
622
raise errors.NoSuchRevisionInTree(self, revision_id)
625
"""What files are present in this tree and unknown.
627
:return: an iterator over the unknown files.
634
def filter_unversioned_files(self, paths):
635
"""Filter out paths that are versioned.
637
:return: set of paths.
639
# NB: we specifically *don't* call self.has_filename, because for
640
# WorkingTrees that can indicate files that exist on disk but that
642
return set(p for p in paths if not self.is_versioned(p))
644
def walkdirs(self, prefix=""):
645
"""Walk the contents of this tree from path down.
647
This yields all the data about the contents of a directory at a time.
648
After each directory has been yielded, if the caller has mutated the
649
list to exclude some directories, they are then not descended into.
651
The data yielded is of the form:
652
((directory-relpath, directory-path-from-root, directory-fileid),
653
[(relpath, basename, kind, lstat, path_from_tree_root, file_id,
654
versioned_kind), ...]),
655
- directory-relpath is the containing dirs relpath from prefix
656
- directory-path-from-root is the containing dirs path from /
657
- directory-fileid is the id of the directory if it is versioned.
658
- relpath is the relative path within the subtree being walked.
659
- basename is the basename
660
- kind is the kind of the file now. If unknonwn then the file is not
661
present within the tree - but it may be recorded as versioned. See
663
- lstat is the stat data *if* the file was statted.
664
- path_from_tree_root is the path from the root of the tree.
665
- file_id is the file_id if the entry is versioned.
666
- versioned_kind is the kind of the file as last recorded in the
667
versioning system. If 'unknown' the file is not versioned.
668
One of 'kind' and 'versioned_kind' must not be 'unknown'.
670
:param prefix: Start walking from prefix within the tree rather than
671
at the root. This allows one to walk a subtree but get paths that are
672
relative to a tree rooted higher up.
673
:return: an iterator over the directory data.
675
raise NotImplementedError(self.walkdirs)
677
def supports_content_filtering(self):
680
def _content_filter_stack(self, path=None):
681
"""The stack of content filters for a path if filtering is supported.
683
Readers will be applied in first-to-last order.
684
Writers will be applied in last-to-first order.
685
Either the path or the file-id needs to be provided.
687
:param path: path relative to the root of the tree
689
:return: the list of filters - [] if there are none
691
filter_pref_names = filters._get_registered_names()
692
if len(filter_pref_names) == 0:
694
prefs = next(self.iter_search_rules([path], filter_pref_names))
695
stk = filters._get_filter_stack_for(prefs)
696
if 'filters' in debug.debug_flags:
698
gettext("*** {0} content-filter: {1} => {2!r}").format(path, prefs, stk))
701
def _content_filter_stack_provider(self):
702
"""A function that returns a stack of ContentFilters.
704
The function takes a path (relative to the top of the tree) and a
705
file-id as parameters.
707
:return: None if content filtering is not supported by this tree.
709
if self.supports_content_filtering():
710
return self._content_filter_stack
714
def iter_search_rules(self, path_names, pref_names=None,
715
_default_searcher=None):
716
"""Find the preferences for filenames in a tree.
718
:param path_names: an iterable of paths to find attributes for.
719
Paths are given relative to the root of the tree.
720
:param pref_names: the list of preferences to lookup - None for all
721
:param _default_searcher: private parameter to assist testing - don't use
722
:return: an iterator of tuple sequences, one per path-name.
723
See _RulesSearcher.get_items for details on the tuple sequence.
725
if _default_searcher is None:
726
_default_searcher = rules._per_user_searcher
727
searcher = self._get_rules_searcher(_default_searcher)
728
if searcher is not None:
729
if pref_names is not None:
730
for path in path_names:
731
yield searcher.get_selected_items(path, pref_names)
733
for path in path_names:
734
yield searcher.get_items(path)
736
def _get_rules_searcher(self, default_searcher):
737
"""Get the RulesSearcher for this tree given the default one."""
738
searcher = default_searcher
741
def archive(self, format, name, root='', subdir=None,
743
"""Create an archive of this tree.
745
:param format: Format name (e.g. 'tar')
746
:param name: target file name
747
:param root: Root directory name (or None)
748
:param subdir: Subdirectory to export (or None)
749
:return: Iterator over archive chunks
751
from .archive import create_archive
752
with self.lock_read():
753
return create_archive(format, self, name, root,
754
subdir, force_mtime=force_mtime)
757
def versionable_kind(cls, kind):
758
"""Check if this tree support versioning a specific file kind."""
759
return (kind in ('file', 'directory', 'symlink', 'tree-reference'))
762
class InterTree(InterObject):
763
"""This class represents operations taking place between two Trees.
765
Its instances have methods like 'compare' and contain references to the
766
source and target trees these operations are to be carried out on.
768
Clients of breezy should not need to use InterTree directly, rather they
769
should use the convenience methods on Tree such as 'Tree.compare()' which
770
will pass through to InterTree as appropriate.
773
# Formats that will be used to test this InterTree. If both are
774
# None, this InterTree will not be tested (e.g. because a complex
776
_matching_from_tree_format = None
777
_matching_to_tree_format = None
782
def is_compatible(kls, source, target):
783
# The default implementation is naive and uses the public API, so
784
# it works for all trees.
787
def _changes_from_entries(self, source_entry, target_entry, source_path,
789
"""Generate a iter_changes tuple between source_entry and target_entry.
791
:param source_entry: An inventory entry from self.source, or None.
792
:param target_entry: An inventory entry from self.target, or None.
793
:param source_path: The path of source_entry.
794
:param target_path: The path of target_entry.
795
:return: A tuple, item 0 of which is an iter_changes result tuple, and
796
item 1 is True if there are any changes in the result tuple.
798
if source_entry is None:
799
if target_entry is None:
801
file_id = target_entry.file_id
803
file_id = source_entry.file_id
804
if source_entry is not None:
805
source_versioned = True
806
source_name = source_entry.name
807
source_parent = source_entry.parent_id
808
source_kind, source_executable, source_stat = \
809
self.source._comparison_data(source_entry, source_path)
811
source_versioned = False
815
source_executable = None
816
if target_entry is not None:
817
target_versioned = True
818
target_name = target_entry.name
819
target_parent = target_entry.parent_id
820
target_kind, target_executable, target_stat = \
821
self.target._comparison_data(target_entry, target_path)
823
target_versioned = False
827
target_executable = None
828
versioned = (source_versioned, target_versioned)
829
kind = (source_kind, target_kind)
830
changed_content = False
831
if source_kind != target_kind:
832
changed_content = True
833
elif source_kind == 'file':
834
if not self.file_content_matches(
835
source_path, target_path,
836
source_stat, target_stat):
837
changed_content = True
838
elif source_kind == 'symlink':
839
if (self.source.get_symlink_target(source_path) !=
840
self.target.get_symlink_target(target_path)):
841
changed_content = True
842
elif source_kind == 'tree-reference':
843
if (self.source.get_reference_revision(source_path)
844
!= self.target.get_reference_revision(target_path)):
845
changed_content = True
846
parent = (source_parent, target_parent)
847
name = (source_name, target_name)
848
executable = (source_executable, target_executable)
849
if (changed_content is not False or versioned[0] != versioned[1] or
850
parent[0] != parent[1] or name[0] != name[1] or
851
executable[0] != executable[1]):
856
file_id, (source_path, target_path), changed_content,
857
versioned, parent, name, kind, executable), changes
859
def compare(self, want_unchanged=False, specific_files=None,
860
extra_trees=None, require_versioned=False, include_root=False,
861
want_unversioned=False):
862
"""Return the changes from source to target.
864
:return: A TreeDelta.
865
:param specific_files: An optional list of file paths to restrict the
866
comparison to. When mapping filenames to ids, all matches in all
867
trees (including optional extra_trees) are used, and all children of
868
matched directories are included.
869
:param want_unchanged: An optional boolean requesting the inclusion of
870
unchanged entries in the result.
871
:param extra_trees: An optional list of additional trees to use when
872
mapping the contents of specific_files (paths) to file_ids.
873
:param require_versioned: An optional boolean (defaults to False). When
874
supplied and True all the 'specific_files' must be versioned, or
875
a PathsNotVersionedError will be thrown.
876
:param want_unversioned: Scan for unversioned paths.
878
trees = (self.source,)
879
if extra_trees is not None:
880
trees = trees + tuple(extra_trees)
881
with self.lock_read():
882
return delta._compare_trees(self.source, self.target, want_unchanged,
883
specific_files, include_root, extra_trees=extra_trees,
884
require_versioned=require_versioned,
885
want_unversioned=want_unversioned)
887
def iter_changes(self, include_unchanged=False,
888
specific_files=None, pb=None, extra_trees=[],
889
require_versioned=True, want_unversioned=False):
890
"""Generate an iterator of changes between trees.
893
(file_id, (path_in_source, path_in_target),
894
changed_content, versioned, parent, name, kind,
897
Changed_content is True if the file's content has changed. This
898
includes changes to its kind, and to a symlink's target.
900
versioned, parent, name, kind, executable are tuples of (from, to).
901
If a file is missing in a tree, its kind is None.
903
Iteration is done in parent-to-child order, relative to the target
906
There is no guarantee that all paths are in sorted order: the
907
requirement to expand the search due to renames may result in children
908
that should be found early being found late in the search, after
909
lexically later results have been returned.
910
:param require_versioned: Raise errors.PathsNotVersionedError if a
911
path in the specific_files list is not versioned in one of
912
source, target or extra_trees.
913
:param specific_files: An optional list of file paths to restrict the
914
comparison to. When mapping filenames to ids, all matches in all
915
trees (including optional extra_trees) are used, and all children
916
of matched directories are included. The parents in the target tree
917
of the specific files up to and including the root of the tree are
918
always evaluated for changes too.
919
:param want_unversioned: Should unversioned files be returned in the
920
output. An unversioned file is defined as one with (False, False)
921
for the versioned pair.
926
extra_trees = list(extra_trees)
927
# The ids of items we need to examine to insure delta consistency.
928
precise_file_ids = set()
929
changed_file_ids = []
930
if specific_files == []:
931
target_specific_files = []
932
source_specific_files = []
934
target_specific_files = self.target.find_related_paths_across_trees(
935
specific_files, [self.source] + extra_trees,
936
require_versioned=require_versioned)
937
source_specific_files = self.source.find_related_paths_across_trees(
938
specific_files, [self.target] + extra_trees,
939
require_versioned=require_versioned)
940
if specific_files is not None:
941
# reparented or added entries must have their parents included
942
# so that valid deltas can be created. The seen_parents set
943
# tracks the parents that we need to have.
944
# The seen_dirs set tracks directory entries we've yielded.
945
# After outputting version object in to_entries we set difference
946
# the two seen sets and start checking parents.
950
all_unversioned = sorted([(p.split('/'), p) for p in
952
if specific_files is None or
953
osutils.is_inside_any(specific_files, p)])
954
all_unversioned = deque(all_unversioned)
956
all_unversioned = deque()
958
from_entries_by_dir = list(self.source.iter_entries_by_dir(
959
specific_files=source_specific_files))
960
from_data = dict(from_entries_by_dir)
961
to_entries_by_dir = list(self.target.iter_entries_by_dir(
962
specific_files=target_specific_files))
963
path_equivs = self.find_source_paths([p for p, e in to_entries_by_dir])
964
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
966
# the unversioned path lookup only occurs on real trees - where there
967
# can be extras. So the fake_entry is solely used to look up
968
# executable it values when execute is not supported.
969
fake_entry = TreeFile()
970
for target_path, target_entry in to_entries_by_dir:
971
while (all_unversioned and
972
all_unversioned[0][0] < target_path.split('/')):
973
unversioned_path = all_unversioned.popleft()
974
target_kind, target_executable, target_stat = \
975
self.target._comparison_data(
976
fake_entry, unversioned_path[1])
978
None, (None, unversioned_path[1]), True, (False, False),
980
(None, unversioned_path[0][-1]),
982
(None, target_executable))
983
source_path = path_equivs[target_path]
984
if source_path is not None:
985
source_entry = from_data.get(source_path)
988
result, changes = self._changes_from_entries(
989
source_entry, target_entry, source_path=source_path, target_path=target_path)
990
to_paths[result.file_id] = result.path[1]
992
if result.versioned[0]:
995
pb.update('comparing files', entry_count, num_entries)
996
if changes or include_unchanged:
997
if specific_files is not None:
998
precise_file_ids.add(result.parent_id[1])
999
changed_file_ids.append(result.file_id)
1001
# Ensure correct behaviour for reparented/added specific files.
1002
if specific_files is not None:
1003
# Record output dirs
1004
if result.kind[1] == 'directory':
1005
seen_dirs.add(result.file_id)
1006
# Record parents of reparented/added entries.
1007
if not result.versioned[0] or result.is_reparented():
1008
seen_parents.add(result.parent_id[1])
1009
while all_unversioned:
1010
# yield any trailing unversioned paths
1011
unversioned_path = all_unversioned.popleft()
1012
to_kind, to_executable, to_stat = \
1013
self.target._comparison_data(fake_entry, unversioned_path[1])
1015
None, (None, unversioned_path[1]), True, (False, False),
1017
(None, unversioned_path[0][-1]),
1019
(None, to_executable))
1020
# Yield all remaining source paths
1021
for path, from_entry in from_entries_by_dir:
1022
file_id = from_entry.file_id
1023
if file_id in to_paths:
1026
to_path = self.find_target_path(path)
1029
pb.update('comparing files', entry_count, num_entries)
1030
versioned = (True, False)
1031
parent = (from_entry.parent_id, None)
1032
name = (from_entry.name, None)
1033
from_kind, from_executable, stat_value = \
1034
self.source._comparison_data(from_entry, path)
1035
kind = (from_kind, None)
1036
executable = (from_executable, None)
1037
changed_content = from_kind is not None
1038
# the parent's path is necessarily known at this point.
1039
changed_file_ids.append(file_id)
1041
file_id, (path, to_path), changed_content, versioned, parent,
1042
name, kind, executable)
1043
changed_file_ids = set(changed_file_ids)
1044
if specific_files is not None:
1045
for result in self._handle_precise_ids(precise_file_ids,
1050
def _get_entry(tree, path):
1051
"""Get an inventory entry from a tree, with missing entries as None.
1053
If the tree raises NotImplementedError on accessing .inventory, then
1054
this is worked around using iter_entries_by_dir on just the file id
1057
:param tree: The tree to lookup the entry in.
1058
:param path: The path to look up
1060
# No inventory available.
1062
iterator = tree.iter_entries_by_dir(specific_files=[path])
1063
return next(iterator)[1]
1064
except StopIteration:
1067
def _handle_precise_ids(self, precise_file_ids, changed_file_ids,
1068
discarded_changes=None):
1069
"""Fill out a partial iter_changes to be consistent.
1071
:param precise_file_ids: The file ids of parents that were seen during
1073
:param changed_file_ids: The file ids of already emitted items.
1074
:param discarded_changes: An optional dict of precalculated
1075
iter_changes items which the partial iter_changes had not output
1077
:return: A generator of iter_changes items to output.
1079
# process parents of things that had changed under the users
1080
# requested paths to prevent incorrect paths or parent ids which
1081
# aren't in the tree.
1082
while precise_file_ids:
1083
precise_file_ids.discard(None)
1084
# Don't emit file_ids twice
1085
precise_file_ids.difference_update(changed_file_ids)
1086
if not precise_file_ids:
1088
# If the there was something at a given output path in source, we
1089
# have to include the entry from source in the delta, or we would
1090
# be putting this entry into a used path.
1092
for parent_id in precise_file_ids:
1094
paths.append(self.target.id2path(parent_id))
1095
except errors.NoSuchId:
1096
# This id has been dragged in from the source by delta
1097
# expansion and isn't present in target at all: we don't
1098
# need to check for path collisions on it.
1101
old_id = self.source.path2id(path)
1102
precise_file_ids.add(old_id)
1103
precise_file_ids.discard(None)
1104
current_ids = precise_file_ids
1105
precise_file_ids = set()
1106
# We have to emit all of precise_file_ids that have been altered.
1107
# We may have to output the children of some of those ids if any
1108
# directories have stopped being directories.
1109
for file_id in current_ids:
1111
if discarded_changes:
1112
result = discarded_changes.get(file_id)
1118
source_path = self.source.id2path(file_id)
1119
except errors.NoSuchId:
1123
source_entry = self._get_entry(
1124
self.source, source_path)
1126
target_path = self.target.id2path(file_id)
1127
except errors.NoSuchId:
1131
target_entry = self._get_entry(
1132
self.target, target_path)
1133
result, changes = self._changes_from_entries(
1134
source_entry, target_entry, source_path, target_path)
1137
# Get this parents parent to examine.
1138
new_parent_id = result.parent_id[1]
1139
precise_file_ids.add(new_parent_id)
1141
if (result.kind[0] == 'directory' and
1142
result.kind[1] != 'directory'):
1143
# This stopped being a directory, the old children have
1145
if source_entry is None:
1146
# Reusing a discarded change.
1147
source_entry = self._get_entry(
1148
self.source, result.path[0])
1149
precise_file_ids.update(
1151
for child in self.source.iter_child_entries(result.path[0]))
1152
changed_file_ids.add(result.file_id)
1155
def file_content_matches(
1156
self, source_path, target_path,
1157
source_stat=None, target_stat=None):
1158
"""Check if two files are the same in the source and target trees.
1160
This only checks that the contents of the files are the same,
1161
it does not touch anything else.
1163
:param source_path: Path of the file in the source tree
1164
:param target_path: Path of the file in the target tree
1165
:param source_stat: Optional stat value of the file in the source tree
1166
:param target_stat: Optional stat value of the file in the target tree
1167
:return: Boolean indicating whether the files have the same contents
1169
with self.lock_read():
1170
source_verifier_kind, source_verifier_data = (
1171
self.source.get_file_verifier(source_path, source_stat))
1172
target_verifier_kind, target_verifier_data = (
1173
self.target.get_file_verifier(
1174
target_path, target_stat))
1175
if source_verifier_kind == target_verifier_kind:
1176
return (source_verifier_data == target_verifier_data)
1177
# Fall back to SHA1 for now
1178
if source_verifier_kind != "SHA1":
1179
source_sha1 = self.source.get_file_sha1(
1180
source_path, source_stat)
1182
source_sha1 = source_verifier_data
1183
if target_verifier_kind != "SHA1":
1184
target_sha1 = self.target.get_file_sha1(
1185
target_path, target_stat)
1187
target_sha1 = target_verifier_data
1188
return (source_sha1 == target_sha1)
1190
def find_target_path(self, path, recurse='none'):
1191
"""Find target tree path.
1193
:param path: Path to search for (exists in source)
1194
:return: path in target, or None if there is no equivalent path.
1195
:raise NoSuchFile: If the path doesn't exist in source
1197
file_id = self.source.path2id(path)
1199
raise errors.NoSuchFile(path)
1201
return self.target.id2path(file_id, recurse=recurse)
1202
except errors.NoSuchId:
1205
def find_source_path(self, path, recurse='none'):
1206
"""Find the source tree path.
1208
:param path: Path to search for (exists in target)
1209
:return: path in source, or None if there is no equivalent path.
1210
:raise NoSuchFile: if the path doesn't exist in target
1212
file_id = self.target.path2id(path)
1214
raise errors.NoSuchFile(path)
1216
return self.source.id2path(file_id, recurse=recurse)
1217
except errors.NoSuchId:
1220
def find_target_paths(self, paths, recurse='none'):
1221
"""Find target tree paths.
1223
:param paths: Iterable over paths in target to search for
1224
:return: Dictionary mapping from source paths to paths in target , or
1225
None if there is no equivalent path.
1229
ret[path] = self.find_target_path(path, recurse=recurse)
1232
def find_source_paths(self, paths, recurse='none'):
1233
"""Find source tree paths.
1235
:param paths: Iterable over paths in target to search for
1236
:return: Dictionary mapping from target paths to paths in source, or
1237
None if there is no equivalent path.
1241
ret[path] = self.find_source_path(path, recurse=recurse)
1245
InterTree.register_optimiser(InterTree)
1248
def find_previous_paths(from_tree, to_tree, paths, recurse='none'):
1249
"""Find previous tree paths.
1251
:param from_tree: From tree
1252
:param to_tree: To tree
1253
:param paths: Iterable over paths in from_tree to search for
1254
:return: Dictionary mapping from from_tree paths to paths in to_tree, or
1255
None if there is no equivalent path.
1257
return InterTree.get(to_tree, from_tree).find_source_paths(paths, recurse=recurse)
1260
def find_previous_path(from_tree, to_tree, path, recurse='none'):
1261
"""Find previous tree path.
1263
:param from_tree: From tree
1264
:param to_tree: To tree
1265
:param path: Path to search for (exists in from_tree)
1266
:return: path in to_tree, or None if there is no equivalent path.
1267
:raise NoSuchFile: If the path doesn't exist in from_tree
1269
return InterTree.get(to_tree, from_tree).find_source_path(
1270
path, recurse=recurse)
1273
def get_canonical_path(tree, path, normalize):
1274
"""Find the canonical path of an item, ignoring case.
1276
:param tree: Tree to traverse
1277
:param path: Case-insensitive path to look up
1278
:param normalize: Function to normalize a filename for comparison
1279
:return: The canonical path
1283
bit_iter = iter(path.split("/"))
1284
for elt in bit_iter:
1285
lelt = normalize(elt)
1288
for child in tree.iter_child_entries(cur_path):
1290
if child.name == elt:
1291
# if we found an exact match, we can stop now; if
1292
# we found an approximate match we need to keep
1293
# searching because there might be an exact match
1295
new_path = osutils.pathjoin(cur_path, child.name)
1297
elif normalize(child.name) == lelt:
1298
new_path = osutils.pathjoin(cur_path, child.name)
1299
except errors.NoSuchId:
1300
# before a change is committed we can see this error...
1302
except errors.NotADirectory:
1307
# got to the end of this directory and no entries matched.
1308
# Return what matched so far, plus the rest as specified.
1309
cur_path = osutils.pathjoin(cur_path, elt, *list(bit_iter))
added
removed
breezy/tree.py
