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 MissingNestedTree(errors.BzrError):
66
_fmt = "The nested tree for %(path)s can not be resolved."""
68
def __init__(self, path):
72
class TreeEntry(object):
73
"""An entry that implements the minimum interface used by commands.
78
def __eq__(self, other):
79
# yes, this is ugly, TODO: best practice __eq__ style.
80
return (isinstance(other, TreeEntry)
81
and other.__class__ == self.__class__)
85
def kind_character(self):
88
def is_unmodified(self, other):
89
"""Does this entry reference the same entry?
91
This is mostly the same as __eq__, but returns False
92
for entries without enough information (i.e. revision is None)
97
class TreeDirectory(TreeEntry):
98
"""See TreeEntry. This is a directory in a working tree."""
104
def kind_character(self):
108
class TreeFile(TreeEntry):
109
"""See TreeEntry. This is a regular file in a working tree."""
115
def kind_character(self):
119
class TreeLink(TreeEntry):
120
"""See TreeEntry. This is a symlink in a working tree."""
126
def kind_character(self):
130
class TreeReference(TreeEntry):
131
"""See TreeEntry. This is a reference to a nested tree in a working tree."""
135
kind = 'tree-reference'
137
def kind_character(self):
141
class TreeChange(object):
142
"""Describes the changes between the same item in two different trees."""
144
__slots__ = ['file_id', 'path', 'changed_content', 'versioned', 'parent_id',
145
'name', 'kind', 'executable', 'copied']
147
def __init__(self, file_id, path, changed_content, versioned, parent_id,
148
name, kind, executable, copied=False):
149
self.file_id = file_id
151
self.changed_content = changed_content
152
self.versioned = versioned
153
self.parent_id = parent_id
156
self.executable = executable
160
return "%s%r" % (self.__class__.__name__, self._as_tuple())
163
return len(self.__slots__)
166
return (self.file_id, self.path, self.changed_content, self.versioned,
167
self.parent_id, self.name, self.kind, self.executable, self.copied)
169
def __eq__(self, other):
170
if isinstance(other, TreeChange):
171
return self._as_tuple() == other._as_tuple()
172
if isinstance(other, tuple):
173
return self._as_tuple() == other
176
def __lt__(self, other):
177
return self._as_tuple() < other._as_tuple()
179
def meta_modified(self):
180
if self.versioned == (True, True):
181
return (self.executable[0] != self.executable[1])
184
def is_reparented(self):
185
return self.parent_id[0] != self.parent_id[1]
187
def discard_new(self):
188
return self.__class__(
189
self.file_id, (self.path[0], None), self.changed_content,
190
(self.versioned[0], None), (self.parent_id[0], None),
191
(self.name[0], None), (self.kind[0], None),
192
(self.executable[0], None),
197
"""Abstract file tree.
199
There are several subclasses:
201
* `WorkingTree` exists as files on disk editable by the user.
203
* `RevisionTree` is a tree as recorded at some point in the past.
205
Trees can be compared, etc, regardless of whether they are working
206
trees or versioned trees.
209
def supports_rename_tracking(self):
210
"""Whether this tree supports rename tracking.
212
This defaults to True, but some implementations may want to override
217
def has_versioned_directories(self):
218
"""Whether this tree can contain explicitly versioned directories.
220
This defaults to True, but some implementations may want to override
225
def supports_symlinks(self):
226
"""Does this tree support symbolic links?
228
return osutils.has_symlinks()
230
def changes_from(self, other, want_unchanged=False, specific_files=None,
231
extra_trees=None, require_versioned=False, include_root=False,
232
want_unversioned=False):
233
"""Return a TreeDelta of the changes from other to this tree.
235
:param other: A tree to compare with.
236
:param specific_files: An optional list of file paths to restrict the
237
comparison to. When mapping filenames to ids, all matches in all
238
trees (including optional extra_trees) are used, and all children of
239
matched directories are included.
240
:param want_unchanged: An optional boolean requesting the inclusion of
241
unchanged entries in the result.
242
:param extra_trees: An optional list of additional trees to use when
243
mapping the contents of specific_files (paths) to their identities.
244
:param require_versioned: An optional boolean (defaults to False). When
245
supplied and True all the 'specific_files' must be versioned, or
246
a PathsNotVersionedError will be thrown.
247
:param want_unversioned: Scan for unversioned paths.
249
The comparison will be performed by an InterTree object looked up on
252
# Martin observes that Tree.changes_from returns a TreeDelta and this
253
# may confuse people, because the class name of the returned object is
254
# a synonym of the object referenced in the method name.
255
return InterTree.get(other, self).compare(
256
want_unchanged=want_unchanged,
257
specific_files=specific_files,
258
extra_trees=extra_trees,
259
require_versioned=require_versioned,
260
include_root=include_root,
261
want_unversioned=want_unversioned,
264
def iter_changes(self, from_tree, include_unchanged=False,
265
specific_files=None, pb=None, extra_trees=None,
266
require_versioned=True, want_unversioned=False):
267
"""See InterTree.iter_changes"""
268
intertree = InterTree.get(from_tree, self)
269
return intertree.iter_changes(include_unchanged, specific_files, pb,
270
extra_trees, require_versioned,
271
want_unversioned=want_unversioned)
274
"""Get a list of the conflicts in the tree.
276
Each conflict is an instance of breezy.conflicts.Conflict.
278
return _mod_conflicts.ConflictList()
281
"""For trees that can have unversioned files, return all such paths."""
284
def get_parent_ids(self):
285
"""Get the parent ids for this tree.
287
:return: a list of parent ids. [] is returned to indicate
288
a tree with no parents.
289
:raises: BzrError if the parents are not known.
291
raise NotImplementedError(self.get_parent_ids)
293
def has_filename(self, filename):
294
"""True if the tree has given filename."""
295
raise NotImplementedError(self.has_filename)
297
def is_ignored(self, filename):
298
"""Check whether the filename is ignored by this tree.
300
:param filename: The relative filename within the tree.
301
:return: True if the filename is ignored.
305
def all_file_ids(self):
306
"""Iterate through all file ids, including ids for missing files."""
307
raise NotImplementedError(self.all_file_ids)
309
def all_versioned_paths(self):
310
"""Iterate through all paths, including paths for missing files."""
311
raise NotImplementedError(self.all_versioned_paths)
313
def id2path(self, file_id, recurse='down'):
314
"""Return the path for a file id.
318
raise NotImplementedError(self.id2path)
320
def iter_entries_by_dir(self, specific_files=None, recurse_nested=False):
321
"""Walk the tree in 'by_dir' order.
323
This will yield each entry in the tree as a (path, entry) tuple.
324
The order that they are yielded is:
326
Directories are walked in a depth-first lexicographical order,
327
however, whenever a directory is reached, all of its direct child
328
nodes are yielded in lexicographical order before yielding the
331
For example, in the tree::
341
The yield order (ignoring root) would be::
343
a, f, a/b, a/d, a/b/c, a/d/e, f/g
345
If recurse_nested is enabled then nested trees are included as if
346
they were a part of the tree. If is disabled then TreeReference
347
objects (without any children) are yielded.
349
raise NotImplementedError(self.iter_entries_by_dir)
351
def iter_child_entries(self, path):
352
"""Iterate over the children of a directory or tree reference.
354
:param path: Path of the directory
355
:raise NoSuchFile: When the path does not exist
356
:return: Iterator over entries in the directory
358
raise NotImplementedError(self.iter_child_entries)
360
def list_files(self, include_root=False, from_dir=None, recursive=True,
361
recurse_nested=False):
362
"""List all files in this tree.
364
:param include_root: Whether to include the entry for the tree root
365
:param from_dir: Directory under which to list files
366
:param recursive: Whether to list files recursively
367
:param recurse_nested: enter nested trees
368
:return: iterator over tuples of
369
(path, versioned, kind, inventory entry)
371
raise NotImplementedError(self.list_files)
373
def iter_references(self):
374
if self.supports_tree_reference():
375
for path, entry in self.iter_entries_by_dir():
376
if entry.kind == 'tree-reference':
379
def get_containing_nested_tree(self, path):
380
"""Find the nested tree that contains a path.
382
:return: tuple with (nested tree and path inside the nested tree)
384
for nested_path in self.iter_references():
386
if path.startswith(nested_path):
387
nested_tree = self.get_nested_tree(nested_path)
388
return nested_tree, path[len(nested_path):]
392
def get_nested_tree(self, path):
393
"""Open the nested tree at the specified path.
395
:param path: Path from which to resolve tree reference.
396
:return: A Tree object for the nested tree
397
:raise MissingNestedTree: If the nested tree can not be resolved
399
raise NotImplementedError(self.get_nested_tree)
401
def kind(self, path):
402
raise NotImplementedError("Tree subclass %s must implement kind"
403
% self.__class__.__name__)
405
def stored_kind(self, path):
406
"""File kind stored for this path.
408
May not match kind on disk for working trees. Always available
409
for versioned files, even when the file itself is missing.
411
return self.kind(path)
413
def path_content_summary(self, path):
414
"""Get a summary of the information about path.
416
All the attributes returned are for the canonical form, not the
417
convenient form (if content filters are in use.)
419
:param path: A relative path within the tree.
420
:return: A tuple containing kind, size, exec, sha1-or-link.
421
Kind is always present (see tree.kind()).
422
size is present if kind is file and the size of the
423
canonical form can be cheaply determined, None otherwise.
424
exec is None unless kind is file and the platform supports the 'x'
426
sha1-or-link is the link target if kind is symlink, or the sha1 if
427
it can be obtained without reading the file.
429
raise NotImplementedError(self.path_content_summary)
431
def get_reference_revision(self, path, branch=None):
432
raise NotImplementedError("Tree subclass %s must implement "
433
"get_reference_revision"
434
% self.__class__.__name__)
436
def _comparison_data(self, entry, path):
437
"""Return a tuple of kind, executable, stat_value for a file.
439
entry may be None if there is no inventory entry for the file, but
440
path must always be supplied.
442
kind is None if there is no file present (even if an inventory id is
443
present). executable is False for non-file entries.
445
raise NotImplementedError(self._comparison_data)
447
def get_file(self, path):
448
"""Return a file object for the file path in the tree.
450
raise NotImplementedError(self.get_file)
452
def get_file_with_stat(self, path):
453
"""Get a file handle and stat object for path.
455
The default implementation returns (self.get_file, None) for backwards
458
:param path: The path of the file.
459
:return: A tuple (file_handle, stat_value_or_None). If the tree has
460
no stat facility, or need for a stat cache feedback during commit,
461
it may return None for the second element of the tuple.
463
return (self.get_file(path), None)
465
def get_file_text(self, path):
466
"""Return the byte content of a file.
468
:param path: The path of the file.
470
:returns: A single byte string for the whole file.
472
with self.get_file(path) as my_file:
473
return my_file.read()
475
def get_file_lines(self, path):
476
"""Return the content of a file, as lines.
478
:param path: The path of the file.
480
return osutils.split_lines(self.get_file_text(path))
482
def get_file_verifier(self, path, stat_value=None):
483
"""Return a verifier for a file.
485
The default implementation returns a sha1.
487
:param path: The path that this file can be found at.
488
These must point to the same object.
489
:param stat_value: Optional stat value for the object
490
:return: Tuple with verifier name and verifier data
492
return ("SHA1", self.get_file_sha1(path, stat_value=stat_value))
494
def get_file_sha1(self, path, stat_value=None):
495
"""Return the SHA1 file for a file.
497
:note: callers should use get_file_verifier instead
498
where possible, as the underlying repository implementation may
499
have quicker access to a non-sha1 verifier.
501
:param path: The path that this file can be found at.
502
:param stat_value: Optional stat value for the object
504
raise NotImplementedError(self.get_file_sha1)
506
def get_file_mtime(self, path):
507
"""Return the modification time for a file.
509
:param path: The path that this file can be found at.
511
raise NotImplementedError(self.get_file_mtime)
513
def get_file_size(self, path):
514
"""Return the size of a file in bytes.
516
This applies only to regular files. If invoked on directories or
517
symlinks, it will return None.
519
raise NotImplementedError(self.get_file_size)
521
def is_executable(self, path):
522
"""Check if a file is executable.
524
:param path: The path that this file can be found at.
526
raise NotImplementedError(self.is_executable)
528
def iter_files_bytes(self, desired_files):
529
"""Iterate through file contents.
531
Files will not necessarily be returned in the order they occur in
532
desired_files. No specific order is guaranteed.
534
Yields pairs of identifier, bytes_iterator. identifier is an opaque
535
value supplied by the caller as part of desired_files. It should
536
uniquely identify the file version in the caller's context. (Examples:
537
an index number or a TreeTransform trans_id.)
539
bytes_iterator is an iterable of bytestrings for the file. The
540
kind of iterable and length of the bytestrings are unspecified, but for
541
this implementation, it is a tuple containing a single bytestring with
542
the complete text of the file.
544
:param desired_files: a list of (path, identifier) pairs
546
for path, identifier in desired_files:
547
# We wrap the string in a tuple so that we can return an iterable
548
# of bytestrings. (Technically, a bytestring is also an iterable
549
# of bytestrings, but iterating through each character is not
551
cur_file = (self.get_file_text(path),)
552
yield identifier, cur_file
554
def get_symlink_target(self, path):
555
"""Get the target for a given path.
557
It is assumed that the caller already knows that path is referencing
559
:param path: The path of the file.
560
:return: The path the symlink points to.
562
raise NotImplementedError(self.get_symlink_target)
564
def annotate_iter(self, path,
565
default_revision=_mod_revision.CURRENT_REVISION):
566
"""Return an iterator of revision_id, line tuples.
568
For working trees (and mutable trees in general), the special
569
revision_id 'current:' will be used for lines that are new in this
570
tree, e.g. uncommitted changes.
571
:param path: The file to produce an annotated version from
572
:param default_revision: For lines that don't match a basis, mark them
573
with this revision id. Not all implementations will make use of
576
raise NotImplementedError(self.annotate_iter)
578
def path2id(self, path):
579
"""Return the id for path in this tree."""
580
raise NotImplementedError(self.path2id)
582
def is_versioned(self, path):
583
"""Check whether path is versioned.
585
:param path: Path to check
588
return self.path2id(path) is not None
590
def find_related_paths_across_trees(self, paths, trees=[],
591
require_versioned=True):
592
"""Find related paths in tree corresponding to specified filenames in any
595
All matches in all trees will be used, and all children of matched
596
directories will be used.
598
:param paths: The filenames to find related paths for (if None, returns
600
:param trees: The trees to find file_ids within
601
:param require_versioned: if true, all specified filenames must occur in
603
:return: a set of paths for the specified filenames and their children
606
raise NotImplementedError(self.find_related_paths_across_trees)
609
"""Lock this tree for multiple read only operations.
611
:return: A breezy.lock.LogicalLockResult.
613
return lock.LogicalLockResult(self.unlock)
615
def revision_tree(self, revision_id):
616
"""Obtain a revision tree for the revision revision_id.
618
The intention of this method is to allow access to possibly cached
619
tree data. Implementors of this method should raise NoSuchRevision if
620
the tree is not locally available, even if they could obtain the
621
tree via a repository or some other means. Callers are responsible
622
for finding the ultimate source for a revision tree.
624
:param revision_id: The revision_id of the requested tree.
626
:raises: NoSuchRevision if the tree cannot be obtained.
628
raise errors.NoSuchRevisionInTree(self, revision_id)
631
"""What files are present in this tree and unknown.
633
:return: an iterator over the unknown files.
640
def filter_unversioned_files(self, paths):
641
"""Filter out paths that are versioned.
643
:return: set of paths.
645
# NB: we specifically *don't* call self.has_filename, because for
646
# WorkingTrees that can indicate files that exist on disk but that
648
return set(p for p in paths if not self.is_versioned(p))
650
def walkdirs(self, prefix=""):
651
"""Walk the contents of this tree from path down.
653
This yields all the data about the contents of a directory at a time.
654
After each directory has been yielded, if the caller has mutated the
655
list to exclude some directories, they are then not descended into.
657
The data yielded is of the form:
658
((directory-relpath, directory-path-from-root, directory-fileid),
659
[(relpath, basename, kind, lstat, path_from_tree_root, file_id,
660
versioned_kind), ...]),
661
- directory-relpath is the containing dirs relpath from prefix
662
- directory-path-from-root is the containing dirs path from /
663
- directory-fileid is the id of the directory if it is versioned.
664
- relpath is the relative path within the subtree being walked.
665
- basename is the basename
666
- kind is the kind of the file now. If unknonwn then the file is not
667
present within the tree - but it may be recorded as versioned. See
669
- lstat is the stat data *if* the file was statted.
670
- path_from_tree_root is the path from the root of the tree.
671
- file_id is the file_id if the entry is versioned.
672
- versioned_kind is the kind of the file as last recorded in the
673
versioning system. If 'unknown' the file is not versioned.
674
One of 'kind' and 'versioned_kind' must not be 'unknown'.
676
:param prefix: Start walking from prefix within the tree rather than
677
at the root. This allows one to walk a subtree but get paths that are
678
relative to a tree rooted higher up.
679
:return: an iterator over the directory data.
681
raise NotImplementedError(self.walkdirs)
683
def supports_content_filtering(self):
686
def _content_filter_stack(self, path=None):
687
"""The stack of content filters for a path if filtering is supported.
689
Readers will be applied in first-to-last order.
690
Writers will be applied in last-to-first order.
691
Either the path or the file-id needs to be provided.
693
:param path: path relative to the root of the tree
695
:return: the list of filters - [] if there are none
697
filter_pref_names = filters._get_registered_names()
698
if len(filter_pref_names) == 0:
700
prefs = next(self.iter_search_rules([path], filter_pref_names))
701
stk = filters._get_filter_stack_for(prefs)
702
if 'filters' in debug.debug_flags:
704
gettext("*** {0} content-filter: {1} => {2!r}").format(path, prefs, stk))
707
def _content_filter_stack_provider(self):
708
"""A function that returns a stack of ContentFilters.
710
The function takes a path (relative to the top of the tree) and a
711
file-id as parameters.
713
:return: None if content filtering is not supported by this tree.
715
if self.supports_content_filtering():
716
return lambda path, file_id: \
717
self._content_filter_stack(path)
721
def iter_search_rules(self, path_names, pref_names=None,
722
_default_searcher=None):
723
"""Find the preferences for filenames in a tree.
725
:param path_names: an iterable of paths to find attributes for.
726
Paths are given relative to the root of the tree.
727
:param pref_names: the list of preferences to lookup - None for all
728
:param _default_searcher: private parameter to assist testing - don't use
729
:return: an iterator of tuple sequences, one per path-name.
730
See _RulesSearcher.get_items for details on the tuple sequence.
732
if _default_searcher is None:
733
_default_searcher = rules._per_user_searcher
734
searcher = self._get_rules_searcher(_default_searcher)
735
if searcher is not None:
736
if pref_names is not None:
737
for path in path_names:
738
yield searcher.get_selected_items(path, pref_names)
740
for path in path_names:
741
yield searcher.get_items(path)
743
def _get_rules_searcher(self, default_searcher):
744
"""Get the RulesSearcher for this tree given the default one."""
745
searcher = default_searcher
748
def archive(self, format, name, root='', subdir=None,
750
"""Create an archive of this tree.
752
:param format: Format name (e.g. 'tar')
753
:param name: target file name
754
:param root: Root directory name (or None)
755
:param subdir: Subdirectory to export (or None)
756
:return: Iterator over archive chunks
758
from .archive import create_archive
759
with self.lock_read():
760
return create_archive(format, self, name, root,
761
subdir, force_mtime=force_mtime)
764
def versionable_kind(cls, kind):
765
"""Check if this tree support versioning a specific file kind."""
766
return (kind in ('file', 'directory', 'symlink', 'tree-reference'))
769
class InterTree(InterObject):
770
"""This class represents operations taking place between two Trees.
772
Its instances have methods like 'compare' and contain references to the
773
source and target trees these operations are to be carried out on.
775
Clients of breezy should not need to use InterTree directly, rather they
776
should use the convenience methods on Tree such as 'Tree.compare()' which
777
will pass through to InterTree as appropriate.
780
# Formats that will be used to test this InterTree. If both are
781
# None, this InterTree will not be tested (e.g. because a complex
783
_matching_from_tree_format = None
784
_matching_to_tree_format = None
789
def is_compatible(kls, source, target):
790
# The default implementation is naive and uses the public API, so
791
# it works for all trees.
794
def _changes_from_entries(self, source_entry, target_entry, source_path,
796
"""Generate a iter_changes tuple between source_entry and target_entry.
798
:param source_entry: An inventory entry from self.source, or None.
799
:param target_entry: An inventory entry from self.target, or None.
800
:param source_path: The path of source_entry.
801
:param target_path: The path of target_entry.
802
:return: A tuple, item 0 of which is an iter_changes result tuple, and
803
item 1 is True if there are any changes in the result tuple.
805
if source_entry is None:
806
if target_entry is None:
808
file_id = target_entry.file_id
810
file_id = source_entry.file_id
811
if source_entry is not None:
812
source_versioned = True
813
source_name = source_entry.name
814
source_parent = source_entry.parent_id
815
source_kind, source_executable, source_stat = \
816
self.source._comparison_data(source_entry, source_path)
818
source_versioned = False
822
source_executable = None
823
if target_entry is not None:
824
target_versioned = True
825
target_name = target_entry.name
826
target_parent = target_entry.parent_id
827
target_kind, target_executable, target_stat = \
828
self.target._comparison_data(target_entry, target_path)
830
target_versioned = False
834
target_executable = None
835
versioned = (source_versioned, target_versioned)
836
kind = (source_kind, target_kind)
837
changed_content = False
838
if source_kind != target_kind:
839
changed_content = True
840
elif source_kind == 'file':
841
if not self.file_content_matches(
842
source_path, target_path,
843
source_stat, target_stat):
844
changed_content = True
845
elif source_kind == 'symlink':
846
if (self.source.get_symlink_target(source_path) !=
847
self.target.get_symlink_target(target_path)):
848
changed_content = True
849
elif source_kind == 'tree-reference':
850
if (self.source.get_reference_revision(source_path)
851
!= self.target.get_reference_revision(target_path)):
852
changed_content = True
853
parent = (source_parent, target_parent)
854
name = (source_name, target_name)
855
executable = (source_executable, target_executable)
856
if (changed_content is not False or versioned[0] != versioned[1] or
857
parent[0] != parent[1] or name[0] != name[1] or
858
executable[0] != executable[1]):
863
file_id, (source_path, target_path), changed_content,
864
versioned, parent, name, kind, executable), changes
866
def compare(self, want_unchanged=False, specific_files=None,
867
extra_trees=None, require_versioned=False, include_root=False,
868
want_unversioned=False):
869
"""Return the changes from source to target.
871
:return: A TreeDelta.
872
:param specific_files: An optional list of file paths to restrict the
873
comparison to. When mapping filenames to ids, all matches in all
874
trees (including optional extra_trees) are used, and all children of
875
matched directories are included.
876
:param want_unchanged: An optional boolean requesting the inclusion of
877
unchanged entries in the result.
878
:param extra_trees: An optional list of additional trees to use when
879
mapping the contents of specific_files (paths) to file_ids.
880
:param require_versioned: An optional boolean (defaults to False). When
881
supplied and True all the 'specific_files' must be versioned, or
882
a PathsNotVersionedError will be thrown.
883
:param want_unversioned: Scan for unversioned paths.
885
trees = (self.source,)
886
if extra_trees is not None:
887
trees = trees + tuple(extra_trees)
888
with self.lock_read():
889
return delta._compare_trees(self.source, self.target, want_unchanged,
890
specific_files, include_root, extra_trees=extra_trees,
891
require_versioned=require_versioned,
892
want_unversioned=want_unversioned)
894
def iter_changes(self, include_unchanged=False,
895
specific_files=None, pb=None, extra_trees=[],
896
require_versioned=True, want_unversioned=False):
897
"""Generate an iterator of changes between trees.
900
(file_id, (path_in_source, path_in_target),
901
changed_content, versioned, parent, name, kind,
904
Changed_content is True if the file's content has changed. This
905
includes changes to its kind, and to a symlink's target.
907
versioned, parent, name, kind, executable are tuples of (from, to).
908
If a file is missing in a tree, its kind is None.
910
Iteration is done in parent-to-child order, relative to the target
913
There is no guarantee that all paths are in sorted order: the
914
requirement to expand the search due to renames may result in children
915
that should be found early being found late in the search, after
916
lexically later results have been returned.
917
:param require_versioned: Raise errors.PathsNotVersionedError if a
918
path in the specific_files list is not versioned in one of
919
source, target or extra_trees.
920
:param specific_files: An optional list of file paths to restrict the
921
comparison to. When mapping filenames to ids, all matches in all
922
trees (including optional extra_trees) are used, and all children
923
of matched directories are included. The parents in the target tree
924
of the specific files up to and including the root of the tree are
925
always evaluated for changes too.
926
:param want_unversioned: Should unversioned files be returned in the
927
output. An unversioned file is defined as one with (False, False)
928
for the versioned pair.
933
extra_trees = list(extra_trees)
934
# The ids of items we need to examine to insure delta consistency.
935
precise_file_ids = set()
936
changed_file_ids = []
937
if specific_files == []:
938
target_specific_files = []
939
source_specific_files = []
941
target_specific_files = self.target.find_related_paths_across_trees(
942
specific_files, [self.source] + extra_trees,
943
require_versioned=require_versioned)
944
source_specific_files = self.source.find_related_paths_across_trees(
945
specific_files, [self.target] + extra_trees,
946
require_versioned=require_versioned)
947
if specific_files is not None:
948
# reparented or added entries must have their parents included
949
# so that valid deltas can be created. The seen_parents set
950
# tracks the parents that we need to have.
951
# The seen_dirs set tracks directory entries we've yielded.
952
# After outputting version object in to_entries we set difference
953
# the two seen sets and start checking parents.
957
all_unversioned = sorted([(p.split('/'), p) for p in
959
if specific_files is None or
960
osutils.is_inside_any(specific_files, p)])
961
all_unversioned = deque(all_unversioned)
963
all_unversioned = deque()
965
from_entries_by_dir = list(self.source.iter_entries_by_dir(
966
specific_files=source_specific_files))
967
from_data = dict(from_entries_by_dir)
968
to_entries_by_dir = list(self.target.iter_entries_by_dir(
969
specific_files=target_specific_files))
970
path_equivs = find_previous_paths(
971
self.target, self.source, [p for p, e in to_entries_by_dir])
972
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
974
# the unversioned path lookup only occurs on real trees - where there
975
# can be extras. So the fake_entry is solely used to look up
976
# executable it values when execute is not supported.
977
fake_entry = TreeFile()
978
for target_path, target_entry in to_entries_by_dir:
979
while (all_unversioned and
980
all_unversioned[0][0] < target_path.split('/')):
981
unversioned_path = all_unversioned.popleft()
982
target_kind, target_executable, target_stat = \
983
self.target._comparison_data(
984
fake_entry, unversioned_path[1])
986
None, (None, unversioned_path[1]), True, (False, False),
988
(None, unversioned_path[0][-1]),
990
(None, target_executable))
991
source_path = path_equivs[target_path]
992
if source_path is not None:
993
source_entry = from_data.get(source_path)
996
result, changes = self._changes_from_entries(
997
source_entry, target_entry, source_path=source_path, target_path=target_path)
998
to_paths[result.file_id] = result.path[1]
1000
if result.versioned[0]:
1003
pb.update('comparing files', entry_count, num_entries)
1004
if changes or include_unchanged:
1005
if specific_files is not None:
1006
precise_file_ids.add(result.parent_id[1])
1007
changed_file_ids.append(result.file_id)
1009
# Ensure correct behaviour for reparented/added specific files.
1010
if specific_files is not None:
1011
# Record output dirs
1012
if result.kind[1] == 'directory':
1013
seen_dirs.add(result.file_id)
1014
# Record parents of reparented/added entries.
1015
if not result.versioned[0] or result.is_reparented():
1016
seen_parents.add(result.parent_id[1])
1017
while all_unversioned:
1018
# yield any trailing unversioned paths
1019
unversioned_path = all_unversioned.popleft()
1020
to_kind, to_executable, to_stat = \
1021
self.target._comparison_data(fake_entry, unversioned_path[1])
1023
None, (None, unversioned_path[1]), True, (False, False),
1025
(None, unversioned_path[0][-1]),
1027
(None, to_executable))
1028
# Yield all remaining source paths
1029
for path, from_entry in from_entries_by_dir:
1030
file_id = from_entry.file_id
1031
if file_id in to_paths:
1034
to_path = find_previous_path(self.source, self.target, path)
1037
pb.update('comparing files', entry_count, num_entries)
1038
versioned = (True, False)
1039
parent = (from_entry.parent_id, None)
1040
name = (from_entry.name, None)
1041
from_kind, from_executable, stat_value = \
1042
self.source._comparison_data(from_entry, path)
1043
kind = (from_kind, None)
1044
executable = (from_executable, None)
1045
changed_content = from_kind is not None
1046
# the parent's path is necessarily known at this point.
1047
changed_file_ids.append(file_id)
1049
file_id, (path, to_path), changed_content, versioned, parent,
1050
name, kind, executable)
1051
changed_file_ids = set(changed_file_ids)
1052
if specific_files is not None:
1053
for result in self._handle_precise_ids(precise_file_ids,
1058
def _get_entry(tree, path):
1059
"""Get an inventory entry from a tree, with missing entries as None.
1061
If the tree raises NotImplementedError on accessing .inventory, then
1062
this is worked around using iter_entries_by_dir on just the file id
1065
:param tree: The tree to lookup the entry in.
1066
:param path: The path to look up
1068
# No inventory available.
1070
iterator = tree.iter_entries_by_dir(specific_files=[path])
1071
return next(iterator)[1]
1072
except StopIteration:
1075
def _handle_precise_ids(self, precise_file_ids, changed_file_ids,
1076
discarded_changes=None):
1077
"""Fill out a partial iter_changes to be consistent.
1079
:param precise_file_ids: The file ids of parents that were seen during
1081
:param changed_file_ids: The file ids of already emitted items.
1082
:param discarded_changes: An optional dict of precalculated
1083
iter_changes items which the partial iter_changes had not output
1085
:return: A generator of iter_changes items to output.
1087
# process parents of things that had changed under the users
1088
# requested paths to prevent incorrect paths or parent ids which
1089
# aren't in the tree.
1090
while precise_file_ids:
1091
precise_file_ids.discard(None)
1092
# Don't emit file_ids twice
1093
precise_file_ids.difference_update(changed_file_ids)
1094
if not precise_file_ids:
1096
# If the there was something at a given output path in source, we
1097
# have to include the entry from source in the delta, or we would
1098
# be putting this entry into a used path.
1100
for parent_id in precise_file_ids:
1102
paths.append(self.target.id2path(parent_id))
1103
except errors.NoSuchId:
1104
# This id has been dragged in from the source by delta
1105
# expansion and isn't present in target at all: we don't
1106
# need to check for path collisions on it.
1109
old_id = self.source.path2id(path)
1110
precise_file_ids.add(old_id)
1111
precise_file_ids.discard(None)
1112
current_ids = precise_file_ids
1113
precise_file_ids = set()
1114
# We have to emit all of precise_file_ids that have been altered.
1115
# We may have to output the children of some of those ids if any
1116
# directories have stopped being directories.
1117
for file_id in current_ids:
1119
if discarded_changes:
1120
result = discarded_changes.get(file_id)
1126
source_path = self.source.id2path(file_id)
1127
except errors.NoSuchId:
1131
source_entry = self._get_entry(
1132
self.source, source_path)
1134
target_path = self.target.id2path(file_id)
1135
except errors.NoSuchId:
1139
target_entry = self._get_entry(
1140
self.target, target_path)
1141
result, changes = self._changes_from_entries(
1142
source_entry, target_entry, source_path, target_path)
1145
# Get this parents parent to examine.
1146
new_parent_id = result.parent_id[1]
1147
precise_file_ids.add(new_parent_id)
1149
if (result.kind[0] == 'directory' and
1150
result.kind[1] != 'directory'):
1151
# This stopped being a directory, the old children have
1153
if source_entry is None:
1154
# Reusing a discarded change.
1155
source_entry = self._get_entry(
1156
self.source, result.path[0])
1157
precise_file_ids.update(
1159
for child in self.source.iter_child_entries(result.path[0]))
1160
changed_file_ids.add(result.file_id)
1163
def file_content_matches(
1164
self, source_path, target_path,
1165
source_stat=None, target_stat=None):
1166
"""Check if two files are the same in the source and target trees.
1168
This only checks that the contents of the files are the same,
1169
it does not touch anything else.
1171
:param source_path: Path of the file in the source tree
1172
:param target_path: Path of the file in the target tree
1173
:param source_stat: Optional stat value of the file in the source tree
1174
:param target_stat: Optional stat value of the file in the target tree
1175
:return: Boolean indicating whether the files have the same contents
1177
with self.lock_read():
1178
source_verifier_kind, source_verifier_data = (
1179
self.source.get_file_verifier(source_path, source_stat))
1180
target_verifier_kind, target_verifier_data = (
1181
self.target.get_file_verifier(
1182
target_path, target_stat))
1183
if source_verifier_kind == target_verifier_kind:
1184
return (source_verifier_data == target_verifier_data)
1185
# Fall back to SHA1 for now
1186
if source_verifier_kind != "SHA1":
1187
source_sha1 = self.source.get_file_sha1(
1188
source_path, source_stat)
1190
source_sha1 = source_verifier_data
1191
if target_verifier_kind != "SHA1":
1192
target_sha1 = self.target.get_file_sha1(
1193
target_path, target_stat)
1195
target_sha1 = target_verifier_data
1196
return (source_sha1 == target_sha1)
1199
InterTree.register_optimiser(InterTree)
1202
def find_previous_paths(from_tree, to_tree, paths, recurse='none'):
1203
"""Find previous tree paths.
1205
:param from_tree: From tree
1206
:param to_tree: To tree
1207
:param paths: Iterable over paths in from_tree to search for
1208
:return: Dictionary mapping from from_tree paths to paths in to_tree, or
1209
None if there is no equivalent path.
1213
ret[path] = find_previous_path(from_tree, to_tree, path, recurse=recurse)
1217
def find_previous_path(from_tree, to_tree, path, file_id=None, recurse='none'):
1218
"""Find previous tree path.
1220
:param from_tree: From tree
1221
:param to_tree: To tree
1222
:param path: Path to search for (exists in from_tree)
1223
:return: path in to_tree, or None if there is no equivalent path.
1224
:raise NoSuchFile: If the path doesn't exist in from_tree
1227
file_id = from_tree.path2id(path)
1229
raise errors.NoSuchFile(path)
1231
return to_tree.id2path(file_id, recurse=recurse)
1232
except errors.NoSuchId:
1236
def get_canonical_path(tree, path, normalize):
1237
"""Find the canonical path of an item, ignoring case.
1239
:param tree: Tree to traverse
1240
:param path: Case-insensitive path to look up
1241
:param normalize: Function to normalize a filename for comparison
1242
:return: The canonical path
1246
bit_iter = iter(path.split("/"))
1247
for elt in bit_iter:
1248
lelt = normalize(elt)
1251
for child in tree.iter_child_entries(cur_path):
1253
if child.name == elt:
1254
# if we found an exact match, we can stop now; if
1255
# we found an approximate match we need to keep
1256
# searching because there might be an exact match
1258
new_path = osutils.pathjoin(cur_path, child.name)
1260
elif normalize(child.name) == lelt:
1261
new_path = osutils.pathjoin(cur_path, child.name)
1262
except errors.NoSuchId:
1263
# before a change is committed we can see this error...
1265
except errors.NotADirectory:
1270
# got to the end of this directory and no entries matched.
1271
# Return what matched so far, plus the rest as specified.
1272
cur_path = osutils.pathjoin(cur_path, elt, *list(bit_iter))
added
removed
breezy/tree.py
