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
50
class FileTimestampUnavailable(errors.BzrError):
52
_fmt = "The filestamp for %(path)s is not available."
56
def __init__(self, path):
60
class MissingNestedTree(errors.BzrError):
62
_fmt = "The nested tree for %(path)s can not be resolved."""
64
def __init__(self, path):
68
class TreeEntry(object):
69
"""An entry that implements the minimum interface used by commands.
74
def __eq__(self, other):
75
# yes, this is ugly, TODO: best practice __eq__ style.
76
return (isinstance(other, TreeEntry)
77
and other.__class__ == self.__class__)
81
def kind_character(self):
84
def is_unmodified(self, other):
85
"""Does this entry reference the same entry?
87
This is mostly the same as __eq__, but returns False
88
for entries without enough information (i.e. revision is None)
93
class TreeDirectory(TreeEntry):
94
"""See TreeEntry. This is a directory in a working tree."""
100
def kind_character(self):
104
class TreeFile(TreeEntry):
105
"""See TreeEntry. This is a regular file in a working tree."""
111
def kind_character(self):
115
class TreeLink(TreeEntry):
116
"""See TreeEntry. This is a symlink in a working tree."""
122
def kind_character(self):
126
class TreeReference(TreeEntry):
127
"""See TreeEntry. This is a reference to a nested tree in a working tree."""
131
kind = 'tree-reference'
133
def kind_character(self):
137
class TreeChange(object):
138
"""Describes the changes between the same item in two different trees."""
140
__slots__ = ['file_id', 'path', 'changed_content', 'versioned', 'parent_id',
141
'name', 'kind', 'executable', 'copied']
143
def __init__(self, file_id, path, changed_content, versioned, parent_id,
144
name, kind, executable, copied=False):
145
self.file_id = file_id
147
self.changed_content = changed_content
148
self.versioned = versioned
149
self.parent_id = parent_id
152
self.executable = executable
156
return "%s%r" % (self.__class__.__name__, self._as_tuple())
159
return len(self.__slots__)
162
return (self.file_id, self.path, self.changed_content, self.versioned,
163
self.parent_id, self.name, self.kind, self.executable, self.copied)
165
def __eq__(self, other):
166
if isinstance(other, TreeChange):
167
return self._as_tuple() == other._as_tuple()
168
if isinstance(other, tuple):
169
return self._as_tuple() == other
172
def __lt__(self, other):
173
return self._as_tuple() < other._as_tuple()
175
def meta_modified(self):
176
if self.versioned == (True, True):
177
return (self.executable[0] != self.executable[1])
180
def is_reparented(self):
181
return self.parent_id[0] != self.parent_id[1]
183
def discard_new(self):
184
return self.__class__(
185
self.file_id, (self.path[0], None), self.changed_content,
186
(self.versioned[0], None), (self.parent_id[0], None),
187
(self.name[0], None), (self.kind[0], None),
188
(self.executable[0], None),
193
"""Abstract file tree.
195
There are several subclasses:
197
* `WorkingTree` exists as files on disk editable by the user.
199
* `RevisionTree` is a tree as recorded at some point in the past.
201
Trees can be compared, etc, regardless of whether they are working
202
trees or versioned trees.
205
def supports_rename_tracking(self):
206
"""Whether this tree supports rename tracking.
208
This defaults to True, but some implementations may want to override
213
def has_versioned_directories(self):
214
"""Whether this tree can contain explicitly versioned directories.
216
This defaults to True, but some implementations may want to override
221
def supports_symlinks(self):
222
"""Does this tree support symbolic links?
224
return osutils.has_symlinks()
226
def changes_from(self, other, want_unchanged=False, specific_files=None,
227
extra_trees=None, require_versioned=False, include_root=False,
228
want_unversioned=False):
229
"""Return a TreeDelta of the changes from other to this tree.
231
:param other: A tree to compare with.
232
:param specific_files: An optional list of file paths to restrict the
233
comparison to. When mapping filenames to ids, all matches in all
234
trees (including optional extra_trees) are used, and all children of
235
matched directories are included.
236
:param want_unchanged: An optional boolean requesting the inclusion of
237
unchanged entries in the result.
238
:param extra_trees: An optional list of additional trees to use when
239
mapping the contents of specific_files (paths) to their identities.
240
:param require_versioned: An optional boolean (defaults to False). When
241
supplied and True all the 'specific_files' must be versioned, or
242
a PathsNotVersionedError will be thrown.
243
:param want_unversioned: Scan for unversioned paths.
245
The comparison will be performed by an InterTree object looked up on
248
# Martin observes that Tree.changes_from returns a TreeDelta and this
249
# may confuse people, because the class name of the returned object is
250
# a synonym of the object referenced in the method name.
251
return InterTree.get(other, self).compare(
252
want_unchanged=want_unchanged,
253
specific_files=specific_files,
254
extra_trees=extra_trees,
255
require_versioned=require_versioned,
256
include_root=include_root,
257
want_unversioned=want_unversioned,
260
def iter_changes(self, from_tree, include_unchanged=False,
261
specific_files=None, pb=None, extra_trees=None,
262
require_versioned=True, want_unversioned=False):
263
"""See InterTree.iter_changes"""
264
intertree = InterTree.get(from_tree, self)
265
return intertree.iter_changes(include_unchanged, specific_files, pb,
266
extra_trees, require_versioned,
267
want_unversioned=want_unversioned)
270
"""Get a list of the conflicts in the tree.
272
Each conflict is an instance of breezy.conflicts.Conflict.
274
return _mod_conflicts.ConflictList()
277
"""For trees that can have unversioned files, return all such paths."""
280
def get_parent_ids(self):
281
"""Get the parent ids for this tree.
283
:return: a list of parent ids. [] is returned to indicate
284
a tree with no parents.
285
:raises: BzrError if the parents are not known.
287
raise NotImplementedError(self.get_parent_ids)
289
def has_filename(self, filename):
290
"""True if the tree has given filename."""
291
raise NotImplementedError(self.has_filename)
293
def is_ignored(self, filename):
294
"""Check whether the filename is ignored by this tree.
296
:param filename: The relative filename within the tree.
297
:return: True if the filename is ignored.
301
def all_file_ids(self):
302
"""Iterate through all file ids, including ids for missing files."""
303
raise NotImplementedError(self.all_file_ids)
305
def all_versioned_paths(self):
306
"""Iterate through all paths, including paths for missing files."""
307
raise NotImplementedError(self.all_versioned_paths)
309
def id2path(self, file_id, recurse='down'):
310
"""Return the path for a file id.
314
raise NotImplementedError(self.id2path)
316
def iter_entries_by_dir(self, specific_files=None, recurse_nested=False):
317
"""Walk the tree in 'by_dir' order.
319
This will yield each entry in the tree as a (path, entry) tuple.
320
The order that they are yielded is:
322
Directories are walked in a depth-first lexicographical order,
323
however, whenever a directory is reached, all of its direct child
324
nodes are yielded in lexicographical order before yielding the
327
For example, in the tree::
337
The yield order (ignoring root) would be::
339
a, f, a/b, a/d, a/b/c, a/d/e, f/g
341
If recurse_nested is enabled then nested trees are included as if
342
they were a part of the tree. If is disabled then TreeReference
343
objects (without any children) are yielded.
345
raise NotImplementedError(self.iter_entries_by_dir)
347
def iter_child_entries(self, path):
348
"""Iterate over the children of a directory or tree reference.
350
:param path: Path of the directory
351
:raise NoSuchFile: When the path does not exist
352
:return: Iterator over entries in the directory
354
raise NotImplementedError(self.iter_child_entries)
356
def list_files(self, include_root=False, from_dir=None, recursive=True,
357
recurse_nested=False):
358
"""List all files in this tree.
360
:param include_root: Whether to include the entry for the tree root
361
:param from_dir: Directory under which to list files
362
:param recursive: Whether to list files recursively
363
:param recurse_nested: enter nested trees
364
:return: iterator over tuples of
365
(path, versioned, kind, inventory entry)
367
raise NotImplementedError(self.list_files)
369
def iter_references(self):
370
if self.supports_tree_reference():
371
for path, entry in self.iter_entries_by_dir():
372
if entry.kind == 'tree-reference':
375
def get_containing_nested_tree(self, path):
376
"""Find the nested tree that contains a path.
378
:return: tuple with (nested tree and path inside the nested tree)
380
for nested_path in self.iter_references():
382
if path.startswith(nested_path):
383
nested_tree = self.get_nested_tree(nested_path)
384
return nested_tree, path[len(nested_path):]
388
def get_nested_tree(self, path):
389
"""Open the nested tree at the specified path.
391
:param path: Path from which to resolve tree reference.
392
:return: A Tree object for the nested tree
393
:raise MissingNestedTree: If the nested tree can not be resolved
395
raise NotImplementedError(self.get_nested_tree)
397
def kind(self, path):
398
raise NotImplementedError("Tree subclass %s must implement kind"
399
% self.__class__.__name__)
401
def stored_kind(self, path):
402
"""File kind stored for this path.
404
May not match kind on disk for working trees. Always available
405
for versioned files, even when the file itself is missing.
407
return self.kind(path)
409
def path_content_summary(self, path):
410
"""Get a summary of the information about path.
412
All the attributes returned are for the canonical form, not the
413
convenient form (if content filters are in use.)
415
:param path: A relative path within the tree.
416
:return: A tuple containing kind, size, exec, sha1-or-link.
417
Kind is always present (see tree.kind()).
418
size is present if kind is file and the size of the
419
canonical form can be cheaply determined, None otherwise.
420
exec is None unless kind is file and the platform supports the 'x'
422
sha1-or-link is the link target if kind is symlink, or the sha1 if
423
it can be obtained without reading the file.
425
raise NotImplementedError(self.path_content_summary)
427
def get_reference_revision(self, path, branch=None):
428
raise NotImplementedError("Tree subclass %s must implement "
429
"get_reference_revision"
430
% self.__class__.__name__)
432
def _comparison_data(self, entry, path):
433
"""Return a tuple of kind, executable, stat_value for a file.
435
entry may be None if there is no inventory entry for the file, but
436
path must always be supplied.
438
kind is None if there is no file present (even if an inventory id is
439
present). executable is False for non-file entries.
441
raise NotImplementedError(self._comparison_data)
443
def get_file(self, path):
444
"""Return a file object for the file path in the tree.
446
raise NotImplementedError(self.get_file)
448
def get_file_with_stat(self, path):
449
"""Get a file handle and stat object for path.
451
The default implementation returns (self.get_file, None) for backwards
454
:param path: The path of the file.
455
:return: A tuple (file_handle, stat_value_or_None). If the tree has
456
no stat facility, or need for a stat cache feedback during commit,
457
it may return None for the second element of the tuple.
459
return (self.get_file(path), None)
461
def get_file_text(self, path):
462
"""Return the byte content of a file.
464
:param path: The path of the file.
466
:returns: A single byte string for the whole file.
468
with self.get_file(path) as my_file:
469
return my_file.read()
471
def get_file_lines(self, path):
472
"""Return the content of a file, as lines.
474
:param path: The path of the file.
476
return osutils.split_lines(self.get_file_text(path))
478
def get_file_verifier(self, path, stat_value=None):
479
"""Return a verifier for a file.
481
The default implementation returns a sha1.
483
:param path: The path that this file can be found at.
484
These must point to the same object.
485
:param stat_value: Optional stat value for the object
486
:return: Tuple with verifier name and verifier data
488
return ("SHA1", self.get_file_sha1(path, stat_value=stat_value))
490
def get_file_sha1(self, path, stat_value=None):
491
"""Return the SHA1 file for a file.
493
:note: callers should use get_file_verifier instead
494
where possible, as the underlying repository implementation may
495
have quicker access to a non-sha1 verifier.
497
:param path: The path that this file can be found at.
498
:param stat_value: Optional stat value for the object
500
raise NotImplementedError(self.get_file_sha1)
502
def get_file_mtime(self, path):
503
"""Return the modification time for a file.
505
:param path: The path that this file can be found at.
507
raise NotImplementedError(self.get_file_mtime)
509
def get_file_size(self, path):
510
"""Return the size of a file in bytes.
512
This applies only to regular files. If invoked on directories or
513
symlinks, it will return None.
515
raise NotImplementedError(self.get_file_size)
517
def is_executable(self, path):
518
"""Check if a file is executable.
520
:param path: The path that this file can be found at.
522
raise NotImplementedError(self.is_executable)
524
def iter_files_bytes(self, desired_files):
525
"""Iterate through file contents.
527
Files will not necessarily be returned in the order they occur in
528
desired_files. No specific order is guaranteed.
530
Yields pairs of identifier, bytes_iterator. identifier is an opaque
531
value supplied by the caller as part of desired_files. It should
532
uniquely identify the file version in the caller's context. (Examples:
533
an index number or a TreeTransform trans_id.)
535
bytes_iterator is an iterable of bytestrings for the file. The
536
kind of iterable and length of the bytestrings are unspecified, but for
537
this implementation, it is a tuple containing a single bytestring with
538
the complete text of the file.
540
:param desired_files: a list of (path, identifier) pairs
542
for path, identifier in desired_files:
543
# We wrap the string in a tuple so that we can return an iterable
544
# of bytestrings. (Technically, a bytestring is also an iterable
545
# of bytestrings, but iterating through each character is not
547
cur_file = (self.get_file_text(path),)
548
yield identifier, cur_file
550
def get_symlink_target(self, path):
551
"""Get the target for a given path.
553
It is assumed that the caller already knows that path is referencing
555
:param path: The path of the file.
556
:return: The path the symlink points to.
558
raise NotImplementedError(self.get_symlink_target)
560
def annotate_iter(self, path,
561
default_revision=_mod_revision.CURRENT_REVISION):
562
"""Return an iterator of revision_id, line tuples.
564
For working trees (and mutable trees in general), the special
565
revision_id 'current:' will be used for lines that are new in this
566
tree, e.g. uncommitted changes.
567
:param path: The file to produce an annotated version from
568
:param default_revision: For lines that don't match a basis, mark them
569
with this revision id. Not all implementations will make use of
572
raise NotImplementedError(self.annotate_iter)
574
def path2id(self, path):
575
"""Return the id for path in this tree."""
576
raise NotImplementedError(self.path2id)
578
def is_versioned(self, path):
579
"""Check whether path is versioned.
581
:param path: Path to check
584
return self.path2id(path) is not None
586
def find_related_paths_across_trees(self, paths, trees=[],
587
require_versioned=True):
588
"""Find related paths in tree corresponding to specified filenames in any
591
All matches in all trees will be used, and all children of matched
592
directories will be used.
594
:param paths: The filenames to find related paths for (if None, returns
596
:param trees: The trees to find file_ids within
597
:param require_versioned: if true, all specified filenames must occur in
599
:return: a set of paths for the specified filenames and their children
602
raise NotImplementedError(self.find_related_paths_across_trees)
605
"""Lock this tree for multiple read only operations.
607
:return: A breezy.lock.LogicalLockResult.
609
return lock.LogicalLockResult(self.unlock)
611
def revision_tree(self, revision_id):
612
"""Obtain a revision tree for the revision revision_id.
614
The intention of this method is to allow access to possibly cached
615
tree data. Implementors of this method should raise NoSuchRevision if
616
the tree is not locally available, even if they could obtain the
617
tree via a repository or some other means. Callers are responsible
618
for finding the ultimate source for a revision tree.
620
:param revision_id: The revision_id of the requested tree.
622
:raises: NoSuchRevision if the tree cannot be obtained.
624
raise errors.NoSuchRevisionInTree(self, revision_id)
627
"""What files are present in this tree and unknown.
629
:return: an iterator over the unknown files.
636
def filter_unversioned_files(self, paths):
637
"""Filter out paths that are versioned.
639
:return: set of paths.
641
# NB: we specifically *don't* call self.has_filename, because for
642
# WorkingTrees that can indicate files that exist on disk but that
644
return set(p for p in paths if not self.is_versioned(p))
646
def walkdirs(self, prefix=""):
647
"""Walk the contents of this tree from path down.
649
This yields all the data about the contents of a directory at a time.
650
After each directory has been yielded, if the caller has mutated the
651
list to exclude some directories, they are then not descended into.
653
The data yielded is of the form:
654
((directory-relpath, directory-path-from-root, directory-fileid),
655
[(relpath, basename, kind, lstat, path_from_tree_root, file_id,
656
versioned_kind), ...]),
657
- directory-relpath is the containing dirs relpath from prefix
658
- directory-path-from-root is the containing dirs path from /
659
- directory-fileid is the id of the directory if it is versioned.
660
- relpath is the relative path within the subtree being walked.
661
- basename is the basename
662
- kind is the kind of the file now. If unknonwn then the file is not
663
present within the tree - but it may be recorded as versioned. See
665
- lstat is the stat data *if* the file was statted.
666
- path_from_tree_root is the path from the root of the tree.
667
- file_id is the file_id if the entry is versioned.
668
- versioned_kind is the kind of the file as last recorded in the
669
versioning system. If 'unknown' the file is not versioned.
670
One of 'kind' and 'versioned_kind' must not be 'unknown'.
672
:param prefix: Start walking from prefix within the tree rather than
673
at the root. This allows one to walk a subtree but get paths that are
674
relative to a tree rooted higher up.
675
:return: an iterator over the directory data.
677
raise NotImplementedError(self.walkdirs)
679
def supports_content_filtering(self):
682
def _content_filter_stack(self, path=None):
683
"""The stack of content filters for a path if filtering is supported.
685
Readers will be applied in first-to-last order.
686
Writers will be applied in last-to-first order.
687
Either the path or the file-id needs to be provided.
689
:param path: path relative to the root of the tree
691
:return: the list of filters - [] if there are none
693
filter_pref_names = filters._get_registered_names()
694
if len(filter_pref_names) == 0:
696
prefs = next(self.iter_search_rules([path], filter_pref_names))
697
stk = filters._get_filter_stack_for(prefs)
698
if 'filters' in debug.debug_flags:
700
gettext("*** {0} content-filter: {1} => {2!r}").format(path, prefs, stk))
703
def _content_filter_stack_provider(self):
704
"""A function that returns a stack of ContentFilters.
706
The function takes a path (relative to the top of the tree) and a
707
file-id as parameters.
709
:return: None if content filtering is not supported by this tree.
711
if self.supports_content_filtering():
712
return lambda path, file_id: \
713
self._content_filter_stack(path)
717
def iter_search_rules(self, path_names, pref_names=None,
718
_default_searcher=None):
719
"""Find the preferences for filenames in a tree.
721
:param path_names: an iterable of paths to find attributes for.
722
Paths are given relative to the root of the tree.
723
:param pref_names: the list of preferences to lookup - None for all
724
:param _default_searcher: private parameter to assist testing - don't use
725
:return: an iterator of tuple sequences, one per path-name.
726
See _RulesSearcher.get_items for details on the tuple sequence.
728
if _default_searcher is None:
729
_default_searcher = rules._per_user_searcher
730
searcher = self._get_rules_searcher(_default_searcher)
731
if searcher is not None:
732
if pref_names is not None:
733
for path in path_names:
734
yield searcher.get_selected_items(path, pref_names)
736
for path in path_names:
737
yield searcher.get_items(path)
739
def _get_rules_searcher(self, default_searcher):
740
"""Get the RulesSearcher for this tree given the default one."""
741
searcher = default_searcher
744
def archive(self, format, name, root='', subdir=None,
746
"""Create an archive of this tree.
748
:param format: Format name (e.g. 'tar')
749
:param name: target file name
750
:param root: Root directory name (or None)
751
:param subdir: Subdirectory to export (or None)
752
:return: Iterator over archive chunks
754
from .archive import create_archive
755
with self.lock_read():
756
return create_archive(format, self, name, root,
757
subdir, force_mtime=force_mtime)
760
def versionable_kind(cls, kind):
761
"""Check if this tree support versioning a specific file kind."""
762
return (kind in ('file', 'directory', 'symlink', 'tree-reference'))
765
class InterTree(InterObject):
766
"""This class represents operations taking place between two Trees.
768
Its instances have methods like 'compare' and contain references to the
769
source and target trees these operations are to be carried out on.
771
Clients of breezy should not need to use InterTree directly, rather they
772
should use the convenience methods on Tree such as 'Tree.compare()' which
773
will pass through to InterTree as appropriate.
776
# Formats that will be used to test this InterTree. If both are
777
# None, this InterTree will not be tested (e.g. because a complex
779
_matching_from_tree_format = None
780
_matching_to_tree_format = None
785
def is_compatible(kls, source, target):
786
# The default implementation is naive and uses the public API, so
787
# it works for all trees.
790
def _changes_from_entries(self, source_entry, target_entry, source_path,
792
"""Generate a iter_changes tuple between source_entry and target_entry.
794
:param source_entry: An inventory entry from self.source, or None.
795
:param target_entry: An inventory entry from self.target, or None.
796
:param source_path: The path of source_entry.
797
:param target_path: The path of target_entry.
798
:return: A tuple, item 0 of which is an iter_changes result tuple, and
799
item 1 is True if there are any changes in the result tuple.
801
if source_entry is None:
802
if target_entry is None:
804
file_id = target_entry.file_id
806
file_id = source_entry.file_id
807
if source_entry is not None:
808
source_versioned = True
809
source_name = source_entry.name
810
source_parent = source_entry.parent_id
811
source_kind, source_executable, source_stat = \
812
self.source._comparison_data(source_entry, source_path)
814
source_versioned = False
818
source_executable = None
819
if target_entry is not None:
820
target_versioned = True
821
target_name = target_entry.name
822
target_parent = target_entry.parent_id
823
target_kind, target_executable, target_stat = \
824
self.target._comparison_data(target_entry, target_path)
826
target_versioned = False
830
target_executable = None
831
versioned = (source_versioned, target_versioned)
832
kind = (source_kind, target_kind)
833
changed_content = False
834
if source_kind != target_kind:
835
changed_content = True
836
elif source_kind == 'file':
837
if not self.file_content_matches(
838
source_path, target_path,
839
source_stat, target_stat):
840
changed_content = True
841
elif source_kind == 'symlink':
842
if (self.source.get_symlink_target(source_path) !=
843
self.target.get_symlink_target(target_path)):
844
changed_content = True
845
elif source_kind == 'tree-reference':
846
if (self.source.get_reference_revision(source_path)
847
!= self.target.get_reference_revision(target_path)):
848
changed_content = True
849
parent = (source_parent, target_parent)
850
name = (source_name, target_name)
851
executable = (source_executable, target_executable)
852
if (changed_content is not False or versioned[0] != versioned[1] or
853
parent[0] != parent[1] or name[0] != name[1] or
854
executable[0] != executable[1]):
859
file_id, (source_path, target_path), changed_content,
860
versioned, parent, name, kind, executable), changes
862
def compare(self, want_unchanged=False, specific_files=None,
863
extra_trees=None, require_versioned=False, include_root=False,
864
want_unversioned=False):
865
"""Return the changes from source to target.
867
:return: A TreeDelta.
868
:param specific_files: An optional list of file paths to restrict the
869
comparison to. When mapping filenames to ids, all matches in all
870
trees (including optional extra_trees) are used, and all children of
871
matched directories are included.
872
:param want_unchanged: An optional boolean requesting the inclusion of
873
unchanged entries in the result.
874
:param extra_trees: An optional list of additional trees to use when
875
mapping the contents of specific_files (paths) to file_ids.
876
:param require_versioned: An optional boolean (defaults to False). When
877
supplied and True all the 'specific_files' must be versioned, or
878
a PathsNotVersionedError will be thrown.
879
:param want_unversioned: Scan for unversioned paths.
881
trees = (self.source,)
882
if extra_trees is not None:
883
trees = trees + tuple(extra_trees)
884
with self.lock_read():
885
return delta._compare_trees(self.source, self.target, want_unchanged,
886
specific_files, include_root, extra_trees=extra_trees,
887
require_versioned=require_versioned,
888
want_unversioned=want_unversioned)
890
def iter_changes(self, include_unchanged=False,
891
specific_files=None, pb=None, extra_trees=[],
892
require_versioned=True, want_unversioned=False):
893
"""Generate an iterator of changes between trees.
896
(file_id, (path_in_source, path_in_target),
897
changed_content, versioned, parent, name, kind,
900
Changed_content is True if the file's content has changed. This
901
includes changes to its kind, and to a symlink's target.
903
versioned, parent, name, kind, executable are tuples of (from, to).
904
If a file is missing in a tree, its kind is None.
906
Iteration is done in parent-to-child order, relative to the target
909
There is no guarantee that all paths are in sorted order: the
910
requirement to expand the search due to renames may result in children
911
that should be found early being found late in the search, after
912
lexically later results have been returned.
913
:param require_versioned: Raise errors.PathsNotVersionedError if a
914
path in the specific_files list is not versioned in one of
915
source, target or extra_trees.
916
:param specific_files: An optional list of file paths to restrict the
917
comparison to. When mapping filenames to ids, all matches in all
918
trees (including optional extra_trees) are used, and all children
919
of matched directories are included. The parents in the target tree
920
of the specific files up to and including the root of the tree are
921
always evaluated for changes too.
922
:param want_unversioned: Should unversioned files be returned in the
923
output. An unversioned file is defined as one with (False, False)
924
for the versioned pair.
929
extra_trees = list(extra_trees)
930
# The ids of items we need to examine to insure delta consistency.
931
precise_file_ids = set()
932
changed_file_ids = []
933
if specific_files == []:
934
target_specific_files = []
935
source_specific_files = []
937
target_specific_files = self.target.find_related_paths_across_trees(
938
specific_files, [self.source] + extra_trees,
939
require_versioned=require_versioned)
940
source_specific_files = self.source.find_related_paths_across_trees(
941
specific_files, [self.target] + extra_trees,
942
require_versioned=require_versioned)
943
if specific_files is not None:
944
# reparented or added entries must have their parents included
945
# so that valid deltas can be created. The seen_parents set
946
# tracks the parents that we need to have.
947
# The seen_dirs set tracks directory entries we've yielded.
948
# After outputting version object in to_entries we set difference
949
# the two seen sets and start checking parents.
953
all_unversioned = sorted([(p.split('/'), p) for p in
955
if specific_files is None or
956
osutils.is_inside_any(specific_files, p)])
957
all_unversioned = deque(all_unversioned)
959
all_unversioned = deque()
961
from_entries_by_dir = list(self.source.iter_entries_by_dir(
962
specific_files=source_specific_files))
963
from_data = dict(from_entries_by_dir)
964
to_entries_by_dir = list(self.target.iter_entries_by_dir(
965
specific_files=target_specific_files))
966
path_equivs = find_previous_paths(
967
self.target, self.source, [p for p, e in to_entries_by_dir])
968
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
970
# the unversioned path lookup only occurs on real trees - where there
971
# can be extras. So the fake_entry is solely used to look up
972
# executable it values when execute is not supported.
973
fake_entry = TreeFile()
974
for target_path, target_entry in to_entries_by_dir:
975
while (all_unversioned and
976
all_unversioned[0][0] < target_path.split('/')):
977
unversioned_path = all_unversioned.popleft()
978
target_kind, target_executable, target_stat = \
979
self.target._comparison_data(
980
fake_entry, unversioned_path[1])
982
None, (None, unversioned_path[1]), True, (False, False),
984
(None, unversioned_path[0][-1]),
986
(None, target_executable))
987
source_path = path_equivs[target_path]
988
if source_path is not None:
989
source_entry = from_data.get(source_path)
992
result, changes = self._changes_from_entries(
993
source_entry, target_entry, source_path=source_path, target_path=target_path)
994
to_paths[result.file_id] = result.path[1]
996
if result.versioned[0]:
999
pb.update('comparing files', entry_count, num_entries)
1000
if changes or include_unchanged:
1001
if specific_files is not None:
1002
precise_file_ids.add(result.parent_id[1])
1003
changed_file_ids.append(result.file_id)
1005
# Ensure correct behaviour for reparented/added specific files.
1006
if specific_files is not None:
1007
# Record output dirs
1008
if result.kind[1] == 'directory':
1009
seen_dirs.add(result.file_id)
1010
# Record parents of reparented/added entries.
1011
if not result.versioned[0] or result.is_reparented():
1012
seen_parents.add(result.parent_id[1])
1013
while all_unversioned:
1014
# yield any trailing unversioned paths
1015
unversioned_path = all_unversioned.popleft()
1016
to_kind, to_executable, to_stat = \
1017
self.target._comparison_data(fake_entry, unversioned_path[1])
1019
None, (None, unversioned_path[1]), True, (False, False),
1021
(None, unversioned_path[0][-1]),
1023
(None, to_executable))
1024
# Yield all remaining source paths
1025
for path, from_entry in from_entries_by_dir:
1026
file_id = from_entry.file_id
1027
if file_id in to_paths:
1030
to_path = find_previous_path(self.source, self.target, path)
1033
pb.update('comparing files', entry_count, num_entries)
1034
versioned = (True, False)
1035
parent = (from_entry.parent_id, None)
1036
name = (from_entry.name, None)
1037
from_kind, from_executable, stat_value = \
1038
self.source._comparison_data(from_entry, path)
1039
kind = (from_kind, None)
1040
executable = (from_executable, None)
1041
changed_content = from_kind is not None
1042
# the parent's path is necessarily known at this point.
1043
changed_file_ids.append(file_id)
1045
file_id, (path, to_path), changed_content, versioned, parent,
1046
name, kind, executable)
1047
changed_file_ids = set(changed_file_ids)
1048
if specific_files is not None:
1049
for result in self._handle_precise_ids(precise_file_ids,
1054
def _get_entry(tree, path):
1055
"""Get an inventory entry from a tree, with missing entries as None.
1057
If the tree raises NotImplementedError on accessing .inventory, then
1058
this is worked around using iter_entries_by_dir on just the file id
1061
:param tree: The tree to lookup the entry in.
1062
:param path: The path to look up
1064
# No inventory available.
1066
iterator = tree.iter_entries_by_dir(specific_files=[path])
1067
return next(iterator)[1]
1068
except StopIteration:
1071
def _handle_precise_ids(self, precise_file_ids, changed_file_ids,
1072
discarded_changes=None):
1073
"""Fill out a partial iter_changes to be consistent.
1075
:param precise_file_ids: The file ids of parents that were seen during
1077
:param changed_file_ids: The file ids of already emitted items.
1078
:param discarded_changes: An optional dict of precalculated
1079
iter_changes items which the partial iter_changes had not output
1081
:return: A generator of iter_changes items to output.
1083
# process parents of things that had changed under the users
1084
# requested paths to prevent incorrect paths or parent ids which
1085
# aren't in the tree.
1086
while precise_file_ids:
1087
precise_file_ids.discard(None)
1088
# Don't emit file_ids twice
1089
precise_file_ids.difference_update(changed_file_ids)
1090
if not precise_file_ids:
1092
# If the there was something at a given output path in source, we
1093
# have to include the entry from source in the delta, or we would
1094
# be putting this entry into a used path.
1096
for parent_id in precise_file_ids:
1098
paths.append(self.target.id2path(parent_id))
1099
except errors.NoSuchId:
1100
# This id has been dragged in from the source by delta
1101
# expansion and isn't present in target at all: we don't
1102
# need to check for path collisions on it.
1105
old_id = self.source.path2id(path)
1106
precise_file_ids.add(old_id)
1107
precise_file_ids.discard(None)
1108
current_ids = precise_file_ids
1109
precise_file_ids = set()
1110
# We have to emit all of precise_file_ids that have been altered.
1111
# We may have to output the children of some of those ids if any
1112
# directories have stopped being directories.
1113
for file_id in current_ids:
1115
if discarded_changes:
1116
result = discarded_changes.get(file_id)
1122
source_path = self.source.id2path(file_id)
1123
except errors.NoSuchId:
1127
source_entry = self._get_entry(
1128
self.source, source_path)
1130
target_path = self.target.id2path(file_id)
1131
except errors.NoSuchId:
1135
target_entry = self._get_entry(
1136
self.target, target_path)
1137
result, changes = self._changes_from_entries(
1138
source_entry, target_entry, source_path, target_path)
1141
# Get this parents parent to examine.
1142
new_parent_id = result.parent_id[1]
1143
precise_file_ids.add(new_parent_id)
1145
if (result.kind[0] == 'directory' and
1146
result.kind[1] != 'directory'):
1147
# This stopped being a directory, the old children have
1149
if source_entry is None:
1150
# Reusing a discarded change.
1151
source_entry = self._get_entry(
1152
self.source, result.path[0])
1153
precise_file_ids.update(
1155
for child in self.source.iter_child_entries(result.path[0]))
1156
changed_file_ids.add(result.file_id)
1159
def file_content_matches(
1160
self, source_path, target_path,
1161
source_stat=None, target_stat=None):
1162
"""Check if two files are the same in the source and target trees.
1164
This only checks that the contents of the files are the same,
1165
it does not touch anything else.
1167
:param source_path: Path of the file in the source tree
1168
:param target_path: Path of the file in the target tree
1169
:param source_stat: Optional stat value of the file in the source tree
1170
:param target_stat: Optional stat value of the file in the target tree
1171
:return: Boolean indicating whether the files have the same contents
1173
with self.lock_read():
1174
source_verifier_kind, source_verifier_data = (
1175
self.source.get_file_verifier(source_path, source_stat))
1176
target_verifier_kind, target_verifier_data = (
1177
self.target.get_file_verifier(
1178
target_path, target_stat))
1179
if source_verifier_kind == target_verifier_kind:
1180
return (source_verifier_data == target_verifier_data)
1181
# Fall back to SHA1 for now
1182
if source_verifier_kind != "SHA1":
1183
source_sha1 = self.source.get_file_sha1(
1184
source_path, source_stat)
1186
source_sha1 = source_verifier_data
1187
if target_verifier_kind != "SHA1":
1188
target_sha1 = self.target.get_file_sha1(
1189
target_path, target_stat)
1191
target_sha1 = target_verifier_data
1192
return (source_sha1 == target_sha1)
1195
InterTree.register_optimiser(InterTree)
1198
def find_previous_paths(from_tree, to_tree, paths, recurse='none'):
1199
"""Find previous tree paths.
1201
:param from_tree: From tree
1202
:param to_tree: To tree
1203
:param paths: Iterable over paths in from_tree to search for
1204
:return: Dictionary mapping from from_tree paths to paths in to_tree, or
1205
None if there is no equivalent path.
1209
ret[path] = find_previous_path(from_tree, to_tree, path, recurse=recurse)
1213
def find_previous_path(from_tree, to_tree, path, file_id=None, recurse='none'):
1214
"""Find previous tree path.
1216
:param from_tree: From tree
1217
:param to_tree: To tree
1218
:param path: Path to search for (exists in from_tree)
1219
:return: path in to_tree, or None if there is no equivalent path.
1220
:raise NoSuchFile: If the path doesn't exist in from_tree
1223
file_id = from_tree.path2id(path)
1225
raise errors.NoSuchFile(path)
1227
return to_tree.id2path(file_id, recurse=recurse)
1228
except errors.NoSuchId:
1232
def get_canonical_path(tree, path, normalize):
1233
"""Find the canonical path of an item, ignoring case.
1235
:param tree: Tree to traverse
1236
:param path: Case-insensitive path to look up
1237
:param normalize: Function to normalize a filename for comparison
1238
:return: The canonical path
1242
bit_iter = iter(path.split("/"))
1243
for elt in bit_iter:
1244
lelt = normalize(elt)
1247
for child in tree.iter_child_entries(cur_path):
1249
if child.name == elt:
1250
# if we found an exact match, we can stop now; if
1251
# we found an approximate match we need to keep
1252
# searching because there might be an exact match
1254
new_path = osutils.pathjoin(cur_path, child.name)
1256
elif normalize(child.name) == lelt:
1257
new_path = osutils.pathjoin(cur_path, child.name)
1258
except errors.NoSuchId:
1259
# before a change is committed we can see this error...
1261
except errors.NotADirectory:
1266
# got to the end of this directory and no entries matched.
1267
# Return what matched so far, plus the rest as specified.
1268
cur_path = osutils.pathjoin(cur_path, elt, *list(bit_iter))
added
removed
breezy/tree.py
