13
13
# You should have received a copy of the GNU General Public License
14
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
15
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
17
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),
21
from cStringIO import StringIO
22
from warnings import warn
25
from bzrlib import delta
26
from bzrlib.decorators import needs_read_lock
27
from bzrlib.errors import BzrError, BzrCheckError
28
from bzrlib import errors
29
from bzrlib.inventory import Inventory
30
from bzrlib.inter import InterObject
31
from bzrlib.osutils import fingerprint_file
32
import bzrlib.revision
33
from bzrlib.trace import mutter, note
196
36
class Tree(object):
197
37
"""Abstract file tree.
199
39
There are several subclasses:
201
41
* `WorkingTree` exists as files on disk editable by the user.
203
43
* `RevisionTree` is a tree as recorded at some point in the past.
45
Trees contain an `Inventory` object, and also know how to retrieve
46
file texts mentioned in the inventory, either from a working
47
directory or from a store.
49
It is possible for trees to contain files that are not described
50
in their inventory or vice versa; for this use `filenames()`.
205
52
Trees can be compared, etc, regardless of whether they are working
206
53
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
56
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):
57
extra_trees=None, require_versioned=False):
233
58
"""Return a TreeDelta of the changes from other to this tree.
235
60
:param other: A tree to compare with.
257
81
specific_files=specific_files,
258
82
extra_trees=extra_trees,
259
83
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)
273
86
def conflicts(self):
274
87
"""Get a list of the conflicts in the tree.
276
Each conflict is an instance of breezy.conflicts.Conflict.
89
Each conflict is an instance of bzrlib.conflicts.Conflict.
278
return _mod_conflicts.ConflictList()
281
"""For trees that can have unversioned files, return all such paths."""
284
93
def get_parent_ids(self):
285
"""Get the parent ids for this tree.
94
"""Get the parent ids for this tree.
287
96
:return: a list of parent ids. [] is returned to indicate
288
97
a tree with no parents.
289
98
:raises: BzrError if the parents are not known.
291
100
raise NotImplementedError(self.get_parent_ids)
293
102
def has_filename(self, filename):
294
103
"""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):
104
raise NotImplementedError()
106
def has_id(self, file_id):
107
return self.inventory.has_id(file_id)
109
__contains__ = has_id
111
def has_or_had_id(self, file_id):
112
if file_id == self.inventory.root.file_id:
114
return self.inventory.has_id(file_id)
117
return iter(self.inventory)
119
def id2path(self, file_id):
120
return self.inventory.id2path(file_id)
122
def iter_entries_by_dir(self):
321
123
"""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)
125
This will yield each entry in the tree as a (path, entry) tuple. The
126
order that they are yielded is: the contents of a directory are
127
preceeded by the parent of a directory, and all the contents of a
128
directory are grouped together.
130
return self.inventory.iter_entries_by_dir()
132
def kind(self, file_id):
133
raise NotImplementedError("subclasses must implement kind")
135
def _get_inventory(self):
136
return self._inventory
138
def get_file_by_path(self, path):
139
return self.get_file(self._inventory.path2id(path))
141
inventory = property(_get_inventory,
142
doc="Inventory of this Tree")
144
def _check_retrieved(self, ie, f):
147
fp = fingerprint_file(f)
150
if ie.text_size != None:
151
if ie.text_size != fp['size']:
152
raise BzrError("mismatched size for file %r in %r" % (ie.file_id, self._store),
153
["inventory expects %d bytes" % ie.text_size,
154
"file is actually %d bytes" % fp['size'],
155
"store is probably damaged/corrupt"])
157
if ie.text_sha1 != fp['sha1']:
158
raise BzrError("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
159
["inventory expects %s" % ie.text_sha1,
160
"file is actually %s" % fp['sha1'],
161
"store is probably damaged/corrupt"])
164
def print_file(self, file_id):
165
"""Print file with id `file_id` to stdout."""
167
sys.stdout.write(self.get_file_text(file_id))
608
169
def lock_read(self):
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)
630
172
def unknowns(self):
631
173
"""What files are present in this tree and unknown.
633
175
:return: an iterator over the unknown files.
640
182
def filter_unversioned_files(self, paths):
641
"""Filter out paths that are versioned.
183
"""Filter out paths that are not versioned.
643
185
:return: set of paths.
645
187
# NB: we specifically *don't* call self.has_filename, because for
646
# WorkingTrees that can indicate files that exist on disk but that
188
# WorkingTrees that can indicate files that exist on disk but that
647
189
# are not versioned.
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):
190
pred = self.inventory.has_filename
191
return set((p for p in paths if not pred(p)))
195
from bzrlib.revisiontree import RevisionTree
198
class EmptyTree(Tree):
201
self._inventory = Inventory()
202
warn('EmptyTree is deprecated as of bzr 0.9 please use '
203
'repository.revision_tree instead.',
204
DeprecationWarning, stacklevel=2)
206
def get_parent_ids(self):
209
def get_symlink_target(self, file_id):
212
def has_filename(self, filename):
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 self._content_filter_stack
720
def iter_search_rules(self, path_names, pref_names=None,
721
_default_searcher=None):
722
"""Find the preferences for filenames in a tree.
724
:param path_names: an iterable of paths to find attributes for.
725
Paths are given relative to the root of the tree.
726
:param pref_names: the list of preferences to lookup - None for all
727
:param _default_searcher: private parameter to assist testing - don't use
728
:return: an iterator of tuple sequences, one per path-name.
729
See _RulesSearcher.get_items for details on the tuple sequence.
731
if _default_searcher is None:
732
_default_searcher = rules._per_user_searcher
733
searcher = self._get_rules_searcher(_default_searcher)
734
if searcher is not None:
735
if pref_names is not None:
736
for path in path_names:
737
yield searcher.get_selected_items(path, pref_names)
739
for path in path_names:
740
yield searcher.get_items(path)
742
def _get_rules_searcher(self, default_searcher):
743
"""Get the RulesSearcher for this tree given the default one."""
744
searcher = default_searcher
747
def archive(self, format, name, root='', subdir=None,
749
"""Create an archive of this tree.
751
:param format: Format name (e.g. 'tar')
752
:param name: target file name
753
:param root: Root directory name (or None)
754
:param subdir: Subdirectory to export (or None)
755
:return: Iterator over archive chunks
757
from .archive import create_archive
758
with self.lock_read():
759
return create_archive(format, self, name, root,
760
subdir, force_mtime=force_mtime)
763
def versionable_kind(cls, kind):
764
"""Check if this tree support versioning a specific file kind."""
765
return (kind in ('file', 'directory', 'symlink', 'tree-reference'))
215
def kind(self, file_id):
216
assert self._inventory[file_id].kind == "root_directory"
217
return "root_directory"
219
def list_files(self):
222
def __contains__(self, file_id):
223
return file_id in self._inventory
225
def get_file_sha1(self, file_id, path=None):
226
assert self._inventory[file_id].kind == "root_directory"
230
######################################################################
233
# TODO: Merge these two functions into a single one that can operate
234
# on either a whole tree or a set of files.
236
# TODO: Return the diff in order by filename, not by category or in
237
# random order. Can probably be done by lock-stepping through the
238
# filenames from both trees.
241
def file_status(filename, old_tree, new_tree):
242
"""Return single-letter status, old and new names for a file.
244
The complexity here is in deciding how to represent renames;
245
many complex cases are possible.
247
old_inv = old_tree.inventory
248
new_inv = new_tree.inventory
249
new_id = new_inv.path2id(filename)
250
old_id = old_inv.path2id(filename)
252
if not new_id and not old_id:
253
# easy: doesn't exist in either; not versioned at all
254
if new_tree.is_ignored(filename):
255
return 'I', None, None
257
return '?', None, None
259
# There is now a file of this name, great.
262
# There is no longer a file of this name, but we can describe
263
# what happened to the file that used to have
264
# this name. There are two possibilities: either it was
265
# deleted entirely, or renamed.
267
if new_inv.has_id(old_id):
268
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
270
return 'D', old_inv.id2path(old_id), None
272
# if the file_id is new in this revision, it is added
273
if new_id and not old_inv.has_id(new_id):
276
# if there used to be a file of this name, but that ID has now
277
# disappeared, it is deleted
278
if old_id and not new_inv.has_id(old_id):
285
def find_renames(old_inv, new_inv):
286
for file_id in old_inv:
287
if file_id not in new_inv:
289
old_name = old_inv.id2path(file_id)
290
new_name = new_inv.id2path(file_id)
291
if old_name != new_name:
292
yield (old_name, new_name)
295
def find_ids_across_trees(filenames, trees, require_versioned=True):
296
"""Find the ids corresponding to specified filenames.
298
All matches in all trees will be used, and all children of matched
299
directories will be used.
301
:param filenames: The filenames to find file_ids for
302
:param trees: The trees to find file_ids within
303
:param require_versioned: if true, all specified filenames must occur in
305
:return: a set of file ids for the specified filenames and their children.
309
specified_ids = _find_filename_ids_across_trees(filenames, trees,
311
return _find_children_across_trees(specified_ids, trees)
314
def _find_filename_ids_across_trees(filenames, trees, require_versioned):
315
"""Find the ids corresponding to specified filenames.
317
All matches in all trees will be used.
319
:param filenames: The filenames to find file_ids for
320
:param trees: The trees to find file_ids within
321
:param require_versioned: if true, all specified filenames must occur in
323
:return: a set of file ids for the specified filenames
326
interesting_ids = set()
327
for tree_path in filenames:
330
file_id = tree.inventory.path2id(tree_path)
331
if file_id is not None:
332
interesting_ids.add(file_id)
335
not_versioned.append(tree_path)
336
if len(not_versioned) > 0 and require_versioned:
337
raise errors.PathsNotVersionedError(not_versioned)
338
return interesting_ids
341
def _find_children_across_trees(specified_ids, trees):
342
"""Return a set including specified ids and their children
344
All matches in all trees will be used.
346
:param trees: The trees to find file_ids within
347
:return: a set containing all specified ids and their children
349
interesting_ids = set(specified_ids)
350
pending = interesting_ids
351
# now handle children of interesting ids
352
# we loop so that we handle all children of each id in both trees
353
while len(pending) > 0:
355
for file_id in pending:
357
if file_id not in tree:
359
entry = tree.inventory[file_id]
360
for child in getattr(entry, 'children', {}).itervalues():
361
if child.file_id not in interesting_ids:
362
new_pending.add(child.file_id)
363
interesting_ids.update(new_pending)
364
pending = new_pending
365
return interesting_ids
768
368
class InterTree(InterObject):
771
371
Its instances have methods like 'compare' and contain references to the
772
372
source and target trees these operations are to be carried out on.
774
Clients of breezy should not need to use InterTree directly, rather they
374
clients of bzrlib should not need to use InterTree directly, rather they
775
375
should use the convenience methods on Tree such as 'Tree.compare()' which
776
376
will pass through to InterTree as appropriate.
779
# Formats that will be used to test this InterTree. If both are
780
# None, this InterTree will not be tested (e.g. because a complex
782
_matching_from_tree_format = None
783
_matching_to_tree_format = None
788
def is_compatible(kls, source, target):
789
# The default implementation is naive and uses the public API, so
790
# it works for all trees.
793
def _changes_from_entries(self, source_entry, target_entry, source_path,
795
"""Generate a iter_changes tuple between source_entry and target_entry.
797
:param source_entry: An inventory entry from self.source, or None.
798
:param target_entry: An inventory entry from self.target, or None.
799
:param source_path: The path of source_entry.
800
:param target_path: The path of target_entry.
801
:return: A tuple, item 0 of which is an iter_changes result tuple, and
802
item 1 is True if there are any changes in the result tuple.
804
if source_entry is None:
805
if target_entry is None:
807
file_id = target_entry.file_id
809
file_id = source_entry.file_id
810
if source_entry is not None:
811
source_versioned = True
812
source_name = source_entry.name
813
source_parent = source_entry.parent_id
814
source_kind, source_executable, source_stat = \
815
self.source._comparison_data(source_entry, source_path)
817
source_versioned = False
821
source_executable = None
822
if target_entry is not None:
823
target_versioned = True
824
target_name = target_entry.name
825
target_parent = target_entry.parent_id
826
target_kind, target_executable, target_stat = \
827
self.target._comparison_data(target_entry, target_path)
829
target_versioned = False
833
target_executable = None
834
versioned = (source_versioned, target_versioned)
835
kind = (source_kind, target_kind)
836
changed_content = False
837
if source_kind != target_kind:
838
changed_content = True
839
elif source_kind == 'file':
840
if not self.file_content_matches(
841
source_path, target_path,
842
source_stat, target_stat):
843
changed_content = True
844
elif source_kind == 'symlink':
845
if (self.source.get_symlink_target(source_path) !=
846
self.target.get_symlink_target(target_path)):
847
changed_content = True
848
elif source_kind == 'tree-reference':
849
if (self.source.get_reference_revision(source_path)
850
!= self.target.get_reference_revision(target_path)):
851
changed_content = True
852
parent = (source_parent, target_parent)
853
name = (source_name, target_name)
854
executable = (source_executable, target_executable)
855
if (changed_content is not False or versioned[0] != versioned[1] or
856
parent[0] != parent[1] or name[0] != name[1] or
857
executable[0] != executable[1]):
862
file_id, (source_path, target_path), changed_content,
863
versioned, parent, name, kind, executable), changes
865
382
def compare(self, want_unchanged=False, specific_files=None,
866
extra_trees=None, require_versioned=False, include_root=False,
867
want_unversioned=False):
383
extra_trees=None, require_versioned=False):
868
384
"""Return the changes from source to target.
870
386
:return: A TreeDelta.
879
395
:param require_versioned: An optional boolean (defaults to False). When
880
396
supplied and True all the 'specific_files' must be versioned, or
881
397
a PathsNotVersionedError will be thrown.
882
:param want_unversioned: Scan for unversioned paths.
884
trees = (self.source,)
399
# NB: show_status depends on being able to pass in non-versioned files and
400
# report them as unknown
401
trees = (self.source, self.target)
885
402
if extra_trees is not None:
886
403
trees = trees + tuple(extra_trees)
887
with self.lock_read():
888
return delta._compare_trees(self.source, self.target, want_unchanged,
889
specific_files, include_root, extra_trees=extra_trees,
890
require_versioned=require_versioned,
891
want_unversioned=want_unversioned)
893
def iter_changes(self, include_unchanged=False,
894
specific_files=None, pb=None, extra_trees=[],
895
require_versioned=True, want_unversioned=False):
896
"""Generate an iterator of changes between trees.
899
(file_id, (path_in_source, path_in_target),
900
changed_content, versioned, parent, name, kind,
903
Changed_content is True if the file's content has changed. This
904
includes changes to its kind, and to a symlink's target.
906
versioned, parent, name, kind, executable are tuples of (from, to).
907
If a file is missing in a tree, its kind is None.
909
Iteration is done in parent-to-child order, relative to the target
912
There is no guarantee that all paths are in sorted order: the
913
requirement to expand the search due to renames may result in children
914
that should be found early being found late in the search, after
915
lexically later results have been returned.
916
:param require_versioned: Raise errors.PathsNotVersionedError if a
917
path in the specific_files list is not versioned in one of
918
source, target or extra_trees.
919
:param specific_files: An optional list of file paths to restrict the
920
comparison to. When mapping filenames to ids, all matches in all
921
trees (including optional extra_trees) are used, and all children
922
of matched directories are included. The parents in the target tree
923
of the specific files up to and including the root of the tree are
924
always evaluated for changes too.
925
:param want_unversioned: Should unversioned files be returned in the
926
output. An unversioned file is defined as one with (False, False)
927
for the versioned pair.
932
extra_trees = list(extra_trees)
933
# The ids of items we need to examine to insure delta consistency.
934
precise_file_ids = set()
935
changed_file_ids = []
936
if specific_files == []:
937
target_specific_files = []
938
source_specific_files = []
940
target_specific_files = self.target.find_related_paths_across_trees(
941
specific_files, [self.source] + extra_trees,
942
require_versioned=require_versioned)
943
source_specific_files = self.source.find_related_paths_across_trees(
944
specific_files, [self.target] + extra_trees,
945
require_versioned=require_versioned)
946
if specific_files is not None:
947
# reparented or added entries must have their parents included
948
# so that valid deltas can be created. The seen_parents set
949
# tracks the parents that we need to have.
950
# The seen_dirs set tracks directory entries we've yielded.
951
# After outputting version object in to_entries we set difference
952
# the two seen sets and start checking parents.
956
all_unversioned = sorted([(p.split('/'), p) for p in
958
if specific_files is None or
959
osutils.is_inside_any(specific_files, p)])
960
all_unversioned = deque(all_unversioned)
962
all_unversioned = deque()
964
from_entries_by_dir = list(self.source.iter_entries_by_dir(
965
specific_files=source_specific_files))
966
from_data = dict(from_entries_by_dir)
967
to_entries_by_dir = list(self.target.iter_entries_by_dir(
968
specific_files=target_specific_files))
969
path_equivs = self.find_source_paths([p for p, e in to_entries_by_dir])
970
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
972
# the unversioned path lookup only occurs on real trees - where there
973
# can be extras. So the fake_entry is solely used to look up
974
# executable it values when execute is not supported.
975
fake_entry = TreeFile()
976
for target_path, target_entry in to_entries_by_dir:
977
while (all_unversioned and
978
all_unversioned[0][0] < target_path.split('/')):
979
unversioned_path = all_unversioned.popleft()
980
target_kind, target_executable, target_stat = \
981
self.target._comparison_data(
982
fake_entry, unversioned_path[1])
984
None, (None, unversioned_path[1]), True, (False, False),
986
(None, unversioned_path[0][-1]),
988
(None, target_executable))
989
source_path = path_equivs[target_path]
990
if source_path is not None:
991
source_entry = from_data.get(source_path)
994
result, changes = self._changes_from_entries(
995
source_entry, target_entry, source_path=source_path, target_path=target_path)
996
to_paths[result.file_id] = result.path[1]
998
if result.versioned[0]:
1001
pb.update('comparing files', entry_count, num_entries)
1002
if changes or include_unchanged:
1003
if specific_files is not None:
1004
precise_file_ids.add(result.parent_id[1])
1005
changed_file_ids.append(result.file_id)
1007
# Ensure correct behaviour for reparented/added specific files.
1008
if specific_files is not None:
1009
# Record output dirs
1010
if result.kind[1] == 'directory':
1011
seen_dirs.add(result.file_id)
1012
# Record parents of reparented/added entries.
1013
if not result.versioned[0] or result.is_reparented():
1014
seen_parents.add(result.parent_id[1])
1015
while all_unversioned:
1016
# yield any trailing unversioned paths
1017
unversioned_path = all_unversioned.popleft()
1018
to_kind, to_executable, to_stat = \
1019
self.target._comparison_data(fake_entry, unversioned_path[1])
1021
None, (None, unversioned_path[1]), True, (False, False),
1023
(None, unversioned_path[0][-1]),
1025
(None, to_executable))
1026
# Yield all remaining source paths
1027
for path, from_entry in from_entries_by_dir:
1028
file_id = from_entry.file_id
1029
if file_id in to_paths:
1032
to_path = self.find_target_path(path)
1035
pb.update('comparing files', entry_count, num_entries)
1036
versioned = (True, False)
1037
parent = (from_entry.parent_id, None)
1038
name = (from_entry.name, None)
1039
from_kind, from_executable, stat_value = \
1040
self.source._comparison_data(from_entry, path)
1041
kind = (from_kind, None)
1042
executable = (from_executable, None)
1043
changed_content = from_kind is not None
1044
# the parent's path is necessarily known at this point.
1045
changed_file_ids.append(file_id)
1047
file_id, (path, to_path), changed_content, versioned, parent,
1048
name, kind, executable)
1049
changed_file_ids = set(changed_file_ids)
1050
if specific_files is not None:
1051
for result in self._handle_precise_ids(precise_file_ids,
1056
def _get_entry(tree, path):
1057
"""Get an inventory entry from a tree, with missing entries as None.
1059
If the tree raises NotImplementedError on accessing .inventory, then
1060
this is worked around using iter_entries_by_dir on just the file id
1063
:param tree: The tree to lookup the entry in.
1064
:param path: The path to look up
1066
# No inventory available.
1068
iterator = tree.iter_entries_by_dir(specific_files=[path])
1069
return next(iterator)[1]
1070
except StopIteration:
1073
def _handle_precise_ids(self, precise_file_ids, changed_file_ids,
1074
discarded_changes=None):
1075
"""Fill out a partial iter_changes to be consistent.
1077
:param precise_file_ids: The file ids of parents that were seen during
1079
:param changed_file_ids: The file ids of already emitted items.
1080
:param discarded_changes: An optional dict of precalculated
1081
iter_changes items which the partial iter_changes had not output
1083
:return: A generator of iter_changes items to output.
1085
# process parents of things that had changed under the users
1086
# requested paths to prevent incorrect paths or parent ids which
1087
# aren't in the tree.
1088
while precise_file_ids:
1089
precise_file_ids.discard(None)
1090
# Don't emit file_ids twice
1091
precise_file_ids.difference_update(changed_file_ids)
1092
if not precise_file_ids:
1094
# If the there was something at a given output path in source, we
1095
# have to include the entry from source in the delta, or we would
1096
# be putting this entry into a used path.
1098
for parent_id in precise_file_ids:
1100
paths.append(self.target.id2path(parent_id))
1101
except errors.NoSuchId:
1102
# This id has been dragged in from the source by delta
1103
# expansion and isn't present in target at all: we don't
1104
# need to check for path collisions on it.
1107
old_id = self.source.path2id(path)
1108
precise_file_ids.add(old_id)
1109
precise_file_ids.discard(None)
1110
current_ids = precise_file_ids
1111
precise_file_ids = set()
1112
# We have to emit all of precise_file_ids that have been altered.
1113
# We may have to output the children of some of those ids if any
1114
# directories have stopped being directories.
1115
for file_id in current_ids:
1117
if discarded_changes:
1118
result = discarded_changes.get(file_id)
1124
source_path = self.source.id2path(file_id)
1125
except errors.NoSuchId:
1129
source_entry = self._get_entry(
1130
self.source, source_path)
1132
target_path = self.target.id2path(file_id)
1133
except errors.NoSuchId:
1137
target_entry = self._get_entry(
1138
self.target, target_path)
1139
result, changes = self._changes_from_entries(
1140
source_entry, target_entry, source_path, target_path)
1143
# Get this parents parent to examine.
1144
new_parent_id = result.parent_id[1]
1145
precise_file_ids.add(new_parent_id)
1147
if (result.kind[0] == 'directory' and
1148
result.kind[1] != 'directory'):
1149
# This stopped being a directory, the old children have
1151
if source_entry is None:
1152
# Reusing a discarded change.
1153
source_entry = self._get_entry(
1154
self.source, result.path[0])
1155
precise_file_ids.update(
1157
for child in self.source.iter_child_entries(result.path[0]))
1158
changed_file_ids.add(result.file_id)
1161
def file_content_matches(
1162
self, source_path, target_path,
1163
source_stat=None, target_stat=None):
1164
"""Check if two files are the same in the source and target trees.
1166
This only checks that the contents of the files are the same,
1167
it does not touch anything else.
1169
:param source_path: Path of the file in the source tree
1170
:param target_path: Path of the file in the target tree
1171
:param source_stat: Optional stat value of the file in the source tree
1172
:param target_stat: Optional stat value of the file in the target tree
1173
:return: Boolean indicating whether the files have the same contents
1175
with self.lock_read():
1176
source_verifier_kind, source_verifier_data = (
1177
self.source.get_file_verifier(source_path, source_stat))
1178
target_verifier_kind, target_verifier_data = (
1179
self.target.get_file_verifier(
1180
target_path, target_stat))
1181
if source_verifier_kind == target_verifier_kind:
1182
return (source_verifier_data == target_verifier_data)
1183
# Fall back to SHA1 for now
1184
if source_verifier_kind != "SHA1":
1185
source_sha1 = self.source.get_file_sha1(
1186
source_path, source_stat)
1188
source_sha1 = source_verifier_data
1189
if target_verifier_kind != "SHA1":
1190
target_sha1 = self.target.get_file_sha1(
1191
target_path, target_stat)
1193
target_sha1 = target_verifier_data
1194
return (source_sha1 == target_sha1)
1196
def find_target_path(self, path, recurse='none'):
1197
"""Find target tree path.
1199
:param path: Path to search for (exists in source)
1200
:return: path in target, or None if there is no equivalent path.
1201
:raise NoSuchFile: If the path doesn't exist in source
1203
file_id = self.source.path2id(path)
1205
raise errors.NoSuchFile(path)
1207
return self.target.id2path(file_id, recurse=recurse)
1208
except errors.NoSuchId:
1211
def find_source_path(self, path, recurse='none'):
1212
"""Find the source tree path.
1214
:param path: Path to search for (exists in target)
1215
:return: path in source, or None if there is no equivalent path.
1216
:raise NoSuchFile: if the path doesn't exist in target
1218
file_id = self.target.path2id(path)
1220
raise errors.NoSuchFile(path)
1222
return self.source.id2path(file_id, recurse=recurse)
1223
except errors.NoSuchId:
1226
def find_target_paths(self, paths, recurse='none'):
1227
"""Find target tree paths.
1229
:param paths: Iterable over paths in target to search for
1230
:return: Dictionary mapping from source paths to paths in target , or
1231
None if there is no equivalent path.
1235
ret[path] = self.find_target_path(path, recurse=recurse)
1238
def find_source_paths(self, paths, recurse='none'):
1239
"""Find source tree paths.
1241
:param paths: Iterable over paths in target to search for
1242
:return: Dictionary mapping from target paths to paths in source, or
1243
None if there is no equivalent path.
1247
ret[path] = self.find_source_path(path, recurse=recurse)
1251
InterTree.register_optimiser(InterTree)
1254
def find_previous_paths(from_tree, to_tree, paths, recurse='none'):
1255
"""Find previous tree paths.
1257
:param from_tree: From tree
1258
:param to_tree: To tree
1259
:param paths: Iterable over paths in from_tree to search for
1260
:return: Dictionary mapping from from_tree paths to paths in to_tree, or
1261
None if there is no equivalent path.
1263
return InterTree.get(to_tree, from_tree).find_source_paths(paths, recurse=recurse)
1266
def find_previous_path(from_tree, to_tree, path, recurse='none'):
1267
"""Find previous tree path.
1269
:param from_tree: From tree
1270
:param to_tree: To tree
1271
:param path: Path to search for (exists in from_tree)
1272
:return: path in to_tree, or None if there is no equivalent path.
1273
:raise NoSuchFile: If the path doesn't exist in from_tree
1275
return InterTree.get(to_tree, from_tree).find_source_path(
1276
path, recurse=recurse)
1279
def get_canonical_path(tree, path, normalize):
1280
"""Find the canonical path of an item, ignoring case.
1282
:param tree: Tree to traverse
1283
:param path: Case-insensitive path to look up
1284
:param normalize: Function to normalize a filename for comparison
1285
:return: The canonical path
1289
bit_iter = iter(path.split("/"))
1290
for elt in bit_iter:
1291
lelt = normalize(elt)
1294
for child in tree.iter_child_entries(cur_path):
1296
if child.name == elt:
1297
# if we found an exact match, we can stop now; if
1298
# we found an approximate match we need to keep
1299
# searching because there might be an exact match
1301
new_path = osutils.pathjoin(cur_path, child.name)
1303
elif normalize(child.name) == lelt:
1304
new_path = osutils.pathjoin(cur_path, child.name)
1305
except errors.NoSuchId:
1306
# before a change is committed we can see this error...
1308
except errors.NotADirectory:
1313
# got to the end of this directory and no entries matched.
1314
# Return what matched so far, plus the rest as specified.
1315
cur_path = osutils.pathjoin(cur_path, elt, *list(bit_iter))
404
specific_file_ids = find_ids_across_trees(specific_files,
405
trees, require_versioned=require_versioned)
406
if specific_files and not specific_file_ids:
407
# All files are unversioned, so just return an empty delta
408
# _compare_trees would think we want a complete delta
409
return delta.TreeDelta()
410
return delta._compare_trees(self.source, self.target, want_unchanged,
added
removed
bzrlib/tree.py
