17
17
"""Tree classes, representing directory at point in time.
21
from collections import deque
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(), """
25
31
conflicts as _mod_conflicts,
30
35
revision as _mod_revision,
33
from bzrlib.decorators import needs_read_lock
34
from bzrlib.errors import BzrError, NoSuchId
35
from bzrlib import errors
36
from bzrlib.inventory import InventoryFile
37
from bzrlib.inter import InterObject
38
from bzrlib.osutils import fingerprint_file
39
from bzrlib.symbol_versioning import deprecated_function, deprecated_in
40
from bzrlib.trace import note
39
from breezy.i18n import gettext
47
from .inter import InterObject
54
class FileTimestampUnavailable(errors.BzrError):
56
_fmt = "The filestamp for %(path)s is not available."
60
def __init__(self, path):
64
class TreeEntry(object):
65
"""An entry that implements the minimum interface used by commands.
68
def __eq__(self, other):
69
# yes, this is ugly, TODO: best practice __eq__ style.
70
return (isinstance(other, TreeEntry)
71
and other.__class__ == self.__class__)
75
def kind_character(self):
79
class TreeDirectory(TreeEntry):
80
"""See TreeEntry. This is a directory in a working tree."""
84
def kind_character(self):
88
class TreeFile(TreeEntry):
89
"""See TreeEntry. This is a regular file in a working tree."""
93
def kind_character(self):
97
class TreeLink(TreeEntry):
98
"""See TreeEntry. This is a symlink in a working tree."""
102
def kind_character(self):
106
class TreeReference(TreeEntry):
107
"""See TreeEntry. This is a reference to a nested tree in a working tree."""
109
kind = 'tree-reference'
111
def kind_character(self):
43
115
class Tree(object):
50
122
* `RevisionTree` is a tree as recorded at some point in the past.
52
Trees contain an `Inventory` object, and also know how to retrieve
53
file texts mentioned in the inventory, either from a working
54
directory or from a store.
56
It is possible for trees to contain files that are not described
57
in their inventory or vice versa; for this use `filenames()`.
59
124
Trees can be compared, etc, regardless of whether they are working
60
125
trees or versioned trees.
128
def supports_rename_tracking(self):
129
"""Whether this tree supports rename tracking.
131
This defaults to True, but some implementations may want to override
136
def has_versioned_directories(self):
137
"""Whether this tree can contain explicitly versioned directories.
139
This defaults to True, but some implementations may want to override
144
def supports_symlinks(self):
145
"""Does this tree support symbolic links?
147
return osutils.has_symlinks()
63
149
def changes_from(self, other, want_unchanged=False, specific_files=None,
64
extra_trees=None, require_versioned=False, include_root=False,
65
want_unversioned=False):
150
extra_trees=None, require_versioned=False, include_root=False,
151
want_unversioned=False):
66
152
"""Return a TreeDelta of the changes from other to this tree.
68
154
:param other: A tree to compare with.
147
return iter(self.inventory)
149
224
def all_file_ids(self):
150
225
"""Iterate through all file ids, including ids for missing files."""
151
return set(self.inventory)
226
raise NotImplementedError(self.all_file_ids)
228
def all_versioned_paths(self):
229
"""Iterate through all paths, including paths for missing files."""
230
raise NotImplementedError(self.all_versioned_paths)
153
232
def id2path(self, file_id):
154
233
"""Return the path for a file id.
156
235
:raises NoSuchId:
158
return self.inventory.id2path(file_id)
160
def is_control_filename(self, filename):
161
"""True if filename is the name of a control file in this tree.
163
:param filename: A filename within the tree. This is a relative path
164
from the root of this tree.
166
This is true IF and ONLY IF the filename is part of the meta data
167
that bzr controls in this tree. I.E. a random .bzr directory placed
168
on disk will not be a control file for this tree.
170
return self.bzrdir.is_control_filename(filename)
173
def iter_entries_by_dir(self, specific_file_ids=None, yield_parents=False):
237
raise NotImplementedError(self.id2path)
239
def iter_entries_by_dir(self, specific_files=None):
174
240
"""Walk the tree in 'by_dir' order.
176
242
This will yield each entry in the tree as a (path, entry) tuple.
194
260
The yield order (ignoring root) would be::
195
262
a, f, a/b, a/d, a/b/c, a/d/e, f/g
197
:param yield_parents: If True, yield the parents from the root leading
198
down to specific_file_ids that have been requested. This has no
199
impact if specific_file_ids is None.
201
return self.inventory.iter_entries_by_dir(
202
specific_file_ids=specific_file_ids, yield_parents=yield_parents)
264
raise NotImplementedError(self.iter_entries_by_dir)
266
def iter_child_entries(self, path):
267
"""Iterate over the children of a directory or tree reference.
269
:param path: Path of the directory
270
:raise NoSuchFile: When the path does not exist
271
:return: Iterator over entries in the directory
273
raise NotImplementedError(self.iter_child_entries)
275
def list_files(self, include_root=False, from_dir=None, recursive=True):
276
"""List all files in this tree.
278
:param include_root: Whether to include the entry for the tree root
279
:param from_dir: Directory under which to list files
280
:param recursive: Whether to list files recursively
281
:return: iterator over tuples of
282
(path, versioned, kind, inventory entry)
284
raise NotImplementedError(self.list_files)
204
286
def iter_references(self):
205
287
if self.supports_tree_reference():
254
336
raise NotImplementedError(self._comparison_data)
256
def _file_size(self, entry, stat_value):
257
raise NotImplementedError(self._file_size)
259
def _get_inventory(self):
260
return self._inventory
262
def get_file(self, file_id, path=None):
263
"""Return a file object for the file file_id in the tree.
265
If both file_id and path are defined, it is implementation defined as
266
to which one is used.
338
def get_file(self, path):
339
"""Return a file object for the file path in the tree.
268
341
raise NotImplementedError(self.get_file)
270
def get_file_with_stat(self, file_id, path=None):
271
"""Get a file handle and stat object for file_id.
343
def get_file_with_stat(self, path):
344
"""Get a file handle and stat object for path.
273
346
The default implementation returns (self.get_file, None) for backwards
276
:param file_id: The file id to read.
277
:param path: The path of the file, if it is known.
349
:param path: The path of the file.
278
350
:return: A tuple (file_handle, stat_value_or_None). If the tree has
279
351
no stat facility, or need for a stat cache feedback during commit,
280
352
it may return None for the second element of the tuple.
282
return (self.get_file(file_id, path), None)
354
return (self.get_file(path), None)
284
def get_file_text(self, file_id, path=None):
356
def get_file_text(self, path):
285
357
"""Return the byte content of a file.
287
:param file_id: The file_id of the file.
288
359
:param path: The path of the file.
289
If both file_id and path are supplied, an implementation may use
361
:returns: A single byte string for the whole file.
292
my_file = self.get_file(file_id, path)
363
with self.get_file(path) as my_file:
294
364
return my_file.read()
298
def get_file_lines(self, file_id, path=None):
366
def get_file_lines(self, path):
299
367
"""Return the content of a file, as lines.
301
:param file_id: The file_id of the file.
302
369
:param path: The path of the file.
303
If both file_id and path are supplied, an implementation may use
306
return osutils.split_lines(self.get_file_text(file_id, path))
308
def get_file_mtime(self, file_id, path=None):
371
return osutils.split_lines(self.get_file_text(path))
373
def get_file_verifier(self, path, stat_value=None):
374
"""Return a verifier for a file.
376
The default implementation returns a sha1.
378
:param path: The path that this file can be found at.
379
These must point to the same object.
380
:param stat_value: Optional stat value for the object
381
:return: Tuple with verifier name and verifier data
383
return ("SHA1", self.get_file_sha1(path, stat_value=stat_value))
385
def get_file_sha1(self, path, stat_value=None):
386
"""Return the SHA1 file for a file.
388
:note: callers should use get_file_verifier instead
389
where possible, as the underlying repository implementation may
390
have quicker access to a non-sha1 verifier.
392
:param path: The path that this file can be found at.
393
:param stat_value: Optional stat value for the object
395
raise NotImplementedError(self.get_file_sha1)
397
def get_file_mtime(self, path):
309
398
"""Return the modification time for a file.
311
:param file_id: The handle for this file.
312
400
:param path: The path that this file can be found at.
313
These must point to the same object.
315
402
raise NotImplementedError(self.get_file_mtime)
317
def get_file_size(self, file_id):
404
def get_file_size(self, path):
318
405
"""Return the size of a file in bytes.
320
407
This applies only to regular files. If invoked on directories or
321
408
symlinks, it will return None.
322
:param file_id: The file-id of the file
324
410
raise NotImplementedError(self.get_file_size)
326
def get_file_by_path(self, path):
327
return self.get_file(self._inventory.path2id(path), path)
412
def is_executable(self, path):
413
"""Check if a file is executable.
415
:param path: The path that this file can be found at.
417
raise NotImplementedError(self.is_executable)
329
419
def iter_files_bytes(self, desired_files):
330
420
"""Iterate through file contents.
342
432
this implementation, it is a tuple containing a single bytestring with
343
433
the complete text of the file.
345
:param desired_files: a list of (file_id, identifier) pairs
435
:param desired_files: a list of (path, identifier) pairs
347
for file_id, identifier in desired_files:
437
for path, identifier in desired_files:
348
438
# We wrap the string in a tuple so that we can return an iterable
349
439
# of bytestrings. (Technically, a bytestring is also an iterable
350
440
# of bytestrings, but iterating through each character is not
352
cur_file = (self.get_file_text(file_id),)
442
cur_file = (self.get_file_text(path),)
353
443
yield identifier, cur_file
355
def get_symlink_target(self, file_id):
356
"""Get the target for a given file_id.
445
def get_symlink_target(self, path):
446
"""Get the target for a given path.
358
It is assumed that the caller already knows that file_id is referencing
448
It is assumed that the caller already knows that path is referencing
360
:param file_id: Handle for the symlink entry.
450
:param path: The path of the file.
361
451
:return: The path the symlink points to.
363
453
raise NotImplementedError(self.get_symlink_target)
365
def get_canonical_inventory_paths(self, paths):
366
"""Like get_canonical_inventory_path() but works on multiple items.
368
:param paths: A sequence of paths relative to the root of the tree.
369
:return: A list of paths, with each item the corresponding input path
370
adjusted to account for existing elements that match case
373
return list(self._yield_canonical_inventory_paths(paths))
375
def get_canonical_inventory_path(self, path):
376
"""Returns the first inventory item that case-insensitively matches path.
378
If a path matches exactly, it is returned. If no path matches exactly
379
but more than one path matches case-insensitively, it is implementation
380
defined which is returned.
382
If no path matches case-insensitively, the input path is returned, but
383
with as many path entries that do exist changed to their canonical
386
If you need to resolve many names from the same tree, you should
387
use get_canonical_inventory_paths() to avoid O(N) behaviour.
389
:param path: A paths relative to the root of the tree.
390
:return: The input path adjusted to account for existing elements
391
that match case insensitively.
393
return self._yield_canonical_inventory_paths([path]).next()
395
def _yield_canonical_inventory_paths(self, paths):
397
# First, if the path as specified exists exactly, just use it.
398
if self.path2id(path) is not None:
402
cur_id = self.get_root_id()
404
bit_iter = iter(path.split("/"))
408
for child in self.iter_children(cur_id):
410
# XXX: it seem like if the child is known to be in the
411
# tree, we shouldn't need to go from its id back to
412
# its path -- mbp 2010-02-11
414
# XXX: it seems like we could be more efficient
415
# by just directly looking up the original name and
416
# only then searching all children; also by not
417
# chopping paths so much. -- mbp 2010-02-11
418
child_base = os.path.basename(self.id2path(child))
419
if (child_base == elt):
420
# if we found an exact match, we can stop now; if
421
# we found an approximate match we need to keep
422
# searching because there might be an exact match
425
new_path = osutils.pathjoin(cur_path, child_base)
427
elif child_base.lower() == lelt:
429
new_path = osutils.pathjoin(cur_path, child_base)
431
# before a change is committed we can see this error...
436
# got to the end of this directory and no entries matched.
437
# Return what matched so far, plus the rest as specified.
438
cur_path = osutils.pathjoin(cur_path, elt, *list(bit_iter))
443
455
def get_root_id(self):
444
456
"""Return the file_id for the root of this tree."""
445
457
raise NotImplementedError(self.get_root_id)
447
def annotate_iter(self, file_id,
459
def annotate_iter(self, path,
448
460
default_revision=_mod_revision.CURRENT_REVISION):
449
461
"""Return an iterator of revision_id, line tuples.
451
463
For working trees (and mutable trees in general), the special
452
464
revision_id 'current:' will be used for lines that are new in this
453
465
tree, e.g. uncommitted changes.
454
:param file_id: The file to produce an annotated version from
466
:param path: The file to produce an annotated version from
455
467
:param default_revision: For lines that don't match a basis, mark them
456
468
with this revision id. Not all implementations will make use of
459
471
raise NotImplementedError(self.annotate_iter)
461
def _get_plan_merge_data(self, file_id, other, base):
462
from bzrlib import versionedfile
463
vf = versionedfile._PlanMergeVersionedFile(file_id)
464
last_revision_a = self._get_file_revision(file_id, vf, 'this:')
465
last_revision_b = other._get_file_revision(file_id, vf, 'other:')
467
last_revision_base = None
469
last_revision_base = base._get_file_revision(file_id, vf, 'base:')
470
return vf, last_revision_a, last_revision_b, last_revision_base
472
def plan_file_merge(self, file_id, other, base=None):
473
"""Generate a merge plan based on annotations.
475
If the file contains uncommitted changes in this tree, they will be
476
attributed to the 'current:' pseudo-revision. If the file contains
477
uncommitted changes in the other tree, they will be assigned to the
478
'other:' pseudo-revision.
480
data = self._get_plan_merge_data(file_id, other, base)
481
vf, last_revision_a, last_revision_b, last_revision_base = data
482
return vf.plan_merge(last_revision_a, last_revision_b,
485
def plan_file_lca_merge(self, file_id, other, base=None):
486
"""Generate a merge plan based lca-newness.
488
If the file contains uncommitted changes in this tree, they will be
489
attributed to the 'current:' pseudo-revision. If the file contains
490
uncommitted changes in the other tree, they will be assigned to the
491
'other:' pseudo-revision.
493
data = self._get_plan_merge_data(file_id, other, base)
494
vf, last_revision_a, last_revision_b, last_revision_base = data
495
return vf.plan_lca_merge(last_revision_a, last_revision_b,
498
473
def _iter_parent_trees(self):
499
474
"""Iterate through parent trees, defaulting to Tree.revision_tree."""
500
475
for revision_id in self.get_parent_ids():
503
478
except errors.NoSuchRevisionInTree:
504
479
yield self.repository.revision_tree(revision_id)
507
def _file_revision(revision_tree, file_id):
508
"""Determine the revision associated with a file in a given tree."""
509
revision_tree.lock_read()
511
return revision_tree.inventory[file_id].revision
513
revision_tree.unlock()
515
def _get_file_revision(self, file_id, vf, tree_revision):
516
"""Ensure that file_id, tree_revision is in vf to plan the merge."""
518
if getattr(self, '_repository', None) is None:
519
last_revision = tree_revision
520
parent_keys = [(file_id, self._file_revision(t, file_id)) for t in
521
self._iter_parent_trees()]
522
vf.add_lines((file_id, last_revision), parent_keys,
523
self.get_file(file_id).readlines())
524
repo = self.branch.repository
527
last_revision = self._file_revision(self, file_id)
528
base_vf = self._repository.texts
529
if base_vf not in vf.fallback_versionedfiles:
530
vf.fallback_versionedfiles.append(base_vf)
533
inventory = property(_get_inventory,
534
doc="Inventory of this Tree")
536
def _check_retrieved(self, ie, f):
539
fp = fingerprint_file(f)
542
if ie.text_size is not None:
543
if ie.text_size != fp['size']:
544
raise BzrError("mismatched size for file %r in %r" % (ie.file_id, self._store),
545
["inventory expects %d bytes" % ie.text_size,
546
"file is actually %d bytes" % fp['size'],
547
"store is probably damaged/corrupt"])
549
if ie.text_sha1 != fp['sha1']:
550
raise BzrError("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
551
["inventory expects %s" % ie.text_sha1,
552
"file is actually %s" % fp['sha1'],
553
"store is probably damaged/corrupt"])
556
481
def path2id(self, path):
557
482
"""Return the id for path in this tree."""
558
return self._inventory.path2id(path)
560
def paths2ids(self, paths, trees=[], require_versioned=True):
561
"""Return all the ids that can be reached by walking from paths.
563
Each path is looked up in this tree and any extras provided in
564
trees, and this is repeated recursively: the children in an extra tree
565
of a directory that has been renamed under a provided path in this tree
566
are all returned, even if none exist under a provided path in this
567
tree, and vice versa.
569
:param paths: An iterable of paths to start converting to ids from.
570
Alternatively, if paths is None, no ids should be calculated and None
571
will be returned. This is offered to make calling the api unconditional
572
for code that *might* take a list of files.
573
:param trees: Additional trees to consider.
574
:param require_versioned: If False, do not raise NotVersionedError if
575
an element of paths is not versioned in this tree and all of trees.
577
return find_ids_across_trees(paths, [self] + list(trees), require_versioned)
579
def iter_children(self, file_id):
580
entry = self.iter_entries_by_dir([file_id]).next()[1]
581
for child in getattr(entry, 'children', {}).itervalues():
483
raise NotImplementedError(self.path2id)
485
def is_versioned(self, path):
486
"""Check whether path is versioned.
488
:param path: Path to check
491
return self.path2id(path) is not None
493
def find_related_paths_across_trees(self, paths, trees=[],
494
require_versioned=True):
495
"""Find related paths in tree corresponding to specified filenames in any
498
All matches in all trees will be used, and all children of matched
499
directories will be used.
501
:param paths: The filenames to find related paths for (if None, returns
503
:param trees: The trees to find file_ids within
504
:param require_versioned: if true, all specified filenames must occur in
506
:return: a set of paths for the specified filenames and their children
509
raise NotImplementedError(self.find_related_paths_across_trees)
584
511
def lock_read(self):
512
"""Lock this tree for multiple read only operations.
514
:return: A breezy.lock.LogicalLockResult.
516
return lock.LogicalLockResult(self.unlock)
587
518
def revision_tree(self, revision_id):
588
519
"""Obtain a revision tree for the revision revision_id.
720
648
searcher = default_searcher
724
######################################################################
727
# TODO: Merge these two functions into a single one that can operate
728
# on either a whole tree or a set of files.
730
# TODO: Return the diff in order by filename, not by category or in
731
# random order. Can probably be done by lock-stepping through the
732
# filenames from both trees.
735
def file_status(filename, old_tree, new_tree):
736
"""Return single-letter status, old and new names for a file.
738
The complexity here is in deciding how to represent renames;
739
many complex cases are possible.
741
old_inv = old_tree.inventory
742
new_inv = new_tree.inventory
743
new_id = new_inv.path2id(filename)
744
old_id = old_inv.path2id(filename)
746
if not new_id and not old_id:
747
# easy: doesn't exist in either; not versioned at all
748
if new_tree.is_ignored(filename):
749
return 'I', None, None
751
return '?', None, None
753
# There is now a file of this name, great.
756
# There is no longer a file of this name, but we can describe
757
# what happened to the file that used to have
758
# this name. There are two possibilities: either it was
759
# deleted entirely, or renamed.
760
if new_inv.has_id(old_id):
761
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
763
return 'D', old_inv.id2path(old_id), None
765
# if the file_id is new in this revision, it is added
766
if new_id and not old_inv.has_id(new_id):
769
# if there used to be a file of this name, but that ID has now
770
# disappeared, it is deleted
771
if old_id and not new_inv.has_id(old_id):
777
@deprecated_function(deprecated_in((1, 9, 0)))
778
def find_renames(old_inv, new_inv):
779
for file_id in old_inv:
780
if file_id not in new_inv:
782
old_name = old_inv.id2path(file_id)
783
new_name = new_inv.id2path(file_id)
784
if old_name != new_name:
785
yield (old_name, new_name)
788
def find_ids_across_trees(filenames, trees, require_versioned=True):
789
"""Find the ids corresponding to specified filenames.
791
All matches in all trees will be used, and all children of matched
792
directories will be used.
794
:param filenames: The filenames to find file_ids for (if None, returns
796
:param trees: The trees to find file_ids within
797
:param require_versioned: if true, all specified filenames must occur in
799
:return: a set of file ids for the specified filenames and their children.
803
specified_path_ids = _find_ids_across_trees(filenames, trees,
805
return _find_children_across_trees(specified_path_ids, trees)
808
def _find_ids_across_trees(filenames, trees, require_versioned):
809
"""Find the ids corresponding to specified filenames.
811
All matches in all trees will be used, but subdirectories are not scanned.
813
:param filenames: The filenames to find file_ids for
814
:param trees: The trees to find file_ids within
815
:param require_versioned: if true, all specified filenames must occur in
817
:return: a set of file ids for the specified filenames
820
interesting_ids = set()
821
for tree_path in filenames:
824
file_id = tree.path2id(tree_path)
825
if file_id is not None:
826
interesting_ids.add(file_id)
829
not_versioned.append(tree_path)
830
if len(not_versioned) > 0 and require_versioned:
831
raise errors.PathsNotVersionedError(not_versioned)
832
return interesting_ids
835
def _find_children_across_trees(specified_ids, trees):
836
"""Return a set including specified ids and their children.
838
All matches in all trees will be used.
840
:param trees: The trees to find file_ids within
841
:return: a set containing all specified ids and their children
843
interesting_ids = set(specified_ids)
844
pending = interesting_ids
845
# now handle children of interesting ids
846
# we loop so that we handle all children of each id in both trees
847
while len(pending) > 0:
849
for file_id in pending:
851
if not tree.has_or_had_id(file_id):
853
for child_id in tree.iter_children(file_id):
854
if child_id not in interesting_ids:
855
new_pending.add(child_id)
856
interesting_ids.update(new_pending)
857
pending = new_pending
858
return interesting_ids
651
def archive(self, format, name, root='', subdir=None,
653
"""Create an archive of this tree.
655
:param format: Format name (e.g. 'tar')
656
:param name: target file name
657
:param root: Root directory name (or None)
658
:param subdir: Subdirectory to export (or None)
659
:return: Iterator over archive chunks
661
from .archive import create_archive
662
with self.lock_read():
663
return create_archive(format, self, name, root,
664
subdir, force_mtime=force_mtime)
667
def versionable_kind(cls, kind):
668
"""Check if this tree support versioning a specific file kind."""
669
return (kind in ('file', 'directory', 'symlink', 'tree-reference'))
861
672
class InterTree(InterObject):
930
741
if source_kind != target_kind:
931
742
changed_content = True
932
743
elif source_kind == 'file':
933
if (self.source.get_file_sha1(file_id, source_path, source_stat) !=
934
self.target.get_file_sha1(file_id, target_path, target_stat)):
744
if not self.file_content_matches(
745
source_path, target_path,
746
source_stat, target_stat):
935
747
changed_content = True
936
748
elif source_kind == 'symlink':
937
if (self.source.get_symlink_target(file_id) !=
938
self.target.get_symlink_target(file_id)):
939
changed_content = True
940
# XXX: Yes, the indentation below is wrong. But fixing it broke
941
# test_merge.TestMergerEntriesLCAOnDisk.
942
# test_nested_tree_subtree_renamed_and_modified. We'll wait for
943
# the fix from bzr.dev -- vila 2009026
944
elif source_kind == 'tree-reference':
945
if (self.source.get_reference_revision(file_id, source_path)
946
!= self.target.get_reference_revision(file_id, target_path)):
947
changed_content = True
749
if (self.source.get_symlink_target(source_path) !=
750
self.target.get_symlink_target(target_path)):
751
changed_content = True
752
elif source_kind == 'tree-reference':
753
if (self.source.get_reference_revision(source_path)
754
!= self.target.get_reference_revision(target_path)):
755
changed_content = True
948
756
parent = (source_parent, target_parent)
949
757
name = (source_name, target_name)
950
758
executable = (source_executable, target_executable)
951
if (changed_content is not False or versioned[0] != versioned[1]
952
or parent[0] != parent[1] or name[0] != name[1] or
953
executable[0] != executable[1]):
759
if (changed_content is not False or versioned[0] != versioned[1] or
760
parent[0] != parent[1] or name[0] != name[1] or
761
executable[0] != executable[1]):
957
765
return (file_id, (source_path, target_path), changed_content,
958
766
versioned, parent, name, kind, executable), changes
961
768
def compare(self, want_unchanged=False, specific_files=None,
962
extra_trees=None, require_versioned=False, include_root=False,
963
want_unversioned=False):
769
extra_trees=None, require_versioned=False, include_root=False,
770
want_unversioned=False):
964
771
"""Return the changes from source to target.
966
773
:return: A TreeDelta.
980
787
trees = (self.source,)
981
788
if extra_trees is not None:
982
789
trees = trees + tuple(extra_trees)
983
# target is usually the newer tree:
984
specific_file_ids = self.target.paths2ids(specific_files, trees,
985
require_versioned=require_versioned)
986
if specific_files and not specific_file_ids:
987
# All files are unversioned, so just return an empty delta
988
# _compare_trees would think we want a complete delta
989
result = delta.TreeDelta()
990
fake_entry = InventoryFile('unused', 'unused', 'unused')
991
result.unversioned = [(path, None,
992
self.target._comparison_data(fake_entry, path)[0]) for path in
995
return delta._compare_trees(self.source, self.target, want_unchanged,
996
specific_files, include_root, extra_trees=extra_trees,
997
require_versioned=require_versioned,
998
want_unversioned=want_unversioned)
790
with self.lock_read():
791
return delta._compare_trees(self.source, self.target, want_unchanged,
792
specific_files, include_root, extra_trees=extra_trees,
793
require_versioned=require_versioned,
794
want_unversioned=want_unversioned)
1000
796
def iter_changes(self, include_unchanged=False,
1001
specific_files=None, pb=None, extra_trees=[],
1002
require_versioned=True, want_unversioned=False):
797
specific_files=None, pb=None, extra_trees=[],
798
require_versioned=True, want_unversioned=False):
1003
799
"""Generate an iterator of changes between trees.
1005
801
A tuple is returned:
1055
857
seen_dirs = set()
1056
858
if want_unversioned:
1057
859
all_unversioned = sorted([(p.split('/'), p) for p in
1058
self.target.extras()
1059
if specific_files is None or
1060
osutils.is_inside_any(specific_files, p)])
861
if specific_files is None or
862
osutils.is_inside_any(specific_files, p)])
1061
863
all_unversioned = deque(all_unversioned)
1063
865
all_unversioned = deque()
1065
867
from_entries_by_dir = list(self.source.iter_entries_by_dir(
1066
specific_file_ids=specific_file_ids))
868
specific_files=source_specific_files))
1067
869
from_data = dict((e.file_id, (p, e)) for p, e in from_entries_by_dir)
1068
870
to_entries_by_dir = list(self.target.iter_entries_by_dir(
1069
specific_file_ids=specific_file_ids))
871
specific_files=target_specific_files))
1070
872
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
1072
874
# the unversioned path lookup only occurs on real trees - where there
1073
875
# can be extras. So the fake_entry is solely used to look up
1074
876
# executable it values when execute is not supported.
1075
fake_entry = InventoryFile('unused', 'unused', 'unused')
877
fake_entry = TreeFile()
1076
878
for target_path, target_entry in to_entries_by_dir:
1077
879
while (all_unversioned and
1078
all_unversioned[0][0] < target_path.split('/')):
880
all_unversioned[0][0] < target_path.split('/')):
1079
881
unversioned_path = all_unversioned.popleft()
1080
882
target_kind, target_executable, target_stat = \
1081
self.target._comparison_data(fake_entry, unversioned_path[1])
883
self.target._comparison_data(
884
fake_entry, unversioned_path[1])
1082
885
yield (None, (None, unversioned_path[1]), True, (False, False),
1084
(None, unversioned_path[0][-1]),
1085
(None, target_kind),
1086
(None, target_executable))
887
(None, unversioned_path[0][-1]),
889
(None, target_executable))
1087
890
source_path, source_entry = from_data.get(target_entry.file_id,
1089
892
result, changes = self._changes_from_entries(source_entry,
1090
target_entry, source_path=source_path, target_path=target_path)
893
target_entry, source_path=source_path, target_path=target_path)
1091
894
to_paths[result[0]] = result[1][1]
1092
895
entry_count += 1
1093
896
if result[3][0]:
1239
1043
precise_file_ids.add(new_parent_id)
1241
1045
if (result[6][0] == 'directory' and
1242
result[6][1] != 'directory'):
1046
result[6][1] != 'directory'):
1243
1047
# This stopped being a directory, the old children have
1244
1048
# to be included.
1245
if old_entry is None:
1049
if source_entry is None:
1246
1050
# Reusing a discarded change.
1247
old_entry = self._get_entry(self.source, file_id)
1248
for child in old_entry.children.values():
1249
precise_file_ids.add(child.file_id)
1051
source_entry = self._get_entry(
1052
self.source, result[1][0])
1053
precise_file_ids.update(
1055
for child in self.source.iter_child_entries(result[1][0]))
1250
1056
changed_file_ids.add(result[0])
1059
def file_content_matches(
1060
self, source_path, target_path,
1061
source_stat=None, target_stat=None):
1062
"""Check if two files are the same in the source and target trees.
1064
This only checks that the contents of the files are the same,
1065
it does not touch anything else.
1067
:param source_path: Path of the file in the source tree
1068
:param target_path: Path of the file in the target tree
1069
:param source_file_id: Optional file id of the file in the source tree
1070
:param target_file_id: Optional file id of the file in the target tree
1071
:param source_stat: Optional stat value of the file in the source tree
1072
:param target_stat: Optional stat value of the file in the target tree
1073
:return: Boolean indicating whether the files have the same contents
1075
with self.lock_read():
1076
source_verifier_kind, source_verifier_data = (
1077
self.source.get_file_verifier(source_path, source_stat))
1078
target_verifier_kind, target_verifier_data = (
1079
self.target.get_file_verifier(
1080
target_path, target_stat))
1081
if source_verifier_kind == target_verifier_kind:
1082
return (source_verifier_data == target_verifier_data)
1083
# Fall back to SHA1 for now
1084
if source_verifier_kind != "SHA1":
1085
source_sha1 = self.source.get_file_sha1(
1086
source_path, source_file_id, source_stat)
1088
source_sha1 = source_verifier_data
1089
if target_verifier_kind != "SHA1":
1090
target_sha1 = self.target.get_file_sha1(
1091
target_path, target_file_id, target_stat)
1093
target_sha1 = target_verifier_data
1094
return (source_sha1 == target_sha1)
1097
InterTree.register_optimiser(InterTree)
1254
1100
class MultiWalker(object):
1255
1101
"""Walk multiple trees simultaneously, getting combined results."""
1473
1319
# the lookup_by_file_id will be removing anything processed
1474
1320
# from the extras cache
1475
1321
other_extra.pop(file_id)
1476
other_values = [(None, None) for i in xrange(idx)]
1322
other_values = [(None, None)] * idx
1477
1323
other_values.append((other_path, other_ie))
1478
for alt_idx, alt_extra in enumerate(self._others_extra[idx+1:]):
1324
for alt_idx, alt_extra in enumerate(self._others_extra[idx + 1:]):
1479
1325
alt_idx = alt_idx + idx + 1
1480
1326
alt_extra = self._others_extra[alt_idx]
1481
1327
alt_tree = self._other_trees[alt_idx]
1482
1328
other_values.append(self._lookup_by_file_id(
1483
alt_extra, alt_tree, file_id))
1329
alt_extra, alt_tree, file_id))
1484
1330
yield other_path, file_id, None, other_values
1333
def find_previous_paths(from_tree, to_tree, paths):
1334
"""Find previous tree paths.
1336
:param from_tree: From tree
1337
:param to_tree: To tree
1338
:param paths: Iterable over paths to search for
1339
:return: Dictionary mapping from from_tree paths to paths in to_tree, or
1340
None if there is no equivalent path.
1344
ret[path] = find_previous_path(from_tree, to_tree, path)
1348
def find_previous_path(from_tree, to_tree, path, file_id=None):
1349
"""Find previous tree path.
1351
:param from_tree: From tree
1352
:param to_tree: To tree
1353
:param path: Path to search for
1354
:return: path in to_tree, or None if there is no equivalent path.
1357
file_id = from_tree.path2id(path)
1359
raise errors.NoSuchFile(path)
1361
return to_tree.id2path(file_id)
1362
except errors.NoSuchId:
1366
def get_canonical_path(tree, path, normalize):
1367
"""Find the canonical path of an item, ignoring case.
1369
:param tree: Tree to traverse
1370
:param path: Case-insensitive path to look up
1371
:param normalize: Function to normalize a filename for comparison
1372
:return: The canonical path
1376
bit_iter = iter(path.split("/"))
1377
for elt in bit_iter:
1378
lelt = normalize(elt)
1381
for child in tree.iter_child_entries(cur_path):
1383
if child.name == elt:
1384
# if we found an exact match, we can stop now; if
1385
# we found an approximate match we need to keep
1386
# searching because there might be an exact match
1388
new_path = osutils.pathjoin(cur_path, child.name)
1390
elif normalize(child.name) == lelt:
1391
new_path = osutils.pathjoin(cur_path, child.name)
1392
except errors.NoSuchId:
1393
# before a change is committed we can see this error...
1395
except errors.NotADirectory:
1400
# got to the end of this directory and no entries matched.
1401
# Return what matched so far, plus the rest as specified.
1402
cur_path = osutils.pathjoin(cur_path, elt, *list(bit_iter))
added
removed
breezy/tree.py
