17
17
"""Tree classes, representing directory at point in time.
20
from __future__ import absolute_import
22
from .lazy_import import lazy_import
23
lazy_import(globals(), """
21
from collections import deque
27
25
conflicts as _mod_conflicts,
31
30
revision as _mod_revision,
35
from breezy.i18n import gettext
43
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 TreeEntry(object):
61
"""An entry that implements the minimum interface used by commands.
64
def __eq__(self, other):
65
# yes, this is ugly, TODO: best practice __eq__ style.
66
return (isinstance(other, TreeEntry)
67
and other.__class__ == self.__class__)
69
def kind_character(self):
73
class TreeDirectory(TreeEntry):
74
"""See TreeEntry. This is a directory in a working tree."""
76
def kind_character(self):
80
class TreeFile(TreeEntry):
81
"""See TreeEntry. This is a regular file in a working tree."""
83
def kind_character(self):
87
class TreeLink(TreeEntry):
88
"""See TreeEntry. This is a symlink in a working tree."""
90
def kind_character(self):
94
class TreeReference(TreeEntry):
95
"""See TreeEntry. This is a reference to a nested tree in a working tree."""
97
def kind_character(self):
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
101
43
class Tree(object):
108
50
* `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()`.
110
59
Trees can be compared, etc, regardless of whether they are working
111
60
trees or versioned trees.
114
def supports_rename_tracking(self):
115
"""Whether this tree supports rename tracking.
117
This defaults to True, but some implementations may want to override
122
def has_versioned_directories(self):
123
"""Whether this tree can contain explicitly versioned directories.
125
This defaults to True, but some implementations may want to override
130
63
def changes_from(self, other, want_unchanged=False, specific_files=None,
131
extra_trees=None, require_versioned=False, include_root=False,
132
want_unversioned=False):
64
extra_trees=None, require_versioned=False, include_root=False,
65
want_unversioned=False):
133
66
"""Return a TreeDelta of the changes from other to this tree.
135
68
:param other: A tree to compare with.
147
return iter(self.inventory)
204
149
def all_file_ids(self):
205
150
"""Iterate through all file ids, including ids for missing files."""
206
raise NotImplementedError(self.all_file_ids)
208
def all_versioned_paths(self):
209
"""Iterate through all paths, including paths for missing files."""
210
raise NotImplementedError(self.all_versioned_paths)
151
return set(self.inventory)
212
153
def id2path(self, file_id):
213
154
"""Return the path for a file id.
215
156
:raises NoSuchId:
217
raise NotImplementedError(self.id2path)
219
def iter_entries_by_dir(self, specific_files=None):
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):
220
174
"""Walk the tree in 'by_dir' order.
222
176
This will yield each entry in the tree as a (path, entry) tuple.
240
194
The yield order (ignoring root) would be::
242
195
a, f, a/b, a/d, a/b/c, a/d/e, f/g
244
raise NotImplementedError(self.iter_entries_by_dir)
246
def iter_child_entries(self, path):
247
"""Iterate over the children of a directory or tree reference.
249
:param path: Path of the directory
250
:raise NoSuchFile: When the path does not exist
251
:return: Iterator over entries in the directory
253
raise NotImplementedError(self.iter_child_entries)
255
def list_files(self, include_root=False, from_dir=None, recursive=True):
256
"""List all files in this tree.
258
:param include_root: Whether to include the entry for the tree root
259
:param from_dir: Directory under which to list files
260
:param recursive: Whether to list files recursively
261
:return: iterator over tuples of (path, versioned, kind, file_id,
264
raise NotImplementedError(self.list_files)
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)
266
204
def iter_references(self):
267
205
if self.supports_tree_reference():
316
254
raise NotImplementedError(self._comparison_data)
318
def get_file(self, path):
319
"""Return a file object for the file path in the tree.
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.
321
268
raise NotImplementedError(self.get_file)
323
def get_file_with_stat(self, path):
324
"""Get a file handle and stat object for path.
270
def get_file_with_stat(self, file_id, path=None):
271
"""Get a file handle and stat object for file_id.
326
273
The default implementation returns (self.get_file, None) for backwards
329
:param path: The path of the file.
276
:param file_id: The file id to read.
277
:param path: The path of the file, if it is known.
330
278
:return: A tuple (file_handle, stat_value_or_None). If the tree has
331
279
no stat facility, or need for a stat cache feedback during commit,
332
280
it may return None for the second element of the tuple.
334
return (self.get_file(path), None)
282
return (self.get_file(file_id, path), None)
336
def get_file_text(self, path):
284
def get_file_text(self, file_id, path=None):
337
285
"""Return the byte content of a file.
287
:param file_id: The file_id of the file.
339
288
:param path: The path of the file.
341
:returns: A single byte string for the whole file.
289
If both file_id and path are supplied, an implementation may use
343
with self.get_file(path) as my_file:
292
my_file = self.get_file(file_id, path)
344
294
return my_file.read()
346
def get_file_lines(self, path):
298
def get_file_lines(self, file_id, path=None):
347
299
"""Return the content of a file, as lines.
301
:param file_id: The file_id of the file.
349
302
:param path: The path of the file.
303
If both file_id and path are supplied, an implementation may use
351
return osutils.split_lines(self.get_file_text(path))
353
def get_file_verifier(self, path, stat_value=None):
354
"""Return a verifier for a file.
356
The default implementation returns a sha1.
306
return osutils.split_lines(self.get_file_text(file_id, path))
308
def get_file_mtime(self, file_id, path=None):
309
"""Return the modification time for a file.
311
:param file_id: The handle for this file.
358
312
:param path: The path that this file can be found at.
359
313
These must point to the same object.
360
:param stat_value: Optional stat value for the object
361
:return: Tuple with verifier name and verifier data
363
return ("SHA1", self.get_file_sha1(path, stat_value=stat_value))
365
def get_file_sha1(self, path, stat_value=None):
366
"""Return the SHA1 file for a file.
368
:note: callers should use get_file_verifier instead
369
where possible, as the underlying repository implementation may
370
have quicker access to a non-sha1 verifier.
372
:param path: The path that this file can be found at.
373
:param stat_value: Optional stat value for the object
375
raise NotImplementedError(self.get_file_sha1)
377
def get_file_mtime(self, path):
378
"""Return the modification time for a file.
380
:param path: The path that this file can be found at.
382
315
raise NotImplementedError(self.get_file_mtime)
384
def get_file_size(self, path):
317
def get_file_size(self, file_id):
385
318
"""Return the size of a file in bytes.
387
320
This applies only to regular files. If invoked on directories or
388
321
symlinks, it will return None.
322
:param file_id: The file-id of the file
390
324
raise NotImplementedError(self.get_file_size)
392
def is_executable(self, path):
393
"""Check if a file is executable.
395
:param path: The path that this file can be found at.
397
raise NotImplementedError(self.is_executable)
326
def get_file_by_path(self, path):
327
return self.get_file(self._inventory.path2id(path), path)
399
329
def iter_files_bytes(self, desired_files):
400
330
"""Iterate through file contents.
412
342
this implementation, it is a tuple containing a single bytestring with
413
343
the complete text of the file.
415
:param desired_files: a list of (path, identifier) pairs
345
:param desired_files: a list of (file_id, identifier) pairs
417
for path, identifier in desired_files:
347
for file_id, identifier in desired_files:
418
348
# We wrap the string in a tuple so that we can return an iterable
419
349
# of bytestrings. (Technically, a bytestring is also an iterable
420
350
# of bytestrings, but iterating through each character is not
422
cur_file = (self.get_file_text(path),)
352
cur_file = (self.get_file_text(file_id),)
423
353
yield identifier, cur_file
425
def get_symlink_target(self, path):
426
"""Get the target for a given path.
355
def get_symlink_target(self, file_id):
356
"""Get the target for a given file_id.
428
It is assumed that the caller already knows that path is referencing
358
It is assumed that the caller already knows that file_id is referencing
430
:param path: The path of the file.
360
:param file_id: Handle for the symlink entry.
431
361
:return: The path the symlink points to.
433
363
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))
435
443
def get_root_id(self):
436
444
"""Return the file_id for the root of this tree."""
437
445
raise NotImplementedError(self.get_root_id)
439
def annotate_iter(self, path,
447
def annotate_iter(self, file_id,
440
448
default_revision=_mod_revision.CURRENT_REVISION):
441
449
"""Return an iterator of revision_id, line tuples.
443
451
For working trees (and mutable trees in general), the special
444
452
revision_id 'current:' will be used for lines that are new in this
445
453
tree, e.g. uncommitted changes.
446
:param path: The file to produce an annotated version from
454
:param file_id: The file to produce an annotated version from
447
455
:param default_revision: For lines that don't match a basis, mark them
448
456
with this revision id. Not all implementations will make use of
451
459
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,
453
498
def _iter_parent_trees(self):
454
499
"""Iterate through parent trees, defaulting to Tree.revision_tree."""
455
500
for revision_id in self.get_parent_ids():
458
503
except errors.NoSuchRevisionInTree:
459
504
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"])
461
556
def path2id(self, path):
462
557
"""Return the id for path in this tree."""
463
raise NotImplementedError(self.path2id)
465
def is_versioned(self, path):
466
"""Check whether path is versioned.
468
:param path: Path to check
471
return self.path2id(path) is not None
473
def find_related_paths_across_trees(self, paths, trees=[],
474
require_versioned=True):
475
"""Find related paths in tree corresponding to specified filenames in any
478
All matches in all trees will be used, and all children of matched
479
directories will be used.
481
:param paths: The filenames to find related paths for (if None, returns
483
:param trees: The trees to find file_ids within
484
:param require_versioned: if true, all specified filenames must occur in
486
:return: a set of paths for the specified filenames and their children
489
raise NotImplementedError(self.find_related_paths_across_trees)
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():
491
584
def lock_read(self):
492
"""Lock this tree for multiple read only operations.
494
:return: A breezy.lock.LogicalLockResult.
496
return lock.LogicalLockResult(self.unlock)
498
587
def revision_tree(self, revision_id):
499
588
"""Obtain a revision tree for the revision revision_id.
628
720
searcher = default_searcher
631
def archive(self, format, name, root='', subdir=None,
633
"""Create an archive of this tree.
635
:param format: Format name (e.g. 'tar')
636
:param name: target file name
637
:param root: Root directory name (or None)
638
:param subdir: Subdirectory to export (or None)
639
:return: Iterator over archive chunks
641
from .archive import create_archive
642
with self.lock_read():
643
return create_archive(format, self, name, root,
644
subdir, force_mtime=force_mtime)
647
def versionable_kind(cls, kind):
648
"""Check if this tree support versioning a specific file kind."""
649
return (kind in ('file', 'directory', 'symlink', 'tree-reference'))
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
652
861
class InterTree(InterObject):
721
930
if source_kind != target_kind:
722
931
changed_content = True
723
932
elif source_kind == 'file':
724
if not self.file_content_matches(
725
source_path, target_path,
726
source_stat, target_stat):
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)):
727
935
changed_content = True
728
936
elif source_kind == 'symlink':
729
if (self.source.get_symlink_target(source_path) !=
730
self.target.get_symlink_target(target_path)):
937
if (self.source.get_symlink_target(file_id) !=
938
self.target.get_symlink_target(file_id)):
731
939
changed_content = True
732
elif source_kind == 'tree-reference':
733
if (self.source.get_reference_revision(source_path)
734
!= self.target.get_reference_revision(target_path)):
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)):
735
947
changed_content = True
736
948
parent = (source_parent, target_parent)
737
949
name = (source_name, target_name)
738
950
executable = (source_executable, target_executable)
739
if (changed_content is not False or versioned[0] != versioned[1] or
740
parent[0] != parent[1] or name[0] != name[1] or
741
executable[0] != executable[1]):
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]):
745
957
return (file_id, (source_path, target_path), changed_content,
746
958
versioned, parent, name, kind, executable), changes
748
961
def compare(self, want_unchanged=False, specific_files=None,
749
extra_trees=None, require_versioned=False, include_root=False,
750
want_unversioned=False):
962
extra_trees=None, require_versioned=False, include_root=False,
963
want_unversioned=False):
751
964
"""Return the changes from source to target.
753
966
:return: A TreeDelta.
767
980
trees = (self.source,)
768
981
if extra_trees is not None:
769
982
trees = trees + tuple(extra_trees)
770
with self.lock_read():
771
return delta._compare_trees(self.source, self.target, want_unchanged,
772
specific_files, include_root, extra_trees=extra_trees,
773
require_versioned=require_versioned,
774
want_unversioned=want_unversioned)
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)
776
1000
def iter_changes(self, include_unchanged=False,
777
specific_files=None, pb=None, extra_trees=[],
778
require_versioned=True, want_unversioned=False):
1001
specific_files=None, pb=None, extra_trees=[],
1002
require_versioned=True, want_unversioned=False):
779
1003
"""Generate an iterator of changes between trees.
781
1005
A tuple is returned:
837
1055
seen_dirs = set()
838
1056
if want_unversioned:
839
1057
all_unversioned = sorted([(p.split('/'), p) for p in
841
if specific_files is None or
842
osutils.is_inside_any(specific_files, p)])
843
all_unversioned = collections.deque(all_unversioned)
1058
self.target.extras()
1059
if specific_files is None or
1060
osutils.is_inside_any(specific_files, p)])
1061
all_unversioned = deque(all_unversioned)
845
all_unversioned = collections.deque()
1063
all_unversioned = deque()
847
1065
from_entries_by_dir = list(self.source.iter_entries_by_dir(
848
specific_files=source_specific_files))
1066
specific_file_ids=specific_file_ids))
849
1067
from_data = dict((e.file_id, (p, e)) for p, e in from_entries_by_dir)
850
1068
to_entries_by_dir = list(self.target.iter_entries_by_dir(
851
specific_files=target_specific_files))
1069
specific_file_ids=specific_file_ids))
852
1070
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
854
1072
# the unversioned path lookup only occurs on real trees - where there
855
1073
# can be extras. So the fake_entry is solely used to look up
856
1074
# executable it values when execute is not supported.
857
fake_entry = TreeFile()
1075
fake_entry = InventoryFile('unused', 'unused', 'unused')
858
1076
for target_path, target_entry in to_entries_by_dir:
859
1077
while (all_unversioned and
860
all_unversioned[0][0] < target_path.split('/')):
1078
all_unversioned[0][0] < target_path.split('/')):
861
1079
unversioned_path = all_unversioned.popleft()
862
1080
target_kind, target_executable, target_stat = \
863
self.target._comparison_data(
864
fake_entry, unversioned_path[1])
1081
self.target._comparison_data(fake_entry, unversioned_path[1])
865
1082
yield (None, (None, unversioned_path[1]), True, (False, False),
867
(None, unversioned_path[0][-1]),
869
(None, target_executable))
1084
(None, unversioned_path[0][-1]),
1085
(None, target_kind),
1086
(None, target_executable))
870
1087
source_path, source_entry = from_data.get(target_entry.file_id,
872
1089
result, changes = self._changes_from_entries(source_entry,
873
target_entry, source_path=source_path, target_path=target_path)
1090
target_entry, source_path=source_path, target_path=target_path)
874
1091
to_paths[result[0]] = result[1][1]
875
1092
entry_count += 1
876
1093
if result[3][0]:
1023
1239
precise_file_ids.add(new_parent_id)
1025
1241
if (result[6][0] == 'directory' and
1026
result[6][1] != 'directory'):
1242
result[6][1] != 'directory'):
1027
1243
# This stopped being a directory, the old children have
1028
1244
# to be included.
1029
if source_entry is None:
1245
if old_entry is None:
1030
1246
# Reusing a discarded change.
1031
source_entry = self._get_entry(
1032
self.source, result[1][0])
1033
precise_file_ids.update(
1035
for child in self.source.iter_child_entries(result[1][0]))
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)
1036
1250
changed_file_ids.add(result[0])
1039
def file_content_matches(
1040
self, source_path, target_path,
1041
source_stat=None, target_stat=None):
1042
"""Check if two files are the same in the source and target trees.
1044
This only checks that the contents of the files are the same,
1045
it does not touch anything else.
1047
:param source_path: Path of the file in the source tree
1048
:param target_path: Path of the file in the target tree
1049
:param source_file_id: Optional file id of the file in the source tree
1050
:param target_file_id: Optional file id of the file in the target tree
1051
:param source_stat: Optional stat value of the file in the source tree
1052
:param target_stat: Optional stat value of the file in the target tree
1053
:return: Boolean indicating whether the files have the same contents
1055
with self.lock_read():
1056
source_verifier_kind, source_verifier_data = (
1057
self.source.get_file_verifier(source_path, source_stat))
1058
target_verifier_kind, target_verifier_data = (
1059
self.target.get_file_verifier(
1060
target_path, target_stat))
1061
if source_verifier_kind == target_verifier_kind:
1062
return (source_verifier_data == target_verifier_data)
1063
# Fall back to SHA1 for now
1064
if source_verifier_kind != "SHA1":
1065
source_sha1 = self.source.get_file_sha1(
1066
source_path, source_file_id, source_stat)
1068
source_sha1 = source_verifier_data
1069
if target_verifier_kind != "SHA1":
1070
target_sha1 = self.target.get_file_sha1(
1071
target_path, target_file_id, target_stat)
1073
target_sha1 = target_verifier_data
1074
return (source_sha1 == target_sha1)
1077
InterTree.register_optimiser(InterTree)
1080
1254
class MultiWalker(object):
1081
1255
"""Walk multiple trees simultaneously, getting combined results."""
1299
1473
# the lookup_by_file_id will be removing anything processed
1300
1474
# from the extras cache
1301
1475
other_extra.pop(file_id)
1302
other_values = [(None, None)] * idx
1476
other_values = [(None, None) for i in xrange(idx)]
1303
1477
other_values.append((other_path, other_ie))
1304
for alt_idx, alt_extra in enumerate(self._others_extra[idx + 1:]):
1478
for alt_idx, alt_extra in enumerate(self._others_extra[idx+1:]):
1305
1479
alt_idx = alt_idx + idx + 1
1306
1480
alt_extra = self._others_extra[alt_idx]
1307
1481
alt_tree = self._other_trees[alt_idx]
1308
1482
other_values.append(self._lookup_by_file_id(
1309
alt_extra, alt_tree, file_id))
1483
alt_extra, alt_tree, file_id))
1310
1484
yield other_path, file_id, None, other_values
1313
def find_previous_paths(from_tree, to_tree, paths):
1314
"""Find previous tree paths.
1316
:param from_tree: From tree
1317
:param to_tree: To tree
1318
:param paths: Iterable over paths to search for
1319
:return: Dictionary mapping from from_tree paths to paths in to_tree, or
1320
None if there is no equivalent path.
1324
ret[path] = find_previous_path(from_tree, to_tree, path)
1328
def find_previous_path(from_tree, to_tree, path, file_id=None):
1329
"""Find previous tree path.
1331
:param from_tree: From tree
1332
:param to_tree: To tree
1333
:param path: Path to search for
1334
:return: path in to_tree, or None if there is no equivalent path.
1337
file_id = from_tree.path2id(path)
1339
raise errors.NoSuchFile(path)
1341
return to_tree.id2path(file_id)
1342
except errors.NoSuchId:
1346
def get_canonical_path(tree, path, normalize):
1347
"""Find the canonical path of an item, ignoring case.
1349
:param tree: Tree to traverse
1350
:param path: Case-insensitive path to look up
1351
:param normalize: Function to normalize a filename for comparison
1352
:return: The canonical path
1355
cur_id = tree.get_root_id()
1357
bit_iter = iter(path.split("/"))
1358
for elt in bit_iter:
1359
lelt = normalize(elt)
1362
for child in tree.iter_child_entries(cur_path, cur_id):
1364
if child.name == elt:
1365
# if we found an exact match, we can stop now; if
1366
# we found an approximate match we need to keep
1367
# searching because there might be an exact match
1369
cur_id = child.file_id
1370
new_path = osutils.pathjoin(cur_path, child.name)
1372
elif normalize(child.name) == lelt:
1373
cur_id = child.file_id
1374
new_path = osutils.pathjoin(cur_path, child.name)
1375
except errors.NoSuchId:
1376
# before a change is committed we can see this error...
1378
except errors.NotADirectory:
1383
# got to the end of this directory and no entries matched.
1384
# Return what matched so far, plus the rest as specified.
1385
cur_path = osutils.pathjoin(cur_path, elt, *list(bit_iter))
added
removed
bzrlib/tree.py
