1
# Copyright (C) 2005-2011 Canonical Ltd
3
# This program is free software; you can redistribute it and/or modify
4
# it under the terms of the GNU General Public License as published by
5
# the Free Software Foundation; either version 2 of the License, or
6
# (at your option) any later version.
8
# This program is distributed in the hope that it will be useful,
9
# but WITHOUT ANY WARRANTY; without even the implied warranty of
10
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11
# GNU General Public License for more details.
13
# You should have received a copy of the GNU General Public License
14
# along with this program; if not, write to the Free Software
15
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
17
"""Tree classes, representing directory at point in time.
20
from __future__ import absolute_import
24
from .lazy_import import lazy_import
25
lazy_import(globals(), """
29
conflicts as _mod_conflicts,
36
revision as _mod_revision,
40
from breezy.i18n import gettext
43
from .decorators import needs_read_lock
44
from .inter import InterObject
51
"""Abstract file tree.
53
There are several subclasses:
55
* `WorkingTree` exists as files on disk editable by the user.
57
* `RevisionTree` is a tree as recorded at some point in the past.
59
Trees can be compared, etc, regardless of whether they are working
60
trees or versioned trees.
63
def has_versioned_directories(self):
64
"""Whether this tree can contain explicitly versioned directories.
66
This defaults to True, but some implementations may want to override
71
def changes_from(self, other, want_unchanged=False, specific_files=None,
72
extra_trees=None, require_versioned=False, include_root=False,
73
want_unversioned=False):
74
"""Return a TreeDelta of the changes from other to this tree.
76
:param other: A tree to compare with.
77
:param specific_files: An optional list of file paths to restrict the
78
comparison to. When mapping filenames to ids, all matches in all
79
trees (including optional extra_trees) are used, and all children of
80
matched directories are included.
81
:param want_unchanged: An optional boolean requesting the inclusion of
82
unchanged entries in the result.
83
:param extra_trees: An optional list of additional trees to use when
84
mapping the contents of specific_files (paths) to file_ids.
85
:param require_versioned: An optional boolean (defaults to False). When
86
supplied and True all the 'specific_files' must be versioned, or
87
a PathsNotVersionedError will be thrown.
88
:param want_unversioned: Scan for unversioned paths.
90
The comparison will be performed by an InterTree object looked up on
93
# Martin observes that Tree.changes_from returns a TreeDelta and this
94
# may confuse people, because the class name of the returned object is
95
# a synonym of the object referenced in the method name.
96
return InterTree.get(other, self).compare(
97
want_unchanged=want_unchanged,
98
specific_files=specific_files,
99
extra_trees=extra_trees,
100
require_versioned=require_versioned,
101
include_root=include_root,
102
want_unversioned=want_unversioned,
105
def iter_changes(self, from_tree, include_unchanged=False,
106
specific_files=None, pb=None, extra_trees=None,
107
require_versioned=True, want_unversioned=False):
108
"""See InterTree.iter_changes"""
109
intertree = InterTree.get(from_tree, self)
110
return intertree.iter_changes(include_unchanged, specific_files, pb,
111
extra_trees, require_versioned, want_unversioned=want_unversioned)
114
"""Get a list of the conflicts in the tree.
116
Each conflict is an instance of breezy.conflicts.Conflict.
118
return _mod_conflicts.ConflictList()
121
"""For trees that can have unversioned files, return all such paths."""
124
def get_parent_ids(self):
125
"""Get the parent ids for this tree.
127
:return: a list of parent ids. [] is returned to indicate
128
a tree with no parents.
129
:raises: BzrError if the parents are not known.
131
raise NotImplementedError(self.get_parent_ids)
133
def has_filename(self, filename):
134
"""True if the tree has given filename."""
135
raise NotImplementedError(self.has_filename)
137
def has_id(self, file_id):
138
raise NotImplementedError(self.has_id)
140
def has_or_had_id(self, file_id):
141
raise NotImplementedError(self.has_or_had_id)
143
def is_ignored(self, filename):
144
"""Check whether the filename is ignored by this tree.
146
:param filename: The relative filename within the tree.
147
:return: True if the filename is ignored.
151
def all_file_ids(self):
152
"""Iterate through all file ids, including ids for missing files."""
153
raise NotImplementedError(self.all_file_ids)
155
def id2path(self, file_id):
156
"""Return the path for a file id.
160
raise NotImplementedError(self.id2path)
162
def iter_entries_by_dir(self, specific_file_ids=None, yield_parents=False):
163
"""Walk the tree in 'by_dir' order.
165
This will yield each entry in the tree as a (path, entry) tuple.
166
The order that they are yielded is:
168
Directories are walked in a depth-first lexicographical order,
169
however, whenever a directory is reached, all of its direct child
170
nodes are yielded in lexicographical order before yielding the
173
For example, in the tree::
183
The yield order (ignoring root) would be::
185
a, f, a/b, a/d, a/b/c, a/d/e, f/g
187
:param yield_parents: If True, yield the parents from the root leading
188
down to specific_file_ids that have been requested. This has no
189
impact if specific_file_ids is None.
191
raise NotImplementedError(self.iter_entries_by_dir)
193
def iter_child_entries(self, file_id, path=None):
194
"""Iterate over the children of a directory or tree reference.
196
:param file_id: File id of the directory/tree-reference
197
:param path: Optional path of the directory
198
:raise NoSuchId: When the file_id does not exist
199
:return: Iterator over entries in the directory
201
raise NotImplementedError(self.iter_child_entries)
203
def list_files(self, include_root=False, from_dir=None, recursive=True):
204
"""List all files in this tree.
206
:param include_root: Whether to include the entry for the tree root
207
:param from_dir: Directory under which to list files
208
:param recursive: Whether to list files recursively
209
:return: iterator over tuples of (path, versioned, kind, file_id,
212
raise NotImplementedError(self.list_files)
214
def iter_references(self):
215
if self.supports_tree_reference():
216
for path, entry in self.iter_entries_by_dir():
217
if entry.kind == 'tree-reference':
218
yield path, entry.file_id
220
def kind(self, file_id):
221
raise NotImplementedError("Tree subclass %s must implement kind"
222
% self.__class__.__name__)
224
def stored_kind(self, file_id):
225
"""File kind stored for this file_id.
227
May not match kind on disk for working trees. Always available
228
for versioned files, even when the file itself is missing.
230
return self.kind(file_id)
232
def path_content_summary(self, path):
233
"""Get a summary of the information about path.
235
All the attributes returned are for the canonical form, not the
236
convenient form (if content filters are in use.)
238
:param path: A relative path within the tree.
239
:return: A tuple containing kind, size, exec, sha1-or-link.
240
Kind is always present (see tree.kind()).
241
size is present if kind is file and the size of the
242
canonical form can be cheaply determined, None otherwise.
243
exec is None unless kind is file and the platform supports the 'x'
245
sha1-or-link is the link target if kind is symlink, or the sha1 if
246
it can be obtained without reading the file.
248
raise NotImplementedError(self.path_content_summary)
250
def get_reference_revision(self, file_id, path=None):
251
raise NotImplementedError("Tree subclass %s must implement "
252
"get_reference_revision"
253
% self.__class__.__name__)
255
def _comparison_data(self, entry, path):
256
"""Return a tuple of kind, executable, stat_value for a file.
258
entry may be None if there is no inventory entry for the file, but
259
path must always be supplied.
261
kind is None if there is no file present (even if an inventory id is
262
present). executable is False for non-file entries.
264
raise NotImplementedError(self._comparison_data)
266
def _file_size(self, entry, stat_value):
267
raise NotImplementedError(self._file_size)
269
def get_file(self, file_id, path=None):
270
"""Return a file object for the file file_id in the tree.
272
If both file_id and path are defined, it is implementation defined as
273
to which one is used.
275
raise NotImplementedError(self.get_file)
277
def get_file_with_stat(self, file_id, path=None):
278
"""Get a file handle and stat object for file_id.
280
The default implementation returns (self.get_file, None) for backwards
283
:param file_id: The file id to read.
284
:param path: The path of the file, if it is known.
285
:return: A tuple (file_handle, stat_value_or_None). If the tree has
286
no stat facility, or need for a stat cache feedback during commit,
287
it may return None for the second element of the tuple.
289
return (self.get_file(file_id, path), None)
291
def get_file_text(self, file_id, path=None):
292
"""Return the byte content of a file.
294
:param file_id: The file_id of the file.
295
:param path: The path of the file.
297
If both file_id and path are supplied, an implementation may use
300
:returns: A single byte string for the whole file.
302
my_file = self.get_file(file_id, path)
304
return my_file.read()
308
def get_file_lines(self, file_id, path=None):
309
"""Return the content of a file, as lines.
311
:param file_id: The file_id of the file.
312
:param path: The path of the file.
314
If both file_id and path are supplied, an implementation may use
317
return osutils.split_lines(self.get_file_text(file_id, path))
319
def get_file_verifier(self, file_id, path=None, stat_value=None):
320
"""Return a verifier for a file.
322
The default implementation returns a sha1.
324
:param file_id: The handle for this file.
325
:param path: The path that this file can be found at.
326
These must point to the same object.
327
:param stat_value: Optional stat value for the object
328
:return: Tuple with verifier name and verifier data
330
return ("SHA1", self.get_file_sha1(file_id, path=path,
331
stat_value=stat_value))
333
def get_file_sha1(self, file_id, path=None, stat_value=None):
334
"""Return the SHA1 file for a file.
336
:note: callers should use get_file_verifier instead
337
where possible, as the underlying repository implementation may
338
have quicker access to a non-sha1 verifier.
340
:param file_id: The handle for this file.
341
:param path: The path that this file can be found at.
342
These must point to the same object.
343
:param stat_value: Optional stat value for the object
345
raise NotImplementedError(self.get_file_sha1)
347
def get_file_mtime(self, file_id, path=None):
348
"""Return the modification time for a file.
350
:param file_id: The handle for this file.
351
:param path: The path that this file can be found at.
352
These must point to the same object.
354
raise NotImplementedError(self.get_file_mtime)
356
def get_file_size(self, file_id):
357
"""Return the size of a file in bytes.
359
This applies only to regular files. If invoked on directories or
360
symlinks, it will return None.
361
:param file_id: The file-id of the file
363
raise NotImplementedError(self.get_file_size)
365
def is_executable(self, file_id, path=None):
366
"""Check if a file is executable.
368
:param file_id: The handle for this file.
369
:param path: The path that this file can be found at.
370
These must point to the same object.
372
raise NotImplementedError(self.is_executable)
374
def iter_files_bytes(self, desired_files):
375
"""Iterate through file contents.
377
Files will not necessarily be returned in the order they occur in
378
desired_files. No specific order is guaranteed.
380
Yields pairs of identifier, bytes_iterator. identifier is an opaque
381
value supplied by the caller as part of desired_files. It should
382
uniquely identify the file version in the caller's context. (Examples:
383
an index number or a TreeTransform trans_id.)
385
bytes_iterator is an iterable of bytestrings for the file. The
386
kind of iterable and length of the bytestrings are unspecified, but for
387
this implementation, it is a tuple containing a single bytestring with
388
the complete text of the file.
390
:param desired_files: a list of (file_id, identifier) pairs
392
for file_id, identifier in desired_files:
393
# We wrap the string in a tuple so that we can return an iterable
394
# of bytestrings. (Technically, a bytestring is also an iterable
395
# of bytestrings, but iterating through each character is not
397
cur_file = (self.get_file_text(file_id),)
398
yield identifier, cur_file
400
def get_symlink_target(self, file_id, path=None):
401
"""Get the target for a given file_id.
403
It is assumed that the caller already knows that file_id is referencing
405
:param file_id: Handle for the symlink entry.
406
:param path: The path of the file.
407
If both file_id and path are supplied, an implementation may use
409
:return: The path the symlink points to.
411
raise NotImplementedError(self.get_symlink_target)
413
def get_root_id(self):
414
"""Return the file_id for the root of this tree."""
415
raise NotImplementedError(self.get_root_id)
417
def annotate_iter(self, file_id,
418
default_revision=_mod_revision.CURRENT_REVISION):
419
"""Return an iterator of revision_id, line tuples.
421
For working trees (and mutable trees in general), the special
422
revision_id 'current:' will be used for lines that are new in this
423
tree, e.g. uncommitted changes.
424
:param file_id: The file to produce an annotated version from
425
:param default_revision: For lines that don't match a basis, mark them
426
with this revision id. Not all implementations will make use of
429
raise NotImplementedError(self.annotate_iter)
431
def _get_plan_merge_data(self, file_id, other, base):
432
from breezy import versionedfile
433
vf = versionedfile._PlanMergeVersionedFile(file_id)
434
last_revision_a = self._get_file_revision(file_id, vf, 'this:')
435
last_revision_b = other._get_file_revision(file_id, vf, 'other:')
437
last_revision_base = None
439
last_revision_base = base._get_file_revision(file_id, vf, 'base:')
440
return vf, last_revision_a, last_revision_b, last_revision_base
442
def plan_file_merge(self, file_id, other, base=None):
443
"""Generate a merge plan based on annotations.
445
If the file contains uncommitted changes in this tree, they will be
446
attributed to the 'current:' pseudo-revision. If the file contains
447
uncommitted changes in the other tree, they will be assigned to the
448
'other:' pseudo-revision.
450
data = self._get_plan_merge_data(file_id, other, base)
451
vf, last_revision_a, last_revision_b, last_revision_base = data
452
return vf.plan_merge(last_revision_a, last_revision_b,
455
def plan_file_lca_merge(self, file_id, other, base=None):
456
"""Generate a merge plan based lca-newness.
458
If the file contains uncommitted changes in this tree, they will be
459
attributed to the 'current:' pseudo-revision. If the file contains
460
uncommitted changes in the other tree, they will be assigned to the
461
'other:' pseudo-revision.
463
data = self._get_plan_merge_data(file_id, other, base)
464
vf, last_revision_a, last_revision_b, last_revision_base = data
465
return vf.plan_lca_merge(last_revision_a, last_revision_b,
468
def _iter_parent_trees(self):
469
"""Iterate through parent trees, defaulting to Tree.revision_tree."""
470
for revision_id in self.get_parent_ids():
472
yield self.revision_tree(revision_id)
473
except errors.NoSuchRevisionInTree:
474
yield self.repository.revision_tree(revision_id)
476
def _get_file_revision(self, file_id, vf, tree_revision):
477
"""Ensure that file_id, tree_revision is in vf to plan the merge."""
479
if getattr(self, '_repository', None) is None:
480
last_revision = tree_revision
481
parent_keys = [(file_id, t.get_file_revision(file_id)) for t in
482
self._iter_parent_trees()]
483
vf.add_lines((file_id, last_revision), parent_keys,
484
self.get_file_lines(file_id))
485
repo = self.branch.repository
488
last_revision = self.get_file_revision(file_id)
489
base_vf = self._repository.texts
490
if base_vf not in vf.fallback_versionedfiles:
491
vf.fallback_versionedfiles.append(base_vf)
494
def _check_retrieved(self, ie, f):
497
fp = osutils.fingerprint_file(f)
500
if ie.text_size is not None:
501
if ie.text_size != fp['size']:
502
raise errors.BzrError(
503
"mismatched size for file %r in %r" %
504
(ie.file_id, self._store),
505
["inventory expects %d bytes" % ie.text_size,
506
"file is actually %d bytes" % fp['size'],
507
"store is probably damaged/corrupt"])
509
if ie.text_sha1 != fp['sha1']:
510
raise errors.BzrError("wrong SHA-1 for file %r in %r" %
511
(ie.file_id, self._store),
512
["inventory expects %s" % ie.text_sha1,
513
"file is actually %s" % fp['sha1'],
514
"store is probably damaged/corrupt"])
516
def path2id(self, path):
517
"""Return the id for path in this tree."""
518
raise NotImplementedError(self.path2id)
520
def paths2ids(self, paths, trees=[], require_versioned=True):
521
"""Return all the ids that can be reached by walking from paths.
523
Each path is looked up in this tree and any extras provided in
524
trees, and this is repeated recursively: the children in an extra tree
525
of a directory that has been renamed under a provided path in this tree
526
are all returned, even if none exist under a provided path in this
527
tree, and vice versa.
529
:param paths: An iterable of paths to start converting to ids from.
530
Alternatively, if paths is None, no ids should be calculated and None
531
will be returned. This is offered to make calling the api unconditional
532
for code that *might* take a list of files.
533
:param trees: Additional trees to consider.
534
:param require_versioned: If False, do not raise NotVersionedError if
535
an element of paths is not versioned in this tree and all of trees.
537
return find_ids_across_trees(paths, [self] + list(trees), require_versioned)
539
def iter_children(self, file_id):
540
"""Iterate over the file ids of the children of an entry.
542
:param file_id: File id of the entry
543
:return: Iterator over child file ids.
545
raise NotImplementedError(self.iter_children)
548
"""Lock this tree for multiple read only operations.
550
:return: A breezy.lock.LogicalLockResult.
554
def revision_tree(self, revision_id):
555
"""Obtain a revision tree for the revision revision_id.
557
The intention of this method is to allow access to possibly cached
558
tree data. Implementors of this method should raise NoSuchRevision if
559
the tree is not locally available, even if they could obtain the
560
tree via a repository or some other means. Callers are responsible
561
for finding the ultimate source for a revision tree.
563
:param revision_id: The revision_id of the requested tree.
565
:raises: NoSuchRevision if the tree cannot be obtained.
567
raise errors.NoSuchRevisionInTree(self, revision_id)
570
"""What files are present in this tree and unknown.
572
:return: an iterator over the unknown files.
579
def filter_unversioned_files(self, paths):
580
"""Filter out paths that are versioned.
582
:return: set of paths.
584
raise NotImplementedError(self.filter_unversioned_files)
586
def walkdirs(self, prefix=""):
587
"""Walk the contents of this tree from path down.
589
This yields all the data about the contents of a directory at a time.
590
After each directory has been yielded, if the caller has mutated the
591
list to exclude some directories, they are then not descended into.
593
The data yielded is of the form:
594
((directory-relpath, directory-path-from-root, directory-fileid),
595
[(relpath, basename, kind, lstat, path_from_tree_root, file_id,
596
versioned_kind), ...]),
597
- directory-relpath is the containing dirs relpath from prefix
598
- directory-path-from-root is the containing dirs path from /
599
- directory-fileid is the id of the directory if it is versioned.
600
- relpath is the relative path within the subtree being walked.
601
- basename is the basename
602
- kind is the kind of the file now. If unknonwn then the file is not
603
present within the tree - but it may be recorded as versioned. See
605
- lstat is the stat data *if* the file was statted.
606
- path_from_tree_root is the path from the root of the tree.
607
- file_id is the file_id if the entry is versioned.
608
- versioned_kind is the kind of the file as last recorded in the
609
versioning system. If 'unknown' the file is not versioned.
610
One of 'kind' and 'versioned_kind' must not be 'unknown'.
612
:param prefix: Start walking from prefix within the tree rather than
613
at the root. This allows one to walk a subtree but get paths that are
614
relative to a tree rooted higher up.
615
:return: an iterator over the directory data.
617
raise NotImplementedError(self.walkdirs)
619
def supports_content_filtering(self):
622
def _content_filter_stack(self, path=None, file_id=None):
623
"""The stack of content filters for a path if filtering is supported.
625
Readers will be applied in first-to-last order.
626
Writers will be applied in last-to-first order.
627
Either the path or the file-id needs to be provided.
629
:param path: path relative to the root of the tree
631
:param file_id: file_id or None if unknown
632
:return: the list of filters - [] if there are none
634
filter_pref_names = filters._get_registered_names()
635
if len(filter_pref_names) == 0:
638
path = self.id2path(file_id)
639
prefs = next(self.iter_search_rules([path], filter_pref_names))
640
stk = filters._get_filter_stack_for(prefs)
641
if 'filters' in debug.debug_flags:
642
trace.note(gettext("*** {0} content-filter: {1} => {2!r}").format(path,prefs,stk))
645
def _content_filter_stack_provider(self):
646
"""A function that returns a stack of ContentFilters.
648
The function takes a path (relative to the top of the tree) and a
649
file-id as parameters.
651
:return: None if content filtering is not supported by this tree.
653
if self.supports_content_filtering():
654
return lambda path, file_id: \
655
self._content_filter_stack(path, file_id)
659
def iter_search_rules(self, path_names, pref_names=None,
660
_default_searcher=None):
661
"""Find the preferences for filenames in a tree.
663
:param path_names: an iterable of paths to find attributes for.
664
Paths are given relative to the root of the tree.
665
:param pref_names: the list of preferences to lookup - None for all
666
:param _default_searcher: private parameter to assist testing - don't use
667
:return: an iterator of tuple sequences, one per path-name.
668
See _RulesSearcher.get_items for details on the tuple sequence.
670
if _default_searcher is None:
671
_default_searcher = rules._per_user_searcher
672
searcher = self._get_rules_searcher(_default_searcher)
673
if searcher is not None:
674
if pref_names is not None:
675
for path in path_names:
676
yield searcher.get_selected_items(path, pref_names)
678
for path in path_names:
679
yield searcher.get_items(path)
681
def _get_rules_searcher(self, default_searcher):
682
"""Get the RulesSearcher for this tree given the default one."""
683
searcher = default_searcher
687
class InventoryTree(Tree):
688
"""A tree that relies on an inventory for its metadata.
690
Trees contain an `Inventory` object, and also know how to retrieve
691
file texts mentioned in the inventory, either from a working
692
directory or from a store.
694
It is possible for trees to contain files that are not described
695
in their inventory or vice versa; for this use `filenames()`.
697
Subclasses should set the _inventory attribute, which is considered
698
private to external API users.
701
def get_canonical_inventory_paths(self, paths):
702
"""Like get_canonical_inventory_path() but works on multiple items.
704
:param paths: A sequence of paths relative to the root of the tree.
705
:return: A list of paths, with each item the corresponding input path
706
adjusted to account for existing elements that match case
709
return list(self._yield_canonical_inventory_paths(paths))
711
def get_canonical_inventory_path(self, path):
712
"""Returns the first inventory item that case-insensitively matches path.
714
If a path matches exactly, it is returned. If no path matches exactly
715
but more than one path matches case-insensitively, it is implementation
716
defined which is returned.
718
If no path matches case-insensitively, the input path is returned, but
719
with as many path entries that do exist changed to their canonical
722
If you need to resolve many names from the same tree, you should
723
use get_canonical_inventory_paths() to avoid O(N) behaviour.
725
:param path: A paths relative to the root of the tree.
726
:return: The input path adjusted to account for existing elements
727
that match case insensitively.
729
return next(self._yield_canonical_inventory_paths([path]))
731
def _yield_canonical_inventory_paths(self, paths):
733
# First, if the path as specified exists exactly, just use it.
734
if self.path2id(path) is not None:
738
cur_id = self.get_root_id()
740
bit_iter = iter(path.split("/"))
744
for child in self.iter_children(cur_id):
746
# XXX: it seem like if the child is known to be in the
747
# tree, we shouldn't need to go from its id back to
748
# its path -- mbp 2010-02-11
750
# XXX: it seems like we could be more efficient
751
# by just directly looking up the original name and
752
# only then searching all children; also by not
753
# chopping paths so much. -- mbp 2010-02-11
754
child_base = os.path.basename(self.id2path(child))
755
if (child_base == elt):
756
# if we found an exact match, we can stop now; if
757
# we found an approximate match we need to keep
758
# searching because there might be an exact match
761
new_path = osutils.pathjoin(cur_path, child_base)
763
elif child_base.lower() == lelt:
765
new_path = osutils.pathjoin(cur_path, child_base)
766
except errors.NoSuchId:
767
# before a change is committed we can see this error...
772
# got to the end of this directory and no entries matched.
773
# Return what matched so far, plus the rest as specified.
774
cur_path = osutils.pathjoin(cur_path, elt, *list(bit_iter))
779
def _get_root_inventory(self):
780
return self._inventory
782
root_inventory = property(_get_root_inventory,
783
doc="Root inventory of this tree")
785
def _unpack_file_id(self, file_id):
786
"""Find the inventory and inventory file id for a tree file id.
788
:param file_id: The tree file id, as bytestring or tuple
789
:return: Inventory and inventory file id
791
if isinstance(file_id, tuple):
792
if len(file_id) != 1:
793
raise ValueError("nested trees not yet supported: %r" % file_id)
795
return self.root_inventory, file_id
798
def path2id(self, path):
799
"""Return the id for path in this tree."""
800
return self._path2inv_file_id(path)[1]
802
def _path2inv_file_id(self, path):
803
"""Lookup a inventory and inventory file id by path.
805
:param path: Path to look up
806
:return: tuple with inventory and inventory file id
808
# FIXME: Support nested trees
809
return self.root_inventory, self.root_inventory.path2id(path)
811
def id2path(self, file_id):
812
"""Return the path for a file id.
816
inventory, file_id = self._unpack_file_id(file_id)
817
return inventory.id2path(file_id)
819
def has_id(self, file_id):
820
inventory, file_id = self._unpack_file_id(file_id)
821
return inventory.has_id(file_id)
823
def has_or_had_id(self, file_id):
824
inventory, file_id = self._unpack_file_id(file_id)
825
return inventory.has_id(file_id)
827
def all_file_ids(self):
828
return {entry.file_id for path, entry in self.iter_entries_by_dir()}
830
def filter_unversioned_files(self, paths):
831
"""Filter out paths that are versioned.
833
:return: set of paths.
835
# NB: we specifically *don't* call self.has_filename, because for
836
# WorkingTrees that can indicate files that exist on disk but that
838
return set((p for p in paths if self.path2id(p) is None))
841
def iter_entries_by_dir(self, specific_file_ids=None, yield_parents=False):
842
"""Walk the tree in 'by_dir' order.
844
This will yield each entry in the tree as a (path, entry) tuple.
845
The order that they are yielded is:
847
See Tree.iter_entries_by_dir for details.
849
:param yield_parents: If True, yield the parents from the root leading
850
down to specific_file_ids that have been requested. This has no
851
impact if specific_file_ids is None.
853
if specific_file_ids is None:
854
inventory_file_ids = None
856
inventory_file_ids = []
857
for tree_file_id in specific_file_ids:
858
inventory, inv_file_id = self._unpack_file_id(tree_file_id)
859
if not inventory is self.root_inventory: # for now
860
raise AssertionError("%r != %r" % (
861
inventory, self.root_inventory))
862
inventory_file_ids.append(inv_file_id)
863
# FIXME: Handle nested trees
864
return self.root_inventory.iter_entries_by_dir(
865
specific_file_ids=inventory_file_ids, yield_parents=yield_parents)
868
def iter_child_entries(self, file_id, path=None):
869
inv, inv_file_id = self._unpack_file_id(file_id)
870
return iter(viewvalues(inv[inv_file_id].children))
872
def iter_children(self, file_id, path=None):
873
"""See Tree.iter_children."""
874
entry = self.iter_entries_by_dir([file_id]).next()[1]
875
for child in viewvalues(getattr(entry, 'children', {})):
879
def find_ids_across_trees(filenames, trees, require_versioned=True):
880
"""Find the ids corresponding to specified filenames.
882
All matches in all trees will be used, and all children of matched
883
directories will be used.
885
:param filenames: The filenames to find file_ids for (if None, returns
887
:param trees: The trees to find file_ids within
888
:param require_versioned: if true, all specified filenames must occur in
890
:return: a set of file ids for the specified filenames and their children.
894
specified_path_ids = _find_ids_across_trees(filenames, trees,
896
return _find_children_across_trees(specified_path_ids, trees)
899
def _find_ids_across_trees(filenames, trees, require_versioned):
900
"""Find the ids corresponding to specified filenames.
902
All matches in all trees will be used, but subdirectories are not scanned.
904
:param filenames: The filenames to find file_ids for
905
:param trees: The trees to find file_ids within
906
:param require_versioned: if true, all specified filenames must occur in
908
:return: a set of file ids for the specified filenames
911
interesting_ids = set()
912
for tree_path in filenames:
915
file_id = tree.path2id(tree_path)
916
if file_id is not None:
917
interesting_ids.add(file_id)
920
not_versioned.append(tree_path)
921
if len(not_versioned) > 0 and require_versioned:
922
raise errors.PathsNotVersionedError(not_versioned)
923
return interesting_ids
926
def _find_children_across_trees(specified_ids, trees):
927
"""Return a set including specified ids and their children.
929
All matches in all trees will be used.
931
:param trees: The trees to find file_ids within
932
:return: a set containing all specified ids and their children
934
interesting_ids = set(specified_ids)
935
pending = interesting_ids
936
# now handle children of interesting ids
937
# we loop so that we handle all children of each id in both trees
938
while len(pending) > 0:
940
for file_id in pending:
942
if not tree.has_or_had_id(file_id):
944
for child_id in tree.iter_children(file_id):
945
if child_id not in interesting_ids:
946
new_pending.add(child_id)
947
interesting_ids.update(new_pending)
948
pending = new_pending
949
return interesting_ids
952
class InterTree(InterObject):
953
"""This class represents operations taking place between two Trees.
955
Its instances have methods like 'compare' and contain references to the
956
source and target trees these operations are to be carried out on.
958
Clients of breezy should not need to use InterTree directly, rather they
959
should use the convenience methods on Tree such as 'Tree.compare()' which
960
will pass through to InterTree as appropriate.
963
# Formats that will be used to test this InterTree. If both are
964
# None, this InterTree will not be tested (e.g. because a complex
966
_matching_from_tree_format = None
967
_matching_to_tree_format = None
972
def is_compatible(kls, source, target):
973
# The default implementation is naive and uses the public API, so
974
# it works for all trees.
977
def _changes_from_entries(self, source_entry, target_entry,
978
source_path=None, target_path=None):
979
"""Generate a iter_changes tuple between source_entry and target_entry.
981
:param source_entry: An inventory entry from self.source, or None.
982
:param target_entry: An inventory entry from self.target, or None.
983
:param source_path: The path of source_entry, if known. If not known
984
it will be looked up.
985
:param target_path: The path of target_entry, if known. If not known
986
it will be looked up.
987
:return: A tuple, item 0 of which is an iter_changes result tuple, and
988
item 1 is True if there are any changes in the result tuple.
990
if source_entry is None:
991
if target_entry is None:
993
file_id = target_entry.file_id
995
file_id = source_entry.file_id
996
if source_entry is not None:
997
source_versioned = True
998
source_name = source_entry.name
999
source_parent = source_entry.parent_id
1000
if source_path is None:
1001
source_path = self.source.id2path(file_id)
1002
source_kind, source_executable, source_stat = \
1003
self.source._comparison_data(source_entry, source_path)
1005
source_versioned = False
1007
source_parent = None
1009
source_executable = None
1010
if target_entry is not None:
1011
target_versioned = True
1012
target_name = target_entry.name
1013
target_parent = target_entry.parent_id
1014
if target_path is None:
1015
target_path = self.target.id2path(file_id)
1016
target_kind, target_executable, target_stat = \
1017
self.target._comparison_data(target_entry, target_path)
1019
target_versioned = False
1021
target_parent = None
1023
target_executable = None
1024
versioned = (source_versioned, target_versioned)
1025
kind = (source_kind, target_kind)
1026
changed_content = False
1027
if source_kind != target_kind:
1028
changed_content = True
1029
elif source_kind == 'file':
1030
if not self.file_content_matches(file_id, file_id, source_path,
1031
target_path, source_stat, target_stat):
1032
changed_content = True
1033
elif source_kind == 'symlink':
1034
if (self.source.get_symlink_target(file_id) !=
1035
self.target.get_symlink_target(file_id)):
1036
changed_content = True
1037
elif source_kind == 'tree-reference':
1038
if (self.source.get_reference_revision(file_id, source_path)
1039
!= self.target.get_reference_revision(file_id, target_path)):
1040
changed_content = True
1041
parent = (source_parent, target_parent)
1042
name = (source_name, target_name)
1043
executable = (source_executable, target_executable)
1044
if (changed_content is not False or versioned[0] != versioned[1]
1045
or parent[0] != parent[1] or name[0] != name[1] or
1046
executable[0] != executable[1]):
1050
return (file_id, (source_path, target_path), changed_content,
1051
versioned, parent, name, kind, executable), changes
1054
def compare(self, want_unchanged=False, specific_files=None,
1055
extra_trees=None, require_versioned=False, include_root=False,
1056
want_unversioned=False):
1057
"""Return the changes from source to target.
1059
:return: A TreeDelta.
1060
:param specific_files: An optional list of file paths to restrict the
1061
comparison to. When mapping filenames to ids, all matches in all
1062
trees (including optional extra_trees) are used, and all children of
1063
matched directories are included.
1064
:param want_unchanged: An optional boolean requesting the inclusion of
1065
unchanged entries in the result.
1066
:param extra_trees: An optional list of additional trees to use when
1067
mapping the contents of specific_files (paths) to file_ids.
1068
:param require_versioned: An optional boolean (defaults to False). When
1069
supplied and True all the 'specific_files' must be versioned, or
1070
a PathsNotVersionedError will be thrown.
1071
:param want_unversioned: Scan for unversioned paths.
1073
trees = (self.source,)
1074
if extra_trees is not None:
1075
trees = trees + tuple(extra_trees)
1076
# target is usually the newer tree:
1077
specific_file_ids = self.target.paths2ids(specific_files, trees,
1078
require_versioned=require_versioned)
1079
if specific_files and not specific_file_ids:
1080
# All files are unversioned, so just return an empty delta
1081
# _compare_trees would think we want a complete delta
1082
result = delta.TreeDelta()
1083
fake_entry = inventory.InventoryFile('unused', 'unused', 'unused')
1084
result.unversioned = [(path, None,
1085
self.target._comparison_data(fake_entry, path)[0]) for path in
1088
return delta._compare_trees(self.source, self.target, want_unchanged,
1089
specific_files, include_root, extra_trees=extra_trees,
1090
require_versioned=require_versioned,
1091
want_unversioned=want_unversioned)
1093
def iter_changes(self, include_unchanged=False,
1094
specific_files=None, pb=None, extra_trees=[],
1095
require_versioned=True, want_unversioned=False):
1096
"""Generate an iterator of changes between trees.
1098
A tuple is returned:
1099
(file_id, (path_in_source, path_in_target),
1100
changed_content, versioned, parent, name, kind,
1103
Changed_content is True if the file's content has changed. This
1104
includes changes to its kind, and to a symlink's target.
1106
versioned, parent, name, kind, executable are tuples of (from, to).
1107
If a file is missing in a tree, its kind is None.
1109
Iteration is done in parent-to-child order, relative to the target
1112
There is no guarantee that all paths are in sorted order: the
1113
requirement to expand the search due to renames may result in children
1114
that should be found early being found late in the search, after
1115
lexically later results have been returned.
1116
:param require_versioned: Raise errors.PathsNotVersionedError if a
1117
path in the specific_files list is not versioned in one of
1118
source, target or extra_trees.
1119
:param specific_files: An optional list of file paths to restrict the
1120
comparison to. When mapping filenames to ids, all matches in all
1121
trees (including optional extra_trees) are used, and all children
1122
of matched directories are included. The parents in the target tree
1123
of the specific files up to and including the root of the tree are
1124
always evaluated for changes too.
1125
:param want_unversioned: Should unversioned files be returned in the
1126
output. An unversioned file is defined as one with (False, False)
1127
for the versioned pair.
1129
lookup_trees = [self.source]
1131
lookup_trees.extend(extra_trees)
1132
# The ids of items we need to examine to insure delta consistency.
1133
precise_file_ids = set()
1134
changed_file_ids = []
1135
if specific_files == []:
1136
specific_file_ids = []
1138
specific_file_ids = self.target.paths2ids(specific_files,
1139
lookup_trees, require_versioned=require_versioned)
1140
if specific_files is not None:
1141
# reparented or added entries must have their parents included
1142
# so that valid deltas can be created. The seen_parents set
1143
# tracks the parents that we need to have.
1144
# The seen_dirs set tracks directory entries we've yielded.
1145
# After outputting version object in to_entries we set difference
1146
# the two seen sets and start checking parents.
1147
seen_parents = set()
1149
if want_unversioned:
1150
all_unversioned = sorted([(p.split('/'), p) for p in
1151
self.target.extras()
1152
if specific_files is None or
1153
osutils.is_inside_any(specific_files, p)])
1154
all_unversioned = collections.deque(all_unversioned)
1156
all_unversioned = collections.deque()
1158
from_entries_by_dir = list(self.source.iter_entries_by_dir(
1159
specific_file_ids=specific_file_ids))
1160
from_data = dict((e.file_id, (p, e)) for p, e in from_entries_by_dir)
1161
to_entries_by_dir = list(self.target.iter_entries_by_dir(
1162
specific_file_ids=specific_file_ids))
1163
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
1165
# the unversioned path lookup only occurs on real trees - where there
1166
# can be extras. So the fake_entry is solely used to look up
1167
# executable it values when execute is not supported.
1168
fake_entry = inventory.InventoryFile('unused', 'unused', 'unused')
1169
for target_path, target_entry in to_entries_by_dir:
1170
while (all_unversioned and
1171
all_unversioned[0][0] < target_path.split('/')):
1172
unversioned_path = all_unversioned.popleft()
1173
target_kind, target_executable, target_stat = \
1174
self.target._comparison_data(fake_entry, unversioned_path[1])
1175
yield (None, (None, unversioned_path[1]), True, (False, False),
1177
(None, unversioned_path[0][-1]),
1178
(None, target_kind),
1179
(None, target_executable))
1180
source_path, source_entry = from_data.get(target_entry.file_id,
1182
result, changes = self._changes_from_entries(source_entry,
1183
target_entry, source_path=source_path, target_path=target_path)
1184
to_paths[result[0]] = result[1][1]
1189
pb.update('comparing files', entry_count, num_entries)
1190
if changes or include_unchanged:
1191
if specific_file_ids is not None:
1192
new_parent_id = result[4][1]
1193
precise_file_ids.add(new_parent_id)
1194
changed_file_ids.append(result[0])
1196
# Ensure correct behaviour for reparented/added specific files.
1197
if specific_files is not None:
1198
# Record output dirs
1199
if result[6][1] == 'directory':
1200
seen_dirs.add(result[0])
1201
# Record parents of reparented/added entries.
1202
versioned = result[3]
1204
if not versioned[0] or parents[0] != parents[1]:
1205
seen_parents.add(parents[1])
1206
while all_unversioned:
1207
# yield any trailing unversioned paths
1208
unversioned_path = all_unversioned.popleft()
1209
to_kind, to_executable, to_stat = \
1210
self.target._comparison_data(fake_entry, unversioned_path[1])
1211
yield (None, (None, unversioned_path[1]), True, (False, False),
1213
(None, unversioned_path[0][-1]),
1215
(None, to_executable))
1216
# Yield all remaining source paths
1217
for path, from_entry in from_entries_by_dir:
1218
file_id = from_entry.file_id
1219
if file_id in to_paths:
1222
if not self.target.has_id(file_id):
1223
# common case - paths we have not emitted are not present in
1227
to_path = self.target.id2path(file_id)
1230
pb.update('comparing files', entry_count, num_entries)
1231
versioned = (True, False)
1232
parent = (from_entry.parent_id, None)
1233
name = (from_entry.name, None)
1234
from_kind, from_executable, stat_value = \
1235
self.source._comparison_data(from_entry, path)
1236
kind = (from_kind, None)
1237
executable = (from_executable, None)
1238
changed_content = from_kind is not None
1239
# the parent's path is necessarily known at this point.
1240
changed_file_ids.append(file_id)
1241
yield(file_id, (path, to_path), changed_content, versioned, parent,
1242
name, kind, executable)
1243
changed_file_ids = set(changed_file_ids)
1244
if specific_file_ids is not None:
1245
for result in self._handle_precise_ids(precise_file_ids,
1249
def _get_entry(self, tree, file_id):
1250
"""Get an inventory entry from a tree, with missing entries as None.
1252
If the tree raises NotImplementedError on accessing .inventory, then
1253
this is worked around using iter_entries_by_dir on just the file id
1256
:param tree: The tree to lookup the entry in.
1257
:param file_id: The file_id to lookup.
1260
inventory = tree.root_inventory
1261
except NotImplementedError:
1262
# No inventory available.
1264
iterator = tree.iter_entries_by_dir(specific_file_ids=[file_id])
1265
return iterator.next()[1]
1266
except StopIteration:
1270
return inventory[file_id]
1271
except errors.NoSuchId:
1274
def _handle_precise_ids(self, precise_file_ids, changed_file_ids,
1275
discarded_changes=None):
1276
"""Fill out a partial iter_changes to be consistent.
1278
:param precise_file_ids: The file ids of parents that were seen during
1280
:param changed_file_ids: The file ids of already emitted items.
1281
:param discarded_changes: An optional dict of precalculated
1282
iter_changes items which the partial iter_changes had not output
1284
:return: A generator of iter_changes items to output.
1286
# process parents of things that had changed under the users
1287
# requested paths to prevent incorrect paths or parent ids which
1288
# aren't in the tree.
1289
while precise_file_ids:
1290
precise_file_ids.discard(None)
1291
# Don't emit file_ids twice
1292
precise_file_ids.difference_update(changed_file_ids)
1293
if not precise_file_ids:
1295
# If the there was something at a given output path in source, we
1296
# have to include the entry from source in the delta, or we would
1297
# be putting this entry into a used path.
1299
for parent_id in precise_file_ids:
1301
paths.append(self.target.id2path(parent_id))
1302
except errors.NoSuchId:
1303
# This id has been dragged in from the source by delta
1304
# expansion and isn't present in target at all: we don't
1305
# need to check for path collisions on it.
1308
old_id = self.source.path2id(path)
1309
precise_file_ids.add(old_id)
1310
precise_file_ids.discard(None)
1311
current_ids = precise_file_ids
1312
precise_file_ids = set()
1313
# We have to emit all of precise_file_ids that have been altered.
1314
# We may have to output the children of some of those ids if any
1315
# directories have stopped being directories.
1316
for file_id in current_ids:
1318
if discarded_changes:
1319
result = discarded_changes.get(file_id)
1324
old_entry = self._get_entry(self.source, file_id)
1325
new_entry = self._get_entry(self.target, file_id)
1326
result, changes = self._changes_from_entries(
1327
old_entry, new_entry)
1330
# Get this parents parent to examine.
1331
new_parent_id = result[4][1]
1332
precise_file_ids.add(new_parent_id)
1334
if (result[6][0] == 'directory' and
1335
result[6][1] != 'directory'):
1336
# This stopped being a directory, the old children have
1338
if old_entry is None:
1339
# Reusing a discarded change.
1340
old_entry = self._get_entry(self.source, file_id)
1341
precise_file_ids.update(
1342
self.source.iter_children(file_id))
1343
changed_file_ids.add(result[0])
1347
def file_content_matches(self, source_file_id, target_file_id,
1348
source_path=None, target_path=None, source_stat=None, target_stat=None):
1349
"""Check if two files are the same in the source and target trees.
1351
This only checks that the contents of the files are the same,
1352
it does not touch anything else.
1354
:param source_file_id: File id of the file in the source tree
1355
:param target_file_id: File id of the file in the target tree
1356
:param source_path: Path of the file in the source tree
1357
:param target_path: Path of the file in the target tree
1358
:param source_stat: Optional stat value of the file in the source tree
1359
:param target_stat: Optional stat value of the file in the target tree
1360
:return: Boolean indicating whether the files have the same contents
1362
source_verifier_kind, source_verifier_data = self.source.get_file_verifier(
1363
source_file_id, source_path, source_stat)
1364
target_verifier_kind, target_verifier_data = self.target.get_file_verifier(
1365
target_file_id, target_path, target_stat)
1366
if source_verifier_kind == target_verifier_kind:
1367
return (source_verifier_data == target_verifier_data)
1368
# Fall back to SHA1 for now
1369
if source_verifier_kind != "SHA1":
1370
source_sha1 = self.source.get_file_sha1(source_file_id,
1371
source_path, source_stat)
1373
source_sha1 = source_verifier_data
1374
if target_verifier_kind != "SHA1":
1375
target_sha1 = self.target.get_file_sha1(target_file_id,
1376
target_path, target_stat)
1378
target_sha1 = target_verifier_data
1379
return (source_sha1 == target_sha1)
1381
InterTree.register_optimiser(InterTree)
1384
class MultiWalker(object):
1385
"""Walk multiple trees simultaneously, getting combined results."""
1387
# Note: This could be written to not assume you can do out-of-order
1388
# lookups. Instead any nodes that don't match in all trees could be
1389
# marked as 'deferred', and then returned in the final cleanup loop.
1390
# For now, I think it is "nicer" to return things as close to the
1391
# "master_tree" order as we can.
1393
def __init__(self, master_tree, other_trees):
1394
"""Create a new MultiWalker.
1396
All trees being walked must implement "iter_entries_by_dir()", such
1397
that they yield (path, object) tuples, where that object will have a
1398
'.file_id' member, that can be used to check equality.
1400
:param master_tree: All trees will be 'slaved' to the master_tree such
1401
that nodes in master_tree will be used as 'first-pass' sync points.
1402
Any nodes that aren't in master_tree will be merged in a second
1404
:param other_trees: A list of other trees to walk simultaneously.
1406
self._master_tree = master_tree
1407
self._other_trees = other_trees
1409
# Keep track of any nodes that were properly processed just out of
1410
# order, that way we don't return them at the end, we don't have to
1411
# track *all* processed file_ids, just the out-of-order ones
1412
self._out_of_order_processed = set()
1415
def _step_one(iterator):
1416
"""Step an iter_entries_by_dir iterator.
1418
:return: (has_more, path, ie)
1419
If has_more is False, path and ie will be None.
1422
path, ie = next(iterator)
1423
except StopIteration:
1424
return False, None, None
1426
return True, path, ie
1429
def _cmp_path_by_dirblock(path1, path2):
1430
"""Compare two paths based on what directory they are in.
1432
This generates a sort order, such that all children of a directory are
1433
sorted together, and grandchildren are in the same order as the
1434
children appear. But all grandchildren come after all children.
1436
:param path1: first path
1437
:param path2: the second path
1438
:return: negative number if ``path1`` comes first,
1439
0 if paths are equal
1440
and a positive number if ``path2`` sorts first
1442
# Shortcut this special case
1445
# This is stolen from _dirstate_helpers_py.py, only switching it to
1446
# Unicode objects. Consider using encode_utf8() and then using the
1447
# optimized versions, or maybe writing optimized unicode versions.
1448
if not isinstance(path1, unicode):
1449
raise TypeError("'path1' must be a unicode string, not %s: %r"
1450
% (type(path1), path1))
1451
if not isinstance(path2, unicode):
1452
raise TypeError("'path2' must be a unicode string, not %s: %r"
1453
% (type(path2), path2))
1454
return cmp(MultiWalker._path_to_key(path1),
1455
MultiWalker._path_to_key(path2))
1458
def _path_to_key(path):
1459
dirname, basename = osutils.split(path)
1460
return (dirname.split(u'/'), basename)
1462
def _lookup_by_file_id(self, extra_entries, other_tree, file_id):
1463
"""Lookup an inventory entry by file_id.
1465
This is called when an entry is missing in the normal order.
1466
Generally this is because a file was either renamed, or it was
1467
deleted/added. If the entry was found in the inventory and not in
1468
extra_entries, it will be added to self._out_of_order_processed
1470
:param extra_entries: A dictionary of {file_id: (path, ie)}. This
1471
should be filled with entries that were found before they were
1472
used. If file_id is present, it will be removed from the
1474
:param other_tree: The Tree to search, in case we didn't find the entry
1476
:param file_id: The file_id to look for
1477
:return: (path, ie) if found or (None, None) if not present.
1479
if file_id in extra_entries:
1480
return extra_entries.pop(file_id)
1481
# TODO: Is id2path better as the first call, or is
1482
# inventory[file_id] better as a first check?
1484
cur_path = other_tree.id2path(file_id)
1485
except errors.NoSuchId:
1487
if cur_path is None:
1490
self._out_of_order_processed.add(file_id)
1491
cur_ie = other_tree.root_inventory[file_id]
1492
return (cur_path, cur_ie)
1495
"""Match up the values in the different trees."""
1496
for result in self._walk_master_tree():
1498
self._finish_others()
1499
for result in self._walk_others():
1502
def _walk_master_tree(self):
1503
"""First pass, walk all trees in lock-step.
1505
When we are done, all nodes in the master_tree will have been
1506
processed. _other_walkers, _other_entries, and _others_extra will be
1507
set on 'self' for future processing.
1509
# This iterator has the most "inlining" done, because it tends to touch
1510
# every file in the tree, while the others only hit nodes that don't
1512
master_iterator = self._master_tree.iter_entries_by_dir()
1514
other_walkers = [other.iter_entries_by_dir()
1515
for other in self._other_trees]
1516
other_entries = [self._step_one(walker) for walker in other_walkers]
1517
# Track extra nodes in the other trees
1518
others_extra = [{} for _ in range(len(self._other_trees))]
1520
master_has_more = True
1521
step_one = self._step_one
1522
lookup_by_file_id = self._lookup_by_file_id
1523
out_of_order_processed = self._out_of_order_processed
1525
while master_has_more:
1526
(master_has_more, path, master_ie) = step_one(master_iterator)
1527
if not master_has_more:
1530
file_id = master_ie.file_id
1532
other_values_append = other_values.append
1533
next_other_entries = []
1534
next_other_entries_append = next_other_entries.append
1535
for idx, (other_has_more, other_path, other_ie) in enumerate(other_entries):
1536
if not other_has_more:
1537
other_values_append(lookup_by_file_id(
1538
others_extra[idx], self._other_trees[idx], file_id))
1539
next_other_entries_append((False, None, None))
1540
elif file_id == other_ie.file_id:
1541
# This is the critical code path, as most of the entries
1542
# should match between most trees.
1543
other_values_append((other_path, other_ie))
1544
next_other_entries_append(step_one(other_walkers[idx]))
1546
# This walker did not match, step it until it either
1547
# matches, or we know we are past the current walker.
1548
other_walker = other_walkers[idx]
1549
other_extra = others_extra[idx]
1550
while (other_has_more and
1551
self._cmp_path_by_dirblock(other_path, path) < 0):
1552
other_file_id = other_ie.file_id
1553
if other_file_id not in out_of_order_processed:
1554
other_extra[other_file_id] = (other_path, other_ie)
1555
other_has_more, other_path, other_ie = \
1556
step_one(other_walker)
1557
if other_has_more and other_ie.file_id == file_id:
1558
# We ended up walking to this point, match and step
1560
other_values_append((other_path, other_ie))
1561
other_has_more, other_path, other_ie = \
1562
step_one(other_walker)
1564
# This record isn't in the normal order, see if it
1566
other_values_append(lookup_by_file_id(
1567
other_extra, self._other_trees[idx], file_id))
1568
next_other_entries_append((other_has_more, other_path,
1570
other_entries = next_other_entries
1572
# We've matched all the walkers, yield this datapoint
1573
yield path, file_id, master_ie, other_values
1574
self._other_walkers = other_walkers
1575
self._other_entries = other_entries
1576
self._others_extra = others_extra
1578
def _finish_others(self):
1579
"""Finish walking the other iterators, so we get all entries."""
1580
for idx, info in enumerate(self._other_entries):
1581
other_extra = self._others_extra[idx]
1582
(other_has_more, other_path, other_ie) = info
1583
while other_has_more:
1584
other_file_id = other_ie.file_id
1585
if other_file_id not in self._out_of_order_processed:
1586
other_extra[other_file_id] = (other_path, other_ie)
1587
other_has_more, other_path, other_ie = \
1588
self._step_one(self._other_walkers[idx])
1589
del self._other_entries
1591
def _walk_others(self):
1592
"""Finish up by walking all the 'deferred' nodes."""
1593
# TODO: One alternative would be to grab all possible unprocessed
1594
# file_ids, and then sort by path, and then yield them. That
1595
# might ensure better ordering, in case a caller strictly
1596
# requires parents before children.
1597
for idx, other_extra in enumerate(self._others_extra):
1598
others = sorted(viewvalues(other_extra),
1599
key=lambda x: self._path_to_key(x[0]))
1600
for other_path, other_ie in others:
1601
file_id = other_ie.file_id
1602
# We don't need to check out_of_order_processed here, because
1603
# the lookup_by_file_id will be removing anything processed
1604
# from the extras cache
1605
other_extra.pop(file_id)
1606
other_values = [(None, None)] * idx
1607
other_values.append((other_path, other_ie))
1608
for alt_idx, alt_extra in enumerate(self._others_extra[idx+1:]):
1609
alt_idx = alt_idx + idx + 1
1610
alt_extra = self._others_extra[alt_idx]
1611
alt_tree = self._other_trees[alt_idx]
1612
other_values.append(self._lookup_by_file_id(
1613
alt_extra, alt_tree, file_id))
1614
yield other_path, file_id, None, other_values
added
removed
breezy/tree.py
