48
61
trees or versioned trees.
64
def changes_from(self, other, want_unchanged=False, specific_files=None,
65
extra_trees=None, require_versioned=False, include_root=False,
66
want_unversioned=False):
67
"""Return a TreeDelta of the changes from other to this tree.
69
:param other: A tree to compare with.
70
:param specific_files: An optional list of file paths to restrict the
71
comparison to. When mapping filenames to ids, all matches in all
72
trees (including optional extra_trees) are used, and all children of
73
matched directories are included.
74
:param want_unchanged: An optional boolean requesting the inclusion of
75
unchanged entries in the result.
76
:param extra_trees: An optional list of additional trees to use when
77
mapping the contents of specific_files (paths) to file_ids.
78
:param require_versioned: An optional boolean (defaults to False). When
79
supplied and True all the 'specific_files' must be versioned, or
80
a PathsNotVersionedError will be thrown.
81
:param want_unversioned: Scan for unversioned paths.
83
The comparison will be performed by an InterTree object looked up on
86
# Martin observes that Tree.changes_from returns a TreeDelta and this
87
# may confuse people, because the class name of the returned object is
88
# a synonym of the object referenced in the method name.
89
return InterTree.get(other, self).compare(
90
want_unchanged=want_unchanged,
91
specific_files=specific_files,
92
extra_trees=extra_trees,
93
require_versioned=require_versioned,
94
include_root=include_root,
95
want_unversioned=want_unversioned,
98
@symbol_versioning.deprecated_method(symbol_versioning.one_three)
99
def _iter_changes(self, *args, **kwargs):
100
return self.iter_changes(*args, **kwargs)
102
def iter_changes(self, from_tree, include_unchanged=False,
103
specific_files=None, pb=None, extra_trees=None,
104
require_versioned=True, want_unversioned=False):
105
intertree = InterTree.get(from_tree, self)
106
return intertree.iter_changes(include_unchanged, specific_files, pb,
107
extra_trees, require_versioned, want_unversioned=want_unversioned)
110
"""Get a list of the conflicts in the tree.
112
Each conflict is an instance of bzrlib.conflicts.Conflict.
114
return _mod_conflicts.ConflictList()
117
"""For trees that can have unversioned files, return all such paths."""
120
def get_parent_ids(self):
121
"""Get the parent ids for this tree.
123
:return: a list of parent ids. [] is returned to indicate
124
a tree with no parents.
125
:raises: BzrError if the parents are not known.
127
raise NotImplementedError(self.get_parent_ids)
51
129
def has_filename(self, filename):
52
130
"""True if the tree has given filename."""
53
raise NotImplementedError()
131
raise NotImplementedError(self.has_filename)
55
133
def has_id(self, file_id):
56
134
return self.inventory.has_id(file_id)
136
def __contains__(self, file_id):
137
return self.has_id(file_id)
58
139
def has_or_had_id(self, file_id):
59
140
if file_id == self.inventory.root.file_id:
61
142
return self.inventory.has_id(file_id)
144
def is_ignored(self, filename):
145
"""Check whether the filename is ignored by this tree.
147
:param filename: The relative filename within the tree.
148
:return: True if the filename is ignored.
65
152
def __iter__(self):
66
153
return iter(self.inventory)
155
def all_file_ids(self):
156
"""Iterate through all file ids, including ids for missing files."""
157
return set(self.inventory)
68
159
def id2path(self, file_id):
160
"""Return the path for a file id.
69
164
return self.inventory.id2path(file_id)
166
def is_control_filename(self, filename):
167
"""True if filename is the name of a control file in this tree.
169
:param filename: A filename within the tree. This is a relative path
170
from the root of this tree.
172
This is true IF and ONLY IF the filename is part of the meta data
173
that bzr controls in this tree. I.E. a random .bzr directory placed
174
on disk will not be a control file for this tree.
176
return self.bzrdir.is_control_filename(filename)
179
def iter_entries_by_dir(self, specific_file_ids=None):
180
"""Walk the tree in 'by_dir' order.
182
This will yield each entry in the tree as a (path, entry) tuple.
183
The order that they are yielded is:
185
Directories are walked in a depth-first lexicographical order,
186
however, whenever a directory is reached, all of its direct child
187
nodes are yielded in lexicographical order before yielding the
190
For example, in the tree::
200
The yield order (ignoring root) would be::
201
a, f, a/b, a/d, a/b/c, a/d/e, f/g
203
return self.inventory.iter_entries_by_dir(
204
specific_file_ids=specific_file_ids)
206
def iter_references(self):
207
for path, entry in self.iter_entries_by_dir():
208
if entry.kind == 'tree-reference':
209
yield path, entry.file_id
71
211
def kind(self, file_id):
72
raise NotImplementedError("subclasses must implement kind")
212
raise NotImplementedError("Tree subclass %s must implement kind"
213
% self.__class__.__name__)
215
def stored_kind(self, file_id):
216
"""File kind stored for this file_id.
218
May not match kind on disk for working trees. Always available
219
for versioned files, even when the file itself is missing.
221
return self.kind(file_id)
223
def path_content_summary(self, path):
224
"""Get a summary of the information about path.
226
:param path: A relative path within the tree.
227
:return: A tuple containing kind, size, exec, sha1-or-link.
228
Kind is always present (see tree.kind()).
229
size is present if kind is file, None otherwise.
230
exec is None unless kind is file and the platform supports the 'x'
232
sha1-or-link is the link target if kind is symlink, or the sha1 if
233
it can be obtained without reading the file.
235
raise NotImplementedError(self.path_content_summary)
237
def get_reference_revision(self, file_id, path=None):
238
raise NotImplementedError("Tree subclass %s must implement "
239
"get_reference_revision"
240
% self.__class__.__name__)
242
def _comparison_data(self, entry, path):
243
"""Return a tuple of kind, executable, stat_value for a file.
245
entry may be None if there is no inventory entry for the file, but
246
path must always be supplied.
248
kind is None if there is no file present (even if an inventory id is
249
present). executable is False for non-file entries.
251
raise NotImplementedError(self._comparison_data)
253
def _file_size(self, entry, stat_value):
254
raise NotImplementedError(self._file_size)
74
256
def _get_inventory(self):
75
257
return self._inventory
259
def get_file(self, file_id, path=None):
260
"""Return a file object for the file file_id in the tree.
262
If both file_id and path are defined, it is implementation defined as
263
to which one is used.
265
raise NotImplementedError(self.get_file)
267
def get_file_text(self, file_id, path=None):
268
"""Return the byte content of a file.
270
:param file_id: The file_id of the file.
271
:param path: The path of the file.
272
If both file_id and path are supplied, an implementation may use
275
my_file = self.get_file(file_id, path)
277
return my_file.read()
281
def get_file_lines(self, file_id, path=None):
282
"""Return the content of a file, as lines.
284
:param file_id: The file_id of the file.
285
:param path: The path of the file.
286
If both file_id and path are supplied, an implementation may use
289
return osutils.split_lines(self.get_file_text(file_id, path))
291
def get_file_mtime(self, file_id, path=None):
292
"""Return the modification time for a file.
294
:param file_id: The handle for this file.
295
:param path: The path that this file can be found at.
296
These must point to the same object.
298
raise NotImplementedError(self.get_file_mtime)
300
def get_file_size(self, file_id):
301
"""Return the size of a file in bytes.
303
This applies only to regular files. If invoked on directories or
304
symlinks, it will return None.
305
:param file_id: The file-id of the file
307
raise NotImplementedError(self.get_file_size)
77
309
def get_file_by_path(self, path):
78
return self.get_file(self._inventory.path2id(path))
310
return self.get_file(self._inventory.path2id(path), path)
312
def iter_files_bytes(self, desired_files):
313
"""Iterate through file contents.
315
Files will not necessarily be returned in the order they occur in
316
desired_files. No specific order is guaranteed.
318
Yields pairs of identifier, bytes_iterator. identifier is an opaque
319
value supplied by the caller as part of desired_files. It should
320
uniquely identify the file version in the caller's context. (Examples:
321
an index number or a TreeTransform trans_id.)
323
bytes_iterator is an iterable of bytestrings for the file. The
324
kind of iterable and length of the bytestrings are unspecified, but for
325
this implementation, it is a tuple containing a single bytestring with
326
the complete text of the file.
328
:param desired_files: a list of (file_id, identifier) pairs
330
for file_id, identifier in desired_files:
331
# We wrap the string in a tuple so that we can return an iterable
332
# of bytestrings. (Technically, a bytestring is also an iterable
333
# of bytestrings, but iterating through each character is not
335
cur_file = (self.get_file_text(file_id),)
336
yield identifier, cur_file
338
def get_symlink_target(self, file_id):
339
"""Get the target for a given file_id.
341
It is assumed that the caller already knows that file_id is referencing
343
:param file_id: Handle for the symlink entry.
344
:return: The path the symlink points to.
346
raise NotImplementedError(self.get_symlink_target)
348
def get_canonical_inventory_paths(self, paths):
349
"""Like get_canonical_inventory_path() but works on multiple items.
351
:param paths: A sequence of paths relative to the root of the tree.
352
:return: A list of paths, with each item the corresponding input path
353
adjusted to account for existing elements that match case
356
return list(self._yield_canonical_inventory_paths(paths))
358
def get_canonical_inventory_path(self, path):
359
"""Returns the first inventory item that case-insensitively matches path.
361
If a path matches exactly, it is returned. If no path matches exactly
362
but more than one path matches case-insensitively, it is implementation
363
defined which is returned.
365
If no path matches case-insensitively, the input path is returned, but
366
with as many path entries that do exist changed to their canonical
369
If you need to resolve many names from the same tree, you should
370
use get_canonical_inventory_paths() to avoid O(N) behaviour.
372
:param path: A paths relative to the root of the tree.
373
:return: The input path adjusted to account for existing elements
374
that match case insensitively.
376
return self._yield_canonical_inventory_paths([path]).next()
378
def _yield_canonical_inventory_paths(self, paths):
380
# First, if the path as specified exists exactly, just use it.
381
if self.path2id(path) is not None:
385
cur_id = self.get_root_id()
387
bit_iter = iter(path.split("/"))
390
for child in self.iter_children(cur_id):
392
child_base = os.path.basename(self.id2path(child))
393
if child_base.lower() == lelt:
395
cur_path = osutils.pathjoin(cur_path, child_base)
398
# before a change is committed we can see this error...
401
# got to the end of this directory and no entries matched.
402
# Return what matched so far, plus the rest as specified.
403
cur_path = osutils.pathjoin(cur_path, elt, *list(bit_iter))
408
def get_root_id(self):
409
"""Return the file_id for the root of this tree."""
410
raise NotImplementedError(self.get_root_id)
412
def annotate_iter(self, file_id,
413
default_revision=_mod_revision.CURRENT_REVISION):
414
"""Return an iterator of revision_id, line tuples.
416
For working trees (and mutable trees in general), the special
417
revision_id 'current:' will be used for lines that are new in this
418
tree, e.g. uncommitted changes.
419
:param file_id: The file to produce an annotated version from
420
:param default_revision: For lines that don't match a basis, mark them
421
with this revision id. Not all implementations will make use of
424
raise NotImplementedError(self.annotate_iter)
426
def _get_plan_merge_data(self, file_id, other, base):
427
from bzrlib import merge, versionedfile
428
vf = versionedfile._PlanMergeVersionedFile(file_id)
429
last_revision_a = self._get_file_revision(file_id, vf, 'this:')
430
last_revision_b = other._get_file_revision(file_id, vf, 'other:')
432
last_revision_base = None
434
last_revision_base = base._get_file_revision(file_id, vf, 'base:')
435
return vf, last_revision_a, last_revision_b, last_revision_base
437
def plan_file_merge(self, file_id, other, base=None):
438
"""Generate a merge plan based on annotations.
440
If the file contains uncommitted changes in this tree, they will be
441
attributed to the 'current:' pseudo-revision. If the file contains
442
uncommitted changes in the other tree, they will be assigned to the
443
'other:' pseudo-revision.
445
data = self._get_plan_merge_data(file_id, other, base)
446
vf, last_revision_a, last_revision_b, last_revision_base = data
447
return vf.plan_merge(last_revision_a, last_revision_b,
450
def plan_file_lca_merge(self, file_id, other, base=None):
451
"""Generate a merge plan based lca-newness.
453
If the file contains uncommitted changes in this tree, they will be
454
attributed to the 'current:' pseudo-revision. If the file contains
455
uncommitted changes in the other tree, they will be assigned to the
456
'other:' pseudo-revision.
458
data = self._get_plan_merge_data(file_id, other, base)
459
vf, last_revision_a, last_revision_b, last_revision_base = data
460
return vf.plan_lca_merge(last_revision_a, last_revision_b,
463
def _iter_parent_trees(self):
464
"""Iterate through parent trees, defaulting to Tree.revision_tree."""
465
for revision_id in self.get_parent_ids():
467
yield self.revision_tree(revision_id)
468
except errors.NoSuchRevisionInTree:
469
yield self.repository.revision_tree(revision_id)
472
def _file_revision(revision_tree, file_id):
473
"""Determine the revision associated with a file in a given tree."""
474
revision_tree.lock_read()
476
return revision_tree.inventory[file_id].revision
478
revision_tree.unlock()
480
def _get_file_revision(self, file_id, vf, tree_revision):
481
"""Ensure that file_id, tree_revision is in vf to plan the merge."""
483
if getattr(self, '_repository', None) is None:
484
last_revision = tree_revision
485
parent_keys = [(file_id, self._file_revision(t, file_id)) for t in
486
self._iter_parent_trees()]
487
vf.add_lines((file_id, last_revision), parent_keys,
488
self.get_file(file_id).readlines())
489
repo = self.branch.repository
492
last_revision = self._file_revision(self, file_id)
493
base_vf = self._repository.texts
494
if base_vf not in vf.fallback_versionedfiles:
495
vf.fallback_versionedfiles.append(base_vf)
80
498
inventory = property(_get_inventory,
81
499
doc="Inventory of this Tree")
121
590
# are not versioned.
122
591
pred = self.inventory.has_filename
123
592
return set((p for p in paths if not pred(p)))
126
class RevisionTree(Tree):
127
"""Tree viewing a previous revision.
129
File text can be retrieved from the text store.
131
TODO: Some kind of `__repr__` method, but a good one
132
probably means knowing the branch and revision number,
133
or at least passing a description to the constructor.
136
def __init__(self, branch, inv, revision_id):
137
self._branch = branch
138
self._weave_store = branch.weave_store
139
self._inventory = inv
140
self._revision_id = revision_id
142
def get_revision_id(self):
143
"""Return the revision id associated with this tree."""
144
return self._revision_id
146
def get_weave(self, file_id):
147
return self._weave_store.get_weave(file_id,
148
self._branch.get_transaction())
150
def get_file_lines(self, file_id):
151
ie = self._inventory[file_id]
152
weave = self.get_weave(file_id)
153
return weave.get_lines(ie.revision)
155
def get_file_text(self, file_id):
156
return ''.join(self.get_file_lines(file_id))
158
def get_file(self, file_id):
159
return StringIO(self.get_file_text(file_id))
161
def get_file_size(self, file_id):
162
return self._inventory[file_id].text_size
164
def get_file_sha1(self, file_id, path=None):
165
ie = self._inventory[file_id]
166
if ie.kind == "file":
170
def get_file_mtime(self, file_id, path=None):
171
ie = self._inventory[file_id]
172
revision = self._branch.get_revision(ie.revision)
173
return revision.timestamp
175
def is_executable(self, file_id, path=None):
176
ie = self._inventory[file_id]
177
if ie.kind != "file":
179
return self._inventory[file_id].executable
181
def has_filename(self, filename):
182
return bool(self.inventory.path2id(filename))
184
def list_files(self):
185
# The only files returned by this are those from the version
186
for path, entry in self.inventory.iter_entries():
187
yield path, 'V', entry.kind, entry.file_id, entry
189
def get_symlink_target(self, file_id):
190
ie = self._inventory[file_id]
191
return ie.symlink_target;
193
def kind(self, file_id):
194
return self._inventory[file_id].kind
197
self._branch.lock_read()
200
self._branch.unlock()
594
def walkdirs(self, prefix=""):
595
"""Walk the contents of this tree from path down.
597
This yields all the data about the contents of a directory at a time.
598
After each directory has been yielded, if the caller has mutated the
599
list to exclude some directories, they are then not descended into.
601
The data yielded is of the form:
602
((directory-relpath, directory-path-from-root, directory-fileid),
603
[(relpath, basename, kind, lstat, path_from_tree_root, file_id,
604
versioned_kind), ...]),
605
- directory-relpath is the containing dirs relpath from prefix
606
- directory-path-from-root is the containing dirs path from /
607
- directory-fileid is the id of the directory if it is versioned.
608
- relpath is the relative path within the subtree being walked.
609
- basename is the basename
610
- kind is the kind of the file now. If unknonwn then the file is not
611
present within the tree - but it may be recorded as versioned. See
613
- lstat is the stat data *if* the file was statted.
614
- path_from_tree_root is the path from the root of the tree.
615
- file_id is the file_id if the entry is versioned.
616
- versioned_kind is the kind of the file as last recorded in the
617
versioning system. If 'unknown' the file is not versioned.
618
One of 'kind' and 'versioned_kind' must not be 'unknown'.
620
:param prefix: Start walking from prefix within the tree rather than
621
at the root. This allows one to walk a subtree but get paths that are
622
relative to a tree rooted higher up.
623
:return: an iterator over the directory data.
625
raise NotImplementedError(self.walkdirs)
627
def supports_content_filtering(self):
630
def iter_search_rules(self, path_names, pref_names=None,
631
_default_searcher=rules._per_user_searcher):
632
"""Find the preferences for filenames in a tree.
634
:param path_names: an iterable of paths to find attributes for.
635
Paths are given relative to the root of the tree.
636
:param pref_names: the list of preferences to lookup - None for all
637
:param _default_searcher: private parameter to assist testing - don't use
638
:return: an iterator of tuple sequences, one per path-name.
639
See _RulesSearcher.get_items for details on the tuple sequence.
641
searcher = self._get_rules_searcher(_default_searcher)
642
if searcher is not None:
643
if pref_names is not None:
644
for path in path_names:
645
yield searcher.get_selected_items(path, pref_names)
647
for path in path_names:
648
yield searcher.get_items(path)
651
def _get_rules_searcher(self, default_searcher):
652
"""Get the RulesSearcher for this tree given the default one."""
653
searcher = default_searcher
203
657
class EmptyTree(Tree):
204
659
def __init__(self):
205
self._inventory = Inventory()
660
self._inventory = Inventory(root_id=None)
661
symbol_versioning.warn('EmptyTree is deprecated as of bzr 0.9 please'
662
' use repository.revision_tree instead.',
663
DeprecationWarning, stacklevel=2)
665
def get_parent_ids(self):
207
668
def get_symlink_target(self, file_id):
288
746
new_name = new_inv.id2path(file_id)
289
747
if old_name != new_name:
290
748
yield (old_name, new_name)
751
def find_ids_across_trees(filenames, trees, require_versioned=True):
752
"""Find the ids corresponding to specified filenames.
754
All matches in all trees will be used, and all children of matched
755
directories will be used.
757
:param filenames: The filenames to find file_ids for (if None, returns
759
:param trees: The trees to find file_ids within
760
:param require_versioned: if true, all specified filenames must occur in
762
:return: a set of file ids for the specified filenames and their children.
766
specified_path_ids = _find_ids_across_trees(filenames, trees,
768
return _find_children_across_trees(specified_path_ids, trees)
771
def _find_ids_across_trees(filenames, trees, require_versioned):
772
"""Find the ids corresponding to specified filenames.
774
All matches in all trees will be used, but subdirectories are not scanned.
776
:param filenames: The filenames to find file_ids for
777
:param trees: The trees to find file_ids within
778
:param require_versioned: if true, all specified filenames must occur in
780
:return: a set of file ids for the specified filenames
783
interesting_ids = set()
784
for tree_path in filenames:
787
file_id = tree.path2id(tree_path)
788
if file_id is not None:
789
interesting_ids.add(file_id)
792
not_versioned.append(tree_path)
793
if len(not_versioned) > 0 and require_versioned:
794
raise errors.PathsNotVersionedError(not_versioned)
795
return interesting_ids
798
def _find_children_across_trees(specified_ids, trees):
799
"""Return a set including specified ids and their children.
801
All matches in all trees will be used.
803
:param trees: The trees to find file_ids within
804
:return: a set containing all specified ids and their children
806
interesting_ids = set(specified_ids)
807
pending = interesting_ids
808
# now handle children of interesting ids
809
# we loop so that we handle all children of each id in both trees
810
while len(pending) > 0:
812
for file_id in pending:
814
if not tree.has_id(file_id):
816
for child_id in tree.iter_children(file_id):
817
if child_id not in interesting_ids:
818
new_pending.add(child_id)
819
interesting_ids.update(new_pending)
820
pending = new_pending
821
return interesting_ids
824
class InterTree(InterObject):
825
"""This class represents operations taking place between two Trees.
827
Its instances have methods like 'compare' and contain references to the
828
source and target trees these operations are to be carried out on.
830
Clients of bzrlib should not need to use InterTree directly, rather they
831
should use the convenience methods on Tree such as 'Tree.compare()' which
832
will pass through to InterTree as appropriate.
838
def compare(self, want_unchanged=False, specific_files=None,
839
extra_trees=None, require_versioned=False, include_root=False,
840
want_unversioned=False):
841
"""Return the changes from source to target.
843
:return: A TreeDelta.
844
:param specific_files: An optional list of file paths to restrict the
845
comparison to. When mapping filenames to ids, all matches in all
846
trees (including optional extra_trees) are used, and all children of
847
matched directories are included.
848
:param want_unchanged: An optional boolean requesting the inclusion of
849
unchanged entries in the result.
850
:param extra_trees: An optional list of additional trees to use when
851
mapping the contents of specific_files (paths) to file_ids.
852
:param require_versioned: An optional boolean (defaults to False). When
853
supplied and True all the 'specific_files' must be versioned, or
854
a PathsNotVersionedError will be thrown.
855
:param want_unversioned: Scan for unversioned paths.
857
# NB: show_status depends on being able to pass in non-versioned files
858
# and report them as unknown
859
trees = (self.source,)
860
if extra_trees is not None:
861
trees = trees + tuple(extra_trees)
862
# target is usually the newer tree:
863
specific_file_ids = self.target.paths2ids(specific_files, trees,
864
require_versioned=require_versioned)
865
if specific_files and not specific_file_ids:
866
# All files are unversioned, so just return an empty delta
867
# _compare_trees would think we want a complete delta
868
result = delta.TreeDelta()
869
fake_entry = InventoryFile('unused', 'unused', 'unused')
870
result.unversioned = [(path, None,
871
self.target._comparison_data(fake_entry, path)[0]) for path in
874
return delta._compare_trees(self.source, self.target, want_unchanged,
875
specific_files, include_root, extra_trees=extra_trees,
876
require_versioned=require_versioned,
877
want_unversioned=want_unversioned)
879
def iter_changes(self, include_unchanged=False,
880
specific_files=None, pb=None, extra_trees=[],
881
require_versioned=True, want_unversioned=False):
882
"""Generate an iterator of changes between trees.
885
(file_id, (path_in_source, path_in_target),
886
changed_content, versioned, parent, name, kind,
889
Changed_content is True if the file's content has changed. This
890
includes changes to its kind, and to a symlink's target.
892
versioned, parent, name, kind, executable are tuples of (from, to).
893
If a file is missing in a tree, its kind is None.
895
Iteration is done in parent-to-child order, relative to the target
898
There is no guarantee that all paths are in sorted order: the
899
requirement to expand the search due to renames may result in children
900
that should be found early being found late in the search, after
901
lexically later results have been returned.
902
:param require_versioned: Raise errors.PathsNotVersionedError if a
903
path in the specific_files list is not versioned in one of
904
source, target or extra_trees.
905
:param want_unversioned: Should unversioned files be returned in the
906
output. An unversioned file is defined as one with (False, False)
907
for the versioned pair.
910
lookup_trees = [self.source]
912
lookup_trees.extend(extra_trees)
913
if specific_files == []:
914
specific_file_ids = []
916
specific_file_ids = self.target.paths2ids(specific_files,
917
lookup_trees, require_versioned=require_versioned)
919
all_unversioned = sorted([(p.split('/'), p) for p in
921
if specific_files is None or
922
osutils.is_inside_any(specific_files, p)])
923
all_unversioned = deque(all_unversioned)
925
all_unversioned = deque()
927
from_entries_by_dir = list(self.source.iter_entries_by_dir(
928
specific_file_ids=specific_file_ids))
929
from_data = dict((e.file_id, (p, e)) for p, e in from_entries_by_dir)
930
to_entries_by_dir = list(self.target.iter_entries_by_dir(
931
specific_file_ids=specific_file_ids))
932
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
934
# the unversioned path lookup only occurs on real trees - where there
935
# can be extras. So the fake_entry is solely used to look up
936
# executable it values when execute is not supported.
937
fake_entry = InventoryFile('unused', 'unused', 'unused')
938
for to_path, to_entry in to_entries_by_dir:
939
while all_unversioned and all_unversioned[0][0] < to_path.split('/'):
940
unversioned_path = all_unversioned.popleft()
941
to_kind, to_executable, to_stat = \
942
self.target._comparison_data(fake_entry, unversioned_path[1])
943
yield (None, (None, unversioned_path[1]), True, (False, False),
945
(None, unversioned_path[0][-1]),
947
(None, to_executable))
948
file_id = to_entry.file_id
949
to_paths[file_id] = to_path
951
changed_content = False
952
from_path, from_entry = from_data.get(file_id, (None, None))
953
from_versioned = (from_entry is not None)
954
if from_entry is not None:
955
from_versioned = True
956
from_name = from_entry.name
957
from_parent = from_entry.parent_id
958
from_kind, from_executable, from_stat = \
959
self.source._comparison_data(from_entry, from_path)
962
from_versioned = False
966
from_executable = None
967
versioned = (from_versioned, True)
968
to_kind, to_executable, to_stat = \
969
self.target._comparison_data(to_entry, to_path)
970
kind = (from_kind, to_kind)
971
if kind[0] != kind[1]:
972
changed_content = True
973
elif from_kind == 'file':
974
from_size = self.source._file_size(from_entry, from_stat)
975
to_size = self.target._file_size(to_entry, to_stat)
976
if from_size != to_size:
977
changed_content = True
978
elif (self.source.get_file_sha1(file_id, from_path, from_stat) !=
979
self.target.get_file_sha1(file_id, to_path, to_stat)):
980
changed_content = True
981
elif from_kind == 'symlink':
982
if (self.source.get_symlink_target(file_id) !=
983
self.target.get_symlink_target(file_id)):
984
changed_content = True
985
elif from_kind == 'tree-reference':
986
if (self.source.get_reference_revision(file_id, from_path)
987
!= self.target.get_reference_revision(file_id, to_path)):
988
changed_content = True
989
parent = (from_parent, to_entry.parent_id)
990
name = (from_name, to_entry.name)
991
executable = (from_executable, to_executable)
993
pb.update('comparing files', entry_count, num_entries)
994
if (changed_content is not False or versioned[0] != versioned[1]
995
or parent[0] != parent[1] or name[0] != name[1] or
996
executable[0] != executable[1] or include_unchanged):
997
yield (file_id, (from_path, to_path), changed_content,
998
versioned, parent, name, kind, executable)
1000
while all_unversioned:
1001
# yield any trailing unversioned paths
1002
unversioned_path = all_unversioned.popleft()
1003
to_kind, to_executable, to_stat = \
1004
self.target._comparison_data(fake_entry, unversioned_path[1])
1005
yield (None, (None, unversioned_path[1]), True, (False, False),
1007
(None, unversioned_path[0][-1]),
1009
(None, to_executable))
1011
def get_to_path(to_entry):
1012
if to_entry.parent_id is None:
1013
to_path = '' # the root
1015
if to_entry.parent_id not in to_paths:
1017
return get_to_path(self.target.inventory[to_entry.parent_id])
1018
to_path = osutils.pathjoin(to_paths[to_entry.parent_id],
1020
to_paths[to_entry.file_id] = to_path
1023
for path, from_entry in from_entries_by_dir:
1024
file_id = from_entry.file_id
1025
if file_id in to_paths:
1028
if not file_id in self.target.all_file_ids():
1029
# common case - paths we have not emitted are not present in
1033
to_path = get_to_path(self.target.inventory[file_id])
1036
pb.update('comparing files', entry_count, num_entries)
1037
versioned = (True, False)
1038
parent = (from_entry.parent_id, None)
1039
name = (from_entry.name, None)
1040
from_kind, from_executable, stat_value = \
1041
self.source._comparison_data(from_entry, path)
1042
kind = (from_kind, None)
1043
executable = (from_executable, None)
1044
changed_content = from_kind is not None
1045
# the parent's path is necessarily known at this point.
1046
yield(file_id, (path, to_path), changed_content, versioned, parent,
1047
name, kind, executable)
1050
class MultiWalker(object):
1051
"""Walk multiple trees simultaneously, getting combined results."""
1053
# Note: This could be written to not assume you can do out-of-order
1054
# lookups. Instead any nodes that don't match in all trees could be
1055
# marked as 'deferred', and then returned in the final cleanup loop.
1056
# For now, I think it is "nicer" to return things as close to the
1057
# "master_tree" order as we can.
1059
def __init__(self, master_tree, other_trees):
1060
"""Create a new MultiWalker.
1062
All trees being walked must implement "iter_entries_by_dir()", such
1063
that they yield (path, object) tuples, where that object will have a
1064
'.file_id' member, that can be used to check equality.
1066
:param master_tree: All trees will be 'slaved' to the master_tree such
1067
that nodes in master_tree will be used as 'first-pass' sync points.
1068
Any nodes that aren't in master_tree will be merged in a second
1070
:param other_trees: A list of other trees to walk simultaneously.
1072
self._master_tree = master_tree
1073
self._other_trees = other_trees
1075
# Keep track of any nodes that were properly processed just out of
1076
# order, that way we don't return them at the end, we don't have to
1077
# track *all* processed file_ids, just the out-of-order ones
1078
self._out_of_order_processed = set()
1081
def _step_one(iterator):
1082
"""Step an iter_entries_by_dir iterator.
1084
:return: (has_more, path, ie)
1085
If has_more is False, path and ie will be None.
1088
path, ie = iterator.next()
1089
except StopIteration:
1090
return False, None, None
1092
return True, path, ie
1095
def _cmp_path_by_dirblock(path1, path2):
1096
"""Compare two paths based on what directory they are in.
1098
This generates a sort order, such that all children of a directory are
1099
sorted together, and grandchildren are in the same order as the
1100
children appear. But all grandchildren come after all children.
1102
:param path1: first path
1103
:param path2: the second path
1104
:return: negative number if ``path1`` comes first,
1105
0 if paths are equal
1106
and a positive number if ``path2`` sorts first
1108
# Shortcut this special case
1111
# This is stolen from _dirstate_helpers_py.py, only switching it to
1112
# Unicode objects. Consider using encode_utf8() and then using the
1113
# optimized versions, or maybe writing optimized unicode versions.
1114
if not isinstance(path1, unicode):
1115
raise TypeError("'path1' must be a unicode string, not %s: %r"
1116
% (type(path1), path1))
1117
if not isinstance(path2, unicode):
1118
raise TypeError("'path2' must be a unicode string, not %s: %r"
1119
% (type(path2), path2))
1120
return cmp(MultiWalker._path_to_key(path1),
1121
MultiWalker._path_to_key(path2))
1124
def _path_to_key(path):
1125
dirname, basename = osutils.split(path)
1126
return (dirname.split(u'/'), basename)
1128
def _lookup_by_file_id(self, extra_entries, other_tree, file_id):
1129
"""Lookup an inventory entry by file_id.
1131
This is called when an entry is missing in the normal order.
1132
Generally this is because a file was either renamed, or it was
1133
deleted/added. If the entry was found in the inventory and not in
1134
extra_entries, it will be added to self._out_of_order_processed
1136
:param extra_entries: A dictionary of {file_id: (path, ie)}. This
1137
should be filled with entries that were found before they were
1138
used. If file_id is present, it will be removed from the
1140
:param other_tree: The Tree to search, in case we didn't find the entry
1142
:param file_id: The file_id to look for
1143
:return: (path, ie) if found or (None, None) if not present.
1145
if file_id in extra_entries:
1146
return extra_entries.pop(file_id)
1147
# TODO: Is id2path better as the first call, or is
1148
# inventory[file_id] better as a first check?
1150
cur_path = other_tree.id2path(file_id)
1151
except errors.NoSuchId:
1153
if cur_path is None:
1156
self._out_of_order_processed.add(file_id)
1157
cur_ie = other_tree.inventory[file_id]
1158
return (cur_path, cur_ie)
1161
"""Match up the values in the different trees."""
1162
for result in self._walk_master_tree():
1164
self._finish_others()
1165
for result in self._walk_others():
1168
def _walk_master_tree(self):
1169
"""First pass, walk all trees in lock-step.
1171
When we are done, all nodes in the master_tree will have been
1172
processed. _other_walkers, _other_entries, and _others_extra will be
1173
set on 'self' for future processing.
1175
# This iterator has the most "inlining" done, because it tends to touch
1176
# every file in the tree, while the others only hit nodes that don't
1178
master_iterator = self._master_tree.iter_entries_by_dir()
1180
other_walkers = [other.iter_entries_by_dir()
1181
for other in self._other_trees]
1182
other_entries = [self._step_one(walker) for walker in other_walkers]
1183
# Track extra nodes in the other trees
1184
others_extra = [{} for i in xrange(len(self._other_trees))]
1186
master_has_more = True
1187
step_one = self._step_one
1188
lookup_by_file_id = self._lookup_by_file_id
1189
out_of_order_processed = self._out_of_order_processed
1191
while master_has_more:
1192
(master_has_more, path, master_ie) = step_one(master_iterator)
1193
if not master_has_more:
1196
file_id = master_ie.file_id
1198
other_values_append = other_values.append
1199
next_other_entries = []
1200
next_other_entries_append = next_other_entries.append
1201
for idx, (other_has_more, other_path, other_ie) in enumerate(other_entries):
1202
if not other_has_more:
1203
other_values_append(lookup_by_file_id(
1204
others_extra[idx], self._other_trees[idx], file_id))
1205
next_other_entries_append((False, None, None))
1206
elif file_id == other_ie.file_id:
1207
# This is the critical code path, as most of the entries
1208
# should match between most trees.
1209
other_values_append((other_path, other_ie))
1210
next_other_entries_append(step_one(other_walkers[idx]))
1212
# This walker did not match, step it until it either
1213
# matches, or we know we are past the current walker.
1214
other_walker = other_walkers[idx]
1215
other_extra = others_extra[idx]
1216
while (other_has_more and
1217
self._cmp_path_by_dirblock(other_path, path) < 0):
1218
other_file_id = other_ie.file_id
1219
if other_file_id not in out_of_order_processed:
1220
other_extra[other_file_id] = (other_path, other_ie)
1221
other_has_more, other_path, other_ie = \
1222
step_one(other_walker)
1223
if other_has_more and other_ie.file_id == file_id:
1224
# We ended up walking to this point, match and step
1226
other_values_append((other_path, other_ie))
1227
other_has_more, other_path, other_ie = \
1228
step_one(other_walker)
1230
# This record isn't in the normal order, see if it
1232
other_values_append(lookup_by_file_id(
1233
other_extra, self._other_trees[idx], file_id))
1234
next_other_entries_append((other_has_more, other_path,
1236
other_entries = next_other_entries
1238
# We've matched all the walkers, yield this datapoint
1239
yield path, file_id, master_ie, other_values
1240
self._other_walkers = other_walkers
1241
self._other_entries = other_entries
1242
self._others_extra = others_extra
1244
def _finish_others(self):
1245
"""Finish walking the other iterators, so we get all entries."""
1246
for idx, info in enumerate(self._other_entries):
1247
other_extra = self._others_extra[idx]
1248
(other_has_more, other_path, other_ie) = info
1249
while other_has_more:
1250
other_file_id = other_ie.file_id
1251
if other_file_id not in self._out_of_order_processed:
1252
other_extra[other_file_id] = (other_path, other_ie)
1253
other_has_more, other_path, other_ie = \
1254
self._step_one(self._other_walkers[idx])
1255
del self._other_entries
1257
def _walk_others(self):
1258
"""Finish up by walking all the 'deferred' nodes."""
1259
# TODO: One alternative would be to grab all possible unprocessed
1260
# file_ids, and then sort by path, and then yield them. That
1261
# might ensure better ordering, in case a caller strictly
1262
# requires parents before children.
1263
for idx, other_extra in enumerate(self._others_extra):
1264
others = sorted(other_extra.itervalues(),
1265
key=lambda x: self._path_to_key(x[0]))
1266
for other_path, other_ie in others:
1267
file_id = other_ie.file_id
1268
# We don't need to check out_of_order_processed here, because
1269
# the lookup_by_file_id will be removing anything processed
1270
# from the extras cache
1271
other_extra.pop(file_id)
1272
other_values = [(None, None) for i in xrange(idx)]
1273
other_values.append((other_path, other_ie))
1274
for alt_idx, alt_extra in enumerate(self._others_extra[idx+1:]):
1275
alt_idx = alt_idx + idx + 1
1276
alt_extra = self._others_extra[alt_idx]
1277
alt_tree = self._other_trees[alt_idx]
1278
other_values.append(self._lookup_by_file_id(
1279
alt_extra, alt_tree, file_id))
1280
yield other_path, file_id, None, other_values
added
removed
bzrlib/tree.py
