1
# Copyright (C) 2005 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., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
17
"""Tree classes, representing directory at point in time.
21
from cStringIO import StringIO
29
from bzrlib.decorators import needs_read_lock
30
from bzrlib.errors import BzrError, BzrCheckError
31
from bzrlib import errors
32
from bzrlib.inventory import Inventory
33
from bzrlib.inter import InterObject
34
from bzrlib.osutils import fingerprint_file
35
import bzrlib.revision
36
from bzrlib.trace import mutter, note
40
"""Abstract file tree.
42
There are several subclasses:
44
* `WorkingTree` exists as files on disk editable by the user.
46
* `RevisionTree` is a tree as recorded at some point in the past.
48
Trees contain an `Inventory` object, and also know how to retrieve
49
file texts mentioned in the inventory, either from a working
50
directory or from a store.
52
It is possible for trees to contain files that are not described
53
in their inventory or vice versa; for this use `filenames()`.
55
Trees can be compared, etc, regardless of whether they are working
56
trees or versioned trees.
59
def changes_from(self, other, want_unchanged=False, specific_files=None,
60
extra_trees=None, require_versioned=False, include_root=False):
61
"""Return a TreeDelta of the changes from other to this tree.
63
:param other: A tree to compare with.
64
:param specific_files: An optional list of file paths to restrict the
65
comparison to. When mapping filenames to ids, all matches in all
66
trees (including optional extra_trees) are used, and all children of
67
matched directories are included.
68
:param want_unchanged: An optional boolean requesting the inclusion of
69
unchanged entries in the result.
70
:param extra_trees: An optional list of additional trees to use when
71
mapping the contents of specific_files (paths) to file_ids.
72
:param require_versioned: An optional boolean (defaults to False). When
73
supplied and True all the 'specific_files' must be versioned, or
74
a PathsNotVersionedError will be thrown.
76
The comparison will be performed by an InterTree object looked up on
79
# Martin observes that Tree.changes_from returns a TreeDelta and this
80
# may confuse people, because the class name of the returned object is
81
# a synonym of the object referenced in the method name.
82
return InterTree.get(other, self).compare(
83
want_unchanged=want_unchanged,
84
specific_files=specific_files,
85
extra_trees=extra_trees,
86
require_versioned=require_versioned,
87
include_root=include_root
90
def _iter_changes(self, from_tree, include_unchanged=False,
91
specific_file_ids=None, pb=None):
92
intertree = InterTree.get(from_tree, self)
93
return intertree._iter_changes(from_tree, self, include_unchanged,
94
specific_file_ids, pb)
97
"""Get a list of the conflicts in the tree.
99
Each conflict is an instance of bzrlib.conflicts.Conflict.
103
def get_parent_ids(self):
104
"""Get the parent ids for this tree.
106
:return: a list of parent ids. [] is returned to indicate
107
a tree with no parents.
108
:raises: BzrError if the parents are not known.
110
raise NotImplementedError(self.get_parent_ids)
112
def has_filename(self, filename):
113
"""True if the tree has given filename."""
114
raise NotImplementedError()
116
def has_id(self, file_id):
117
return self.inventory.has_id(file_id)
119
__contains__ = has_id
121
def has_or_had_id(self, file_id):
122
if file_id == self.inventory.root.file_id:
124
return self.inventory.has_id(file_id)
127
return iter(self.inventory)
129
def id2path(self, file_id):
130
return self.inventory.id2path(file_id)
132
def is_control_filename(self, filename):
133
"""True if filename is the name of a control file in this tree.
135
:param filename: A filename within the tree. This is a relative path
136
from the root of this tree.
138
This is true IF and ONLY IF the filename is part of the meta data
139
that bzr controls in this tree. I.E. a random .bzr directory placed
140
on disk will not be a control file for this tree.
142
return self.bzrdir.is_control_filename(filename)
144
def iter_entries_by_dir(self):
145
"""Walk the tree in 'by_dir' order.
147
This will yield each entry in the tree as a (path, entry) tuple. The
148
order that they are yielded is: the contents of a directory are
149
preceeded by the parent of a directory, and all the contents of a
150
directory are grouped together.
152
return self.inventory.iter_entries_by_dir()
154
def kind(self, file_id):
155
raise NotImplementedError("subclasses must implement kind")
157
def _comparison_data(self, entry, path):
158
"""Return a tuple of kind, executable, stat_value for a file.
160
entry may be None if there is no inventory entry for the file, but
161
path must always be supplied.
163
kind is None if there is no file present (even if an inventory id is
164
present). executable is False for non-file entries.
166
raise NotImplementedError(self._comparison_data)
168
def _file_size(entry, stat_value):
169
raise NotImplementedError(self._file_size)
171
def _get_inventory(self):
172
return self._inventory
174
def get_file(self, file_id):
175
"""Return a file object for the file file_id in the tree."""
176
raise NotImplementedError(self.get_file)
178
def get_file_by_path(self, path):
179
return self.get_file(self._inventory.path2id(path))
181
def annotate_iter(self, file_id):
182
"""Return an iterator of revision_id, line tuples
184
For working trees (and mutable trees in general), the special
185
revision_id 'current:' will be used for lines that are new in this
186
tree, e.g. uncommitted changes.
187
:param file_id: The file to produce an annotated version from
189
raise NotImplementedError(self.annotate_iter)
191
inventory = property(_get_inventory,
192
doc="Inventory of this Tree")
194
def _check_retrieved(self, ie, f):
197
fp = fingerprint_file(f)
200
if ie.text_size is not None:
201
if ie.text_size != fp['size']:
202
raise BzrError("mismatched size for file %r in %r" % (ie.file_id, self._store),
203
["inventory expects %d bytes" % ie.text_size,
204
"file is actually %d bytes" % fp['size'],
205
"store is probably damaged/corrupt"])
207
if ie.text_sha1 != fp['sha1']:
208
raise BzrError("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
209
["inventory expects %s" % ie.text_sha1,
210
"file is actually %s" % fp['sha1'],
211
"store is probably damaged/corrupt"])
213
def path2id(self, path):
214
"""Return the id for path in this tree."""
215
return self._inventory.path2id(path)
217
def print_file(self, file_id):
218
"""Print file with id `file_id` to stdout."""
220
sys.stdout.write(self.get_file_text(file_id))
226
"""What files are present in this tree and unknown.
228
:return: an iterator over the unknown files.
235
def filter_unversioned_files(self, paths):
236
"""Filter out paths that are not versioned.
238
:return: set of paths.
240
# NB: we specifically *don't* call self.has_filename, because for
241
# WorkingTrees that can indicate files that exist on disk but that
243
pred = self.inventory.has_filename
244
return set((p for p in paths if not pred(p)))
247
class EmptyTree(Tree):
250
self._inventory = Inventory(root_id=None)
251
symbol_versioning.warn('EmptyTree is deprecated as of bzr 0.9 please'
252
' use repository.revision_tree instead.',
253
DeprecationWarning, stacklevel=2)
255
def get_parent_ids(self):
258
def get_symlink_target(self, file_id):
261
def has_filename(self, filename):
264
def kind(self, file_id):
265
assert self._inventory[file_id].kind == "directory"
268
def list_files(self, include_root=False):
271
def __contains__(self, file_id):
272
return (file_id in self._inventory)
274
def get_file_sha1(self, file_id, path=None, stat_value=None):
278
######################################################################
281
# TODO: Merge these two functions into a single one that can operate
282
# on either a whole tree or a set of files.
284
# TODO: Return the diff in order by filename, not by category or in
285
# random order. Can probably be done by lock-stepping through the
286
# filenames from both trees.
289
def file_status(filename, old_tree, new_tree):
290
"""Return single-letter status, old and new names for a file.
292
The complexity here is in deciding how to represent renames;
293
many complex cases are possible.
295
old_inv = old_tree.inventory
296
new_inv = new_tree.inventory
297
new_id = new_inv.path2id(filename)
298
old_id = old_inv.path2id(filename)
300
if not new_id and not old_id:
301
# easy: doesn't exist in either; not versioned at all
302
if new_tree.is_ignored(filename):
303
return 'I', None, None
305
return '?', None, None
307
# There is now a file of this name, great.
310
# There is no longer a file of this name, but we can describe
311
# what happened to the file that used to have
312
# this name. There are two possibilities: either it was
313
# deleted entirely, or renamed.
315
if new_inv.has_id(old_id):
316
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
318
return 'D', old_inv.id2path(old_id), None
320
# if the file_id is new in this revision, it is added
321
if new_id and not old_inv.has_id(new_id):
324
# if there used to be a file of this name, but that ID has now
325
# disappeared, it is deleted
326
if old_id and not new_inv.has_id(old_id):
333
def find_renames(old_inv, new_inv):
334
for file_id in old_inv:
335
if file_id not in new_inv:
337
old_name = old_inv.id2path(file_id)
338
new_name = new_inv.id2path(file_id)
339
if old_name != new_name:
340
yield (old_name, new_name)
343
def find_ids_across_trees(filenames, trees, require_versioned=True):
344
"""Find the ids corresponding to specified filenames.
346
All matches in all trees will be used, and all children of matched
347
directories will be used.
349
:param filenames: The filenames to find file_ids for
350
:param trees: The trees to find file_ids within
351
:param require_versioned: if true, all specified filenames must occur in
353
:return: a set of file ids for the specified filenames and their children.
357
specified_ids = _find_filename_ids_across_trees(filenames, trees,
359
return _find_children_across_trees(specified_ids, trees)
362
def _find_filename_ids_across_trees(filenames, trees, require_versioned):
363
"""Find the ids corresponding to specified filenames.
365
All matches in all trees will be used.
367
:param filenames: The filenames to find file_ids for
368
:param trees: The trees to find file_ids within
369
:param require_versioned: if true, all specified filenames must occur in
371
:return: a set of file ids for the specified filenames
374
interesting_ids = set()
375
for tree_path in filenames:
378
file_id = tree.inventory.path2id(tree_path)
379
if file_id is not None:
380
interesting_ids.add(file_id)
383
not_versioned.append(tree_path)
384
if len(not_versioned) > 0 and require_versioned:
385
raise errors.PathsNotVersionedError(not_versioned)
386
return interesting_ids
389
def _find_children_across_trees(specified_ids, trees):
390
"""Return a set including specified ids and their children
392
All matches in all trees will be used.
394
:param trees: The trees to find file_ids within
395
:return: a set containing all specified ids and their children
397
interesting_ids = set(specified_ids)
398
pending = interesting_ids
399
# now handle children of interesting ids
400
# we loop so that we handle all children of each id in both trees
401
while len(pending) > 0:
403
for file_id in pending:
405
if file_id not in tree:
407
entry = tree.inventory[file_id]
408
for child in getattr(entry, 'children', {}).itervalues():
409
if child.file_id not in interesting_ids:
410
new_pending.add(child.file_id)
411
interesting_ids.update(new_pending)
412
pending = new_pending
413
return interesting_ids
416
class InterTree(InterObject):
417
"""This class represents operations taking place between two Trees.
419
Its instances have methods like 'compare' and contain references to the
420
source and target trees these operations are to be carried out on.
422
clients of bzrlib should not need to use InterTree directly, rather they
423
should use the convenience methods on Tree such as 'Tree.compare()' which
424
will pass through to InterTree as appropriate.
430
def compare(self, want_unchanged=False, specific_files=None,
431
extra_trees=None, require_versioned=False, include_root=False):
432
"""Return the changes from source to target.
434
:return: A TreeDelta.
435
:param specific_files: An optional list of file paths to restrict the
436
comparison to. When mapping filenames to ids, all matches in all
437
trees (including optional extra_trees) are used, and all children of
438
matched directories are included.
439
:param want_unchanged: An optional boolean requesting the inclusion of
440
unchanged entries in the result.
441
:param extra_trees: An optional list of additional trees to use when
442
mapping the contents of specific_files (paths) to file_ids.
443
:param require_versioned: An optional boolean (defaults to False). When
444
supplied and True all the 'specific_files' must be versioned, or
445
a PathsNotVersionedError will be thrown.
447
# NB: show_status depends on being able to pass in non-versioned files and
448
# report them as unknown
449
trees = (self.source, self.target)
450
if extra_trees is not None:
451
trees = trees + tuple(extra_trees)
452
specific_file_ids = find_ids_across_trees(specific_files,
453
trees, require_versioned=require_versioned)
454
if specific_files and not specific_file_ids:
455
# All files are unversioned, so just return an empty delta
456
# _compare_trees would think we want a complete delta
457
return delta.TreeDelta()
458
return delta._compare_trees(self.source, self.target, want_unchanged,
459
specific_file_ids, include_root)
461
def _iter_changes(self, from_tree, to_tree, include_unchanged,
462
specific_file_ids, pb):
463
"""Generate an iterator of changes between trees.
466
(file_id, path, changed_content, versioned, parent, name, kind,
469
Path is relative to the to_tree. changed_content is True if the file's
470
content has changed. This includes changes to its kind, and to
473
versioned, parent, name, kind, executable are tuples of (from, to).
474
If a file is missing in a tree, its kind is None.
476
Iteration is done in parent-to-child order, relative to the to_tree.
479
from_entries_by_dir = list(from_tree.inventory.iter_entries_by_dir())
480
from_data = dict((e.file_id, (p, e)) for p, e in from_entries_by_dir)
481
to_entries_by_dir = list(to_tree.inventory.iter_entries_by_dir())
482
if specific_file_ids is not None:
483
specific_file_ids = set(specific_file_ids)
484
num_entries = len(specific_file_ids)
486
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
488
for to_path, to_entry in to_entries_by_dir:
489
file_id = to_entry.file_id
490
to_paths[file_id] = to_path
491
if (specific_file_ids is not None and
492
file_id not in specific_file_ids):
495
changed_content = False
496
from_path, from_entry = from_data.get(file_id, (None, None))
497
from_versioned = (from_entry is not None)
498
if from_entry is not None:
499
from_versioned = True
500
from_name = from_entry.name
501
from_parent = from_entry.parent_id
502
from_kind, from_executable, from_stat = \
503
from_tree._comparison_data(from_entry, from_path)
504
if specific_file_ids is None:
507
from_versioned = False
511
from_executable = None
512
versioned = (from_versioned, True)
513
to_kind, to_executable, to_stat = \
514
to_tree._comparison_data(to_entry, to_path)
515
kind = (from_kind, to_kind)
516
if kind[0] != kind[1]:
517
changed_content = True
518
elif from_kind == 'file':
519
from_size = from_tree._file_size(from_entry, from_stat)
520
to_size = to_tree._file_size(to_entry, to_stat)
521
if from_size != to_size:
522
changed_content = True
523
elif (from_tree.get_file_sha1(file_id, from_path, from_stat) !=
524
to_tree.get_file_sha1(file_id, to_path, to_stat)):
525
changed_content = True
526
elif from_kind == 'symlink':
527
if (from_tree.get_symlink_target(file_id) !=
528
to_tree.get_symlink_target(file_id)):
529
changed_content = True
530
parent = (from_parent, to_entry.parent_id)
531
name = (from_name, to_entry.name)
532
executable = (from_executable, to_executable)
534
pb.update('comparing files', entry_count, num_entries)
535
if (changed_content is not False or versioned[0] != versioned[1]
536
or parent[0] != parent[1] or name[0] != name[1] or
537
executable[0] != executable[1] or include_unchanged):
538
yield (file_id, to_path, changed_content, versioned, parent,
539
name, kind, executable)
541
for path, from_entry in from_entries_by_dir:
542
file_id = from_entry.file_id
543
if file_id in to_paths:
545
if from_entry.parent_id is None:
548
to_path = osutils.pathjoin(to_paths[from_entry.parent_id],
550
to_paths[file_id] = to_path
551
if (specific_file_ids is not None and
552
file_id not in specific_file_ids):
556
pb.update('comparing files', entry_count, num_entries)
557
versioned = (True, False)
558
parent = (from_entry.parent_id, None)
559
name = (from_entry.name, None)
560
from_kind, from_executable, stat_value = \
561
from_tree._comparison_data(from_entry, path)
562
kind = (from_kind, None)
563
executable = (from_executable, None)
564
changed_content = True
565
# the parent's path is necessarily known at this point.
566
yield(file_id, to_path, changed_content, versioned, parent,
567
name, kind, executable)
570
# This was deprecated before 0.12, but did not have an official warning
571
@symbol_versioning.deprecated_function(symbol_versioning.zero_twelve)
572
def RevisionTree(*args, **kwargs):
573
"""RevisionTree has moved to bzrlib.revisiontree.RevisionTree()
575
Accessing it as bzrlib.tree.RevisionTree has been deprecated as of
578
from bzrlib.revisiontree import RevisionTree as _RevisionTree
579
return _RevisionTree(*args, **kwargs)
added
removed
bzrlib/tree.py
