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
inventory = property(_get_inventory,
182
doc="Inventory of this Tree")
184
def _check_retrieved(self, ie, f):
187
fp = fingerprint_file(f)
190
if ie.text_size is not None:
191
if ie.text_size != fp['size']:
192
raise BzrError("mismatched size for file %r in %r" % (ie.file_id, self._store),
193
["inventory expects %d bytes" % ie.text_size,
194
"file is actually %d bytes" % fp['size'],
195
"store is probably damaged/corrupt"])
197
if ie.text_sha1 != fp['sha1']:
198
raise BzrError("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
199
["inventory expects %s" % ie.text_sha1,
200
"file is actually %s" % fp['sha1'],
201
"store is probably damaged/corrupt"])
203
def path2id(self, path):
204
"""Return the id for path in this tree."""
205
return self._inventory.path2id(path)
207
def print_file(self, file_id):
208
"""Print file with id `file_id` to stdout."""
210
sys.stdout.write(self.get_file_text(file_id))
216
"""What files are present in this tree and unknown.
218
:return: an iterator over the unknown files.
225
def filter_unversioned_files(self, paths):
226
"""Filter out paths that are not versioned.
228
:return: set of paths.
230
# NB: we specifically *don't* call self.has_filename, because for
231
# WorkingTrees that can indicate files that exist on disk but that
233
pred = self.inventory.has_filename
234
return set((p for p in paths if not pred(p)))
237
class EmptyTree(Tree):
240
self._inventory = Inventory(root_id=None)
241
symbol_versioning.warn('EmptyTree is deprecated as of bzr 0.9 please'
242
' use repository.revision_tree instead.',
243
DeprecationWarning, stacklevel=2)
245
def get_parent_ids(self):
248
def get_symlink_target(self, file_id):
251
def has_filename(self, filename):
254
def kind(self, file_id):
255
assert self._inventory[file_id].kind == "directory"
258
def list_files(self, include_root=False):
261
def __contains__(self, file_id):
262
return (file_id in self._inventory)
264
def get_file_sha1(self, file_id, path=None, stat_value=None):
268
######################################################################
271
# TODO: Merge these two functions into a single one that can operate
272
# on either a whole tree or a set of files.
274
# TODO: Return the diff in order by filename, not by category or in
275
# random order. Can probably be done by lock-stepping through the
276
# filenames from both trees.
279
def file_status(filename, old_tree, new_tree):
280
"""Return single-letter status, old and new names for a file.
282
The complexity here is in deciding how to represent renames;
283
many complex cases are possible.
285
old_inv = old_tree.inventory
286
new_inv = new_tree.inventory
287
new_id = new_inv.path2id(filename)
288
old_id = old_inv.path2id(filename)
290
if not new_id and not old_id:
291
# easy: doesn't exist in either; not versioned at all
292
if new_tree.is_ignored(filename):
293
return 'I', None, None
295
return '?', None, None
297
# There is now a file of this name, great.
300
# There is no longer a file of this name, but we can describe
301
# what happened to the file that used to have
302
# this name. There are two possibilities: either it was
303
# deleted entirely, or renamed.
305
if new_inv.has_id(old_id):
306
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
308
return 'D', old_inv.id2path(old_id), None
310
# if the file_id is new in this revision, it is added
311
if new_id and not old_inv.has_id(new_id):
314
# if there used to be a file of this name, but that ID has now
315
# disappeared, it is deleted
316
if old_id and not new_inv.has_id(old_id):
323
def find_renames(old_inv, new_inv):
324
for file_id in old_inv:
325
if file_id not in new_inv:
327
old_name = old_inv.id2path(file_id)
328
new_name = new_inv.id2path(file_id)
329
if old_name != new_name:
330
yield (old_name, new_name)
333
def find_ids_across_trees(filenames, trees, require_versioned=True):
334
"""Find the ids corresponding to specified filenames.
336
All matches in all trees will be used, and all children of matched
337
directories will be used.
339
:param filenames: The filenames to find file_ids for
340
:param trees: The trees to find file_ids within
341
:param require_versioned: if true, all specified filenames must occur in
343
:return: a set of file ids for the specified filenames and their children.
347
specified_ids = _find_filename_ids_across_trees(filenames, trees,
349
return _find_children_across_trees(specified_ids, trees)
352
def _find_filename_ids_across_trees(filenames, trees, require_versioned):
353
"""Find the ids corresponding to specified filenames.
355
All matches in all trees will be used.
357
:param filenames: The filenames to find file_ids for
358
:param trees: The trees to find file_ids within
359
:param require_versioned: if true, all specified filenames must occur in
361
:return: a set of file ids for the specified filenames
364
interesting_ids = set()
365
for tree_path in filenames:
368
file_id = tree.inventory.path2id(tree_path)
369
if file_id is not None:
370
interesting_ids.add(file_id)
373
not_versioned.append(tree_path)
374
if len(not_versioned) > 0 and require_versioned:
375
raise errors.PathsNotVersionedError(not_versioned)
376
return interesting_ids
379
def _find_children_across_trees(specified_ids, trees):
380
"""Return a set including specified ids and their children
382
All matches in all trees will be used.
384
:param trees: The trees to find file_ids within
385
:return: a set containing all specified ids and their children
387
interesting_ids = set(specified_ids)
388
pending = interesting_ids
389
# now handle children of interesting ids
390
# we loop so that we handle all children of each id in both trees
391
while len(pending) > 0:
393
for file_id in pending:
395
if file_id not in tree:
397
entry = tree.inventory[file_id]
398
for child in getattr(entry, 'children', {}).itervalues():
399
if child.file_id not in interesting_ids:
400
new_pending.add(child.file_id)
401
interesting_ids.update(new_pending)
402
pending = new_pending
403
return interesting_ids
406
class InterTree(InterObject):
407
"""This class represents operations taking place between two Trees.
409
Its instances have methods like 'compare' and contain references to the
410
source and target trees these operations are to be carried out on.
412
clients of bzrlib should not need to use InterTree directly, rather they
413
should use the convenience methods on Tree such as 'Tree.compare()' which
414
will pass through to InterTree as appropriate.
420
def compare(self, want_unchanged=False, specific_files=None,
421
extra_trees=None, require_versioned=False, include_root=False):
422
"""Return the changes from source to target.
424
:return: A TreeDelta.
425
:param specific_files: An optional list of file paths to restrict the
426
comparison to. When mapping filenames to ids, all matches in all
427
trees (including optional extra_trees) are used, and all children of
428
matched directories are included.
429
:param want_unchanged: An optional boolean requesting the inclusion of
430
unchanged entries in the result.
431
:param extra_trees: An optional list of additional trees to use when
432
mapping the contents of specific_files (paths) to file_ids.
433
:param require_versioned: An optional boolean (defaults to False). When
434
supplied and True all the 'specific_files' must be versioned, or
435
a PathsNotVersionedError will be thrown.
437
# NB: show_status depends on being able to pass in non-versioned files and
438
# report them as unknown
439
trees = (self.source, self.target)
440
if extra_trees is not None:
441
trees = trees + tuple(extra_trees)
442
specific_file_ids = find_ids_across_trees(specific_files,
443
trees, require_versioned=require_versioned)
444
if specific_files and not specific_file_ids:
445
# All files are unversioned, so just return an empty delta
446
# _compare_trees would think we want a complete delta
447
return delta.TreeDelta()
448
return delta._compare_trees(self.source, self.target, want_unchanged,
449
specific_file_ids, include_root)
451
def _iter_changes(self, from_tree, to_tree, include_unchanged,
452
specific_file_ids, pb):
453
"""Generate an iterator of changes between trees.
456
(file_id, path, changed_content, versioned, parent, name, kind,
459
Path is relative to the to_tree. changed_content is True if the file's
460
content has changed. This includes changes to its kind, and to
463
versioned, parent, name, kind, executable are tuples of (from, to).
464
If a file is missing in a tree, its kind is None.
466
Iteration is done in parent-to-child order, relative to the to_tree.
469
from_entries_by_dir = list(from_tree.inventory.iter_entries_by_dir())
470
from_data = dict((e.file_id, (p, e)) for p, e in from_entries_by_dir)
471
to_entries_by_dir = list(to_tree.inventory.iter_entries_by_dir())
472
if specific_file_ids is not None:
473
specific_file_ids = set(specific_file_ids)
474
num_entries = len(specific_file_ids)
476
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
478
for to_path, to_entry in to_entries_by_dir:
479
file_id = to_entry.file_id
480
to_paths[file_id] = to_path
481
if (specific_file_ids is not None and
482
file_id not in specific_file_ids):
485
changed_content = False
486
from_path, from_entry = from_data.get(file_id, (None, None))
487
from_versioned = (from_entry is not None)
488
if from_entry is not None:
489
from_versioned = True
490
from_name = from_entry.name
491
from_parent = from_entry.parent_id
492
from_kind, from_executable, from_stat = \
493
from_tree._comparison_data(from_entry, from_path)
494
if specific_file_ids is None:
497
from_versioned = False
501
from_executable = None
502
versioned = (from_versioned, True)
503
to_kind, to_executable, to_stat = \
504
to_tree._comparison_data(to_entry, to_path)
505
kind = (from_kind, to_kind)
506
if kind[0] != kind[1]:
507
changed_content = True
508
elif from_kind == 'file':
509
from_size = from_tree._file_size(from_entry, from_stat)
510
to_size = to_tree._file_size(to_entry, to_stat)
511
if from_size != to_size:
512
changed_content = True
513
elif (from_tree.get_file_sha1(file_id, from_path, from_stat) !=
514
to_tree.get_file_sha1(file_id, to_path, to_stat)):
515
changed_content = True
516
elif from_kind == 'symlink':
517
if (from_tree.get_symlink_target(file_id) !=
518
to_tree.get_symlink_target(file_id)):
519
changed_content = True
520
parent = (from_parent, to_entry.parent_id)
521
name = (from_name, to_entry.name)
522
executable = (from_executable, to_executable)
524
pb.update('comparing files', entry_count, num_entries)
525
if (changed_content is not False or versioned[0] != versioned[1]
526
or parent[0] != parent[1] or name[0] != name[1] or
527
executable[0] != executable[1] or include_unchanged):
528
yield (file_id, to_path, changed_content, versioned, parent,
529
name, kind, executable)
531
for path, from_entry in from_entries_by_dir:
532
file_id = from_entry.file_id
533
if file_id in to_paths:
535
if from_entry.parent_id is None:
538
to_path = osutils.pathjoin(to_paths[from_entry.parent_id],
540
to_paths[file_id] = to_path
541
if (specific_file_ids is not None and
542
file_id not in specific_file_ids):
546
pb.update('comparing files', entry_count, num_entries)
547
versioned = (True, False)
548
parent = (from_entry.parent_id, None)
549
name = (from_entry.name, None)
550
from_kind, from_executable, stat_value = \
551
from_tree._comparison_data(from_entry, path)
552
kind = (from_kind, None)
553
executable = (from_executable, None)
554
changed_content = True
555
# the parent's path is necessarily known at this point.
556
yield(file_id, to_path, changed_content, versioned, parent,
557
name, kind, executable)
560
# This was deprecated before 0.12, but did not have an official warning
561
@symbol_versioning.deprecated_function(symbol_versioning.zero_twelve)
562
def RevisionTree(*args, **kwargs):
563
"""RevisionTree has moved to bzrlib.revisiontree.RevisionTree()
565
Accessing it as bzrlib.tree.RevisionTree has been deprecated as of
568
from bzrlib.revisiontree import RevisionTree as _RevisionTree
569
return _RevisionTree(*args, **kwargs)
added
removed
bzrlib/tree.py
