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, specific_file_ids=None):
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(
153
specific_file_ids=specific_file_ids)
155
def iter_reference_entries(self):
156
for path, entry in self.iter_entries_by_dir():
157
if entry.kind == 'tree-reference':
160
def kind(self, file_id):
161
raise NotImplementedError("subclasses must implement kind")
163
def get_reference_revision(self, entry, path=None):
164
raise NotImplementedError("subclasses must implement "
165
"get_reference_revision")
167
def _comparison_data(self, entry, path):
168
"""Return a tuple of kind, executable, stat_value for a file.
170
entry may be None if there is no inventory entry for the file, but
171
path must always be supplied.
173
kind is None if there is no file present (even if an inventory id is
174
present). executable is False for non-file entries.
176
raise NotImplementedError(self._comparison_data)
178
def _file_size(entry, stat_value):
179
raise NotImplementedError(self._file_size)
181
def _get_inventory(self):
182
return self._inventory
184
def get_file(self, file_id):
185
"""Return a file object for the file file_id in the tree."""
186
raise NotImplementedError(self.get_file)
188
def get_file_by_path(self, path):
189
return self.get_file(self._inventory.path2id(path))
191
def annotate_iter(self, file_id):
192
"""Return an iterator of revision_id, line tuples
194
For working trees (and mutable trees in general), the special
195
revision_id 'current:' will be used for lines that are new in this
196
tree, e.g. uncommitted changes.
197
:param file_id: The file to produce an annotated version from
199
raise NotImplementedError(self.annotate_iter)
201
inventory = property(_get_inventory,
202
doc="Inventory of this Tree")
204
def _check_retrieved(self, ie, f):
207
fp = fingerprint_file(f)
210
if ie.text_size is not None:
211
if ie.text_size != fp['size']:
212
raise BzrError("mismatched size for file %r in %r" % (ie.file_id, self._store),
213
["inventory expects %d bytes" % ie.text_size,
214
"file is actually %d bytes" % fp['size'],
215
"store is probably damaged/corrupt"])
217
if ie.text_sha1 != fp['sha1']:
218
raise BzrError("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
219
["inventory expects %s" % ie.text_sha1,
220
"file is actually %s" % fp['sha1'],
221
"store is probably damaged/corrupt"])
223
def path2id(self, path):
224
"""Return the id for path in this tree."""
225
return self._inventory.path2id(path)
227
def print_file(self, file_id):
228
"""Print file with id `file_id` to stdout."""
230
sys.stdout.write(self.get_file_text(file_id))
236
"""What files are present in this tree and unknown.
238
:return: an iterator over the unknown files.
245
def filter_unversioned_files(self, paths):
246
"""Filter out paths that are not versioned.
248
:return: set of paths.
250
# NB: we specifically *don't* call self.has_filename, because for
251
# WorkingTrees that can indicate files that exist on disk but that
253
pred = self.inventory.has_filename
254
return set((p for p in paths if not pred(p)))
257
class EmptyTree(Tree):
260
self._inventory = Inventory(root_id=None)
261
symbol_versioning.warn('EmptyTree is deprecated as of bzr 0.9 please'
262
' use repository.revision_tree instead.',
263
DeprecationWarning, stacklevel=2)
265
def get_parent_ids(self):
268
def get_symlink_target(self, file_id):
271
def has_filename(self, filename):
274
def kind(self, file_id):
275
assert self._inventory[file_id].kind == "directory"
278
def list_files(self, include_root=False):
281
def __contains__(self, file_id):
282
return (file_id in self._inventory)
284
def get_file_sha1(self, file_id, path=None, stat_value=None):
288
######################################################################
291
# TODO: Merge these two functions into a single one that can operate
292
# on either a whole tree or a set of files.
294
# TODO: Return the diff in order by filename, not by category or in
295
# random order. Can probably be done by lock-stepping through the
296
# filenames from both trees.
299
def file_status(filename, old_tree, new_tree):
300
"""Return single-letter status, old and new names for a file.
302
The complexity here is in deciding how to represent renames;
303
many complex cases are possible.
305
old_inv = old_tree.inventory
306
new_inv = new_tree.inventory
307
new_id = new_inv.path2id(filename)
308
old_id = old_inv.path2id(filename)
310
if not new_id and not old_id:
311
# easy: doesn't exist in either; not versioned at all
312
if new_tree.is_ignored(filename):
313
return 'I', None, None
315
return '?', None, None
317
# There is now a file of this name, great.
320
# There is no longer a file of this name, but we can describe
321
# what happened to the file that used to have
322
# this name. There are two possibilities: either it was
323
# deleted entirely, or renamed.
325
if new_inv.has_id(old_id):
326
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
328
return 'D', old_inv.id2path(old_id), None
330
# if the file_id is new in this revision, it is added
331
if new_id and not old_inv.has_id(new_id):
334
# if there used to be a file of this name, but that ID has now
335
# disappeared, it is deleted
336
if old_id and not new_inv.has_id(old_id):
343
def find_renames(old_inv, new_inv):
344
for file_id in old_inv:
345
if file_id not in new_inv:
347
old_name = old_inv.id2path(file_id)
348
new_name = new_inv.id2path(file_id)
349
if old_name != new_name:
350
yield (old_name, new_name)
353
def find_ids_across_trees(filenames, trees, require_versioned=True):
354
"""Find the ids corresponding to specified filenames.
356
All matches in all trees will be used, and all children of matched
357
directories will be used.
359
:param filenames: The filenames to find file_ids for (if None, returns
361
:param trees: The trees to find file_ids within
362
:param require_versioned: if true, all specified filenames must occur in
364
:return: a set of file ids for the specified filenames and their children.
368
specified_ids = _find_filename_ids_across_trees(filenames, trees,
370
return _find_children_across_trees(specified_ids, trees)
373
def _find_filename_ids_across_trees(filenames, trees, require_versioned):
374
"""Find the ids corresponding to specified filenames.
376
All matches in all trees will be used.
378
:param filenames: The filenames to find file_ids for
379
:param trees: The trees to find file_ids within
380
:param require_versioned: if true, all specified filenames must occur in
382
:return: a set of file ids for the specified filenames
385
interesting_ids = set()
386
for tree_path in filenames:
389
file_id = tree.inventory.path2id(tree_path)
390
if file_id is not None:
391
interesting_ids.add(file_id)
394
not_versioned.append(tree_path)
395
if len(not_versioned) > 0 and require_versioned:
396
raise errors.PathsNotVersionedError(not_versioned)
397
return interesting_ids
400
def _find_children_across_trees(specified_ids, trees):
401
"""Return a set including specified ids and their children
403
All matches in all trees will be used.
405
:param trees: The trees to find file_ids within
406
:return: a set containing all specified ids and their children
408
interesting_ids = set(specified_ids)
409
pending = interesting_ids
410
# now handle children of interesting ids
411
# we loop so that we handle all children of each id in both trees
412
while len(pending) > 0:
414
for file_id in pending:
416
if file_id not in tree:
418
entry = tree.inventory[file_id]
419
for child in getattr(entry, 'children', {}).itervalues():
420
if child.file_id not in interesting_ids:
421
new_pending.add(child.file_id)
422
interesting_ids.update(new_pending)
423
pending = new_pending
424
return interesting_ids
427
class InterTree(InterObject):
428
"""This class represents operations taking place between two Trees.
430
Its instances have methods like 'compare' and contain references to the
431
source and target trees these operations are to be carried out on.
433
clients of bzrlib should not need to use InterTree directly, rather they
434
should use the convenience methods on Tree such as 'Tree.compare()' which
435
will pass through to InterTree as appropriate.
441
def compare(self, want_unchanged=False, specific_files=None,
442
extra_trees=None, require_versioned=False, include_root=False):
443
"""Return the changes from source to target.
445
:return: A TreeDelta.
446
:param specific_files: An optional list of file paths to restrict the
447
comparison to. When mapping filenames to ids, all matches in all
448
trees (including optional extra_trees) are used, and all children of
449
matched directories are included.
450
:param want_unchanged: An optional boolean requesting the inclusion of
451
unchanged entries in the result.
452
:param extra_trees: An optional list of additional trees to use when
453
mapping the contents of specific_files (paths) to file_ids.
454
:param require_versioned: An optional boolean (defaults to False). When
455
supplied and True all the 'specific_files' must be versioned, or
456
a PathsNotVersionedError will be thrown.
458
# NB: show_status depends on being able to pass in non-versioned files and
459
# report them as unknown
460
trees = (self.source, self.target)
461
if extra_trees is not None:
462
trees = trees + tuple(extra_trees)
463
specific_file_ids = find_ids_across_trees(specific_files,
464
trees, require_versioned=require_versioned)
465
if specific_files and not specific_file_ids:
466
# All files are unversioned, so just return an empty delta
467
# _compare_trees would think we want a complete delta
468
return delta.TreeDelta()
469
return delta._compare_trees(self.source, self.target, want_unchanged,
470
specific_file_ids, include_root)
472
def _iter_changes(self, from_tree, to_tree, include_unchanged,
473
specific_file_ids, pb):
474
"""Generate an iterator of changes between trees.
477
(file_id, path, changed_content, versioned, parent, name, kind,
480
Path is relative to the to_tree. changed_content is True if the file's
481
content has changed. This includes changes to its kind, and to
484
versioned, parent, name, kind, executable are tuples of (from, to).
485
If a file is missing in a tree, its kind is None.
487
Iteration is done in parent-to-child order, relative to the to_tree.
490
from_entries_by_dir = list(from_tree.inventory.iter_entries_by_dir(
491
specific_file_ids=specific_file_ids))
492
from_data = dict((e.file_id, (p, e)) for p, e in from_entries_by_dir)
493
to_entries_by_dir = list(to_tree.inventory.iter_entries_by_dir(
494
specific_file_ids=specific_file_ids))
495
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
497
for to_path, to_entry in to_entries_by_dir:
498
file_id = to_entry.file_id
499
to_paths[file_id] = to_path
501
changed_content = False
502
from_path, from_entry = from_data.get(file_id, (None, None))
503
from_versioned = (from_entry is not None)
504
if from_entry is not None:
505
from_versioned = True
506
from_name = from_entry.name
507
from_parent = from_entry.parent_id
508
from_kind, from_executable, from_stat = \
509
from_tree._comparison_data(from_entry, from_path)
512
from_versioned = False
516
from_executable = None
517
versioned = (from_versioned, True)
518
to_kind, to_executable, to_stat = \
519
to_tree._comparison_data(to_entry, to_path)
520
kind = (from_kind, to_kind)
521
if kind[0] != kind[1]:
522
changed_content = True
523
elif from_kind == 'file':
524
from_size = from_tree._file_size(from_entry, from_stat)
525
to_size = to_tree._file_size(to_entry, to_stat)
526
if from_size != to_size:
527
changed_content = True
528
elif (from_tree.get_file_sha1(file_id, from_path, from_stat) !=
529
to_tree.get_file_sha1(file_id, to_path, to_stat)):
530
changed_content = True
531
elif from_kind == 'symlink':
532
if (from_tree.get_symlink_target(file_id) !=
533
to_tree.get_symlink_target(file_id)):
534
changed_content = True
535
elif from_kind == 'tree-reference':
536
if (from_tree.get_reference_revision(from_entry, from_path) !=
537
to_tree.get_reference_revision(to_entry, to_path)):
538
changed_content = True
539
parent = (from_parent, to_entry.parent_id)
540
name = (from_name, to_entry.name)
541
executable = (from_executable, to_executable)
543
pb.update('comparing files', entry_count, num_entries)
544
if (changed_content is not False or versioned[0] != versioned[1]
545
or parent[0] != parent[1] or name[0] != name[1] or
546
executable[0] != executable[1] or include_unchanged):
547
yield (file_id, to_path, changed_content, versioned, parent,
548
name, kind, executable)
550
def get_to_path(from_entry):
551
if from_entry.parent_id is None:
554
if from_entry.parent_id not in to_paths:
555
get_to_path(from_tree.inventory[from_entry.parent_id])
556
to_path = osutils.pathjoin(to_paths[from_entry.parent_id],
558
to_paths[from_entry.file_id] = to_path
561
for path, from_entry in from_entries_by_dir:
562
file_id = from_entry.file_id
563
if file_id in to_paths:
565
to_path = get_to_path(from_entry)
568
pb.update('comparing files', entry_count, num_entries)
569
versioned = (True, False)
570
parent = (from_entry.parent_id, None)
571
name = (from_entry.name, None)
572
from_kind, from_executable, stat_value = \
573
from_tree._comparison_data(from_entry, path)
574
kind = (from_kind, None)
575
executable = (from_executable, None)
576
changed_content = True
577
# the parent's path is necessarily known at this point.
578
yield(file_id, to_path, changed_content, versioned, parent,
579
name, kind, executable)
582
# This was deprecated before 0.12, but did not have an official warning
583
@symbol_versioning.deprecated_function(symbol_versioning.zero_twelve)
584
def RevisionTree(*args, **kwargs):
585
"""RevisionTree has moved to bzrlib.revisiontree.RevisionTree()
587
Accessing it as bzrlib.tree.RevisionTree has been deprecated as of
590
from bzrlib.revisiontree import RevisionTree as _RevisionTree
591
return _RevisionTree(*args, **kwargs)
added
removed
bzrlib/tree.py
