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)
145
def iter_entries_by_dir(self, specific_file_ids=None):
146
"""Walk the tree in 'by_dir' order.
148
This will yield each entry in the tree as a (path, entry) tuple. The
149
order that they are yielded is: the contents of a directory are
150
preceeded by the parent of a directory, and all the contents of a
151
directory are grouped together.
153
return self.inventory.iter_entries_by_dir(
154
specific_file_ids=specific_file_ids)
156
def kind(self, file_id):
157
raise NotImplementedError("subclasses must implement kind")
159
def _comparison_data(self, entry, path):
160
"""Return a tuple of kind, executable, stat_value for a file.
162
entry may be None if there is no inventory entry for the file, but
163
path must always be supplied.
165
kind is None if there is no file present (even if an inventory id is
166
present). executable is False for non-file entries.
168
raise NotImplementedError(self._comparison_data)
170
def _file_size(self, entry, stat_value):
171
raise NotImplementedError(self._file_size)
173
def _get_inventory(self):
174
return self._inventory
176
def get_file(self, file_id):
177
"""Return a file object for the file file_id in the tree."""
178
raise NotImplementedError(self.get_file)
180
def get_file_by_path(self, path):
181
return self.get_file(self._inventory.path2id(path))
183
def annotate_iter(self, file_id):
184
"""Return an iterator of revision_id, line tuples
186
For working trees (and mutable trees in general), the special
187
revision_id 'current:' will be used for lines that are new in this
188
tree, e.g. uncommitted changes.
189
:param file_id: The file to produce an annotated version from
191
raise NotImplementedError(self.annotate_iter)
193
inventory = property(_get_inventory,
194
doc="Inventory of this Tree")
196
def _check_retrieved(self, ie, f):
199
fp = fingerprint_file(f)
202
if ie.text_size is not None:
203
if ie.text_size != fp['size']:
204
raise BzrError("mismatched size for file %r in %r" % (ie.file_id, self._store),
205
["inventory expects %d bytes" % ie.text_size,
206
"file is actually %d bytes" % fp['size'],
207
"store is probably damaged/corrupt"])
209
if ie.text_sha1 != fp['sha1']:
210
raise BzrError("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
211
["inventory expects %s" % ie.text_sha1,
212
"file is actually %s" % fp['sha1'],
213
"store is probably damaged/corrupt"])
215
def path2id(self, path):
216
"""Return the id for path in this tree."""
217
return self._inventory.path2id(path)
219
def print_file(self, file_id):
220
"""Print file with id `file_id` to stdout."""
222
sys.stdout.write(self.get_file_text(file_id))
227
def revision_tree(self, revision_id):
228
"""Obtain a revision tree for the revision revision_id.
230
The intention of this method is to allow access to possibly cached
231
tree data. Implementors of this method should raise NoSuchRevision if
232
the tree is not locally available, even if they could obtain the
233
tree via a repository or some other means. Callers are responsible
234
for finding the ultimate source for a revision tree.
236
:param revision_id: The revision_id of the requested tree.
238
:raises: NoSuchRevision if the tree cannot be obtained.
240
raise errors.NoSuchRevisionInTree(self, revision_id)
243
"""What files are present in this tree and unknown.
245
:return: an iterator over the unknown files.
252
def filter_unversioned_files(self, paths):
253
"""Filter out paths that are not versioned.
255
:return: set of paths.
257
# NB: we specifically *don't* call self.has_filename, because for
258
# WorkingTrees that can indicate files that exist on disk but that
260
pred = self.inventory.has_filename
261
return set((p for p in paths if not pred(p)))
263
def walkdirs(self, prefix=""):
264
"""Walk the contents of this tree from path down.
266
This yields all the data about the contents of a directory at a time.
267
After each directory has been yielded, if the caller has mutated the
268
list to exclude some directories, they are then not descended into.
270
The data yielded is of the form:
271
((directory-relpath, directory-path-from-root, directory-fileid),
272
[(relpath, basename, kind, lstat, path_from_tree_root, file_id,
273
versioned_kind), ...]),
274
- directory-relpath is the containing dirs relpath from prefix
275
- directory-path-from-root is the containing dirs path from /
276
- directory-fileid is the id of the directory if it is versioned.
277
- relpath is the relative path within the subtree being walked.
278
- basename is the basename
279
- kind is the kind of the file now. If unknonwn then the file is not
280
present within the tree - but it may be recorded as versioned. See
282
- lstat is the stat data *if* the file was statted.
283
- path_from_tree_root is the path from the root of the tree.
284
- file_id is the file_id is the entry is versioned.
285
- versioned_kind is the kind of the file as last recorded in the
286
versioning system. If 'unknown' the file is not versioned.
287
One of 'kind' and 'versioned_kind' must not be 'unknown'.
289
:param prefix: Start walking from prefix within the tree rather than
290
at the root. This allows one to walk a subtree but get paths that are
291
relative to a tree rooted higher up.
292
:return: an iterator over the directory data.
294
raise NotImplementedError(self.walkdirs)
297
class EmptyTree(Tree):
300
self._inventory = Inventory(root_id=None)
301
symbol_versioning.warn('EmptyTree is deprecated as of bzr 0.9 please'
302
' use repository.revision_tree instead.',
303
DeprecationWarning, stacklevel=2)
305
def get_parent_ids(self):
308
def get_symlink_target(self, file_id):
311
def has_filename(self, filename):
314
def kind(self, file_id):
315
assert self._inventory[file_id].kind == "directory"
318
def list_files(self, include_root=False):
321
def __contains__(self, file_id):
322
return (file_id in self._inventory)
324
def get_file_sha1(self, file_id, path=None, stat_value=None):
328
######################################################################
331
# TODO: Merge these two functions into a single one that can operate
332
# on either a whole tree or a set of files.
334
# TODO: Return the diff in order by filename, not by category or in
335
# random order. Can probably be done by lock-stepping through the
336
# filenames from both trees.
339
def file_status(filename, old_tree, new_tree):
340
"""Return single-letter status, old and new names for a file.
342
The complexity here is in deciding how to represent renames;
343
many complex cases are possible.
345
old_inv = old_tree.inventory
346
new_inv = new_tree.inventory
347
new_id = new_inv.path2id(filename)
348
old_id = old_inv.path2id(filename)
350
if not new_id and not old_id:
351
# easy: doesn't exist in either; not versioned at all
352
if new_tree.is_ignored(filename):
353
return 'I', None, None
355
return '?', None, None
357
# There is now a file of this name, great.
360
# There is no longer a file of this name, but we can describe
361
# what happened to the file that used to have
362
# this name. There are two possibilities: either it was
363
# deleted entirely, or renamed.
365
if new_inv.has_id(old_id):
366
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
368
return 'D', old_inv.id2path(old_id), None
370
# if the file_id is new in this revision, it is added
371
if new_id and not old_inv.has_id(new_id):
374
# if there used to be a file of this name, but that ID has now
375
# disappeared, it is deleted
376
if old_id and not new_inv.has_id(old_id):
383
def find_renames(old_inv, new_inv):
384
for file_id in old_inv:
385
if file_id not in new_inv:
387
old_name = old_inv.id2path(file_id)
388
new_name = new_inv.id2path(file_id)
389
if old_name != new_name:
390
yield (old_name, new_name)
393
def find_ids_across_trees(filenames, trees, require_versioned=True):
394
"""Find the ids corresponding to specified filenames.
396
All matches in all trees will be used, and all children of matched
397
directories will be used.
399
:param filenames: The filenames to find file_ids for
400
:param trees: The trees to find file_ids within
401
:param require_versioned: if true, all specified filenames must occur in
403
:return: a set of file ids for the specified filenames and their children.
407
specified_ids = _find_filename_ids_across_trees(filenames, trees,
409
return _find_children_across_trees(specified_ids, trees)
412
def _find_filename_ids_across_trees(filenames, trees, require_versioned):
413
"""Find the ids corresponding to specified filenames.
415
All matches in all trees will be used.
417
:param filenames: The filenames to find file_ids for
418
:param trees: The trees to find file_ids within
419
:param require_versioned: if true, all specified filenames must occur in
421
:return: a set of file ids for the specified filenames
424
interesting_ids = set()
425
for tree_path in filenames:
428
file_id = tree.inventory.path2id(tree_path)
429
if file_id is not None:
430
interesting_ids.add(file_id)
433
not_versioned.append(tree_path)
434
if len(not_versioned) > 0 and require_versioned:
435
raise errors.PathsNotVersionedError(not_versioned)
436
return interesting_ids
439
def _find_children_across_trees(specified_ids, trees):
440
"""Return a set including specified ids and their children
442
All matches in all trees will be used.
444
:param trees: The trees to find file_ids within
445
:return: a set containing all specified ids and their children
447
interesting_ids = set(specified_ids)
448
pending = interesting_ids
449
# now handle children of interesting ids
450
# we loop so that we handle all children of each id in both trees
451
while len(pending) > 0:
453
for file_id in pending:
455
if file_id not in tree:
457
entry = tree.inventory[file_id]
458
for child in getattr(entry, 'children', {}).itervalues():
459
if child.file_id not in interesting_ids:
460
new_pending.add(child.file_id)
461
interesting_ids.update(new_pending)
462
pending = new_pending
463
return interesting_ids
466
class InterTree(InterObject):
467
"""This class represents operations taking place between two Trees.
469
Its instances have methods like 'compare' and contain references to the
470
source and target trees these operations are to be carried out on.
472
clients of bzrlib should not need to use InterTree directly, rather they
473
should use the convenience methods on Tree such as 'Tree.compare()' which
474
will pass through to InterTree as appropriate.
480
def compare(self, want_unchanged=False, specific_files=None,
481
extra_trees=None, require_versioned=False, include_root=False):
482
"""Return the changes from source to target.
484
:return: A TreeDelta.
485
:param specific_files: An optional list of file paths to restrict the
486
comparison to. When mapping filenames to ids, all matches in all
487
trees (including optional extra_trees) are used, and all children of
488
matched directories are included.
489
:param want_unchanged: An optional boolean requesting the inclusion of
490
unchanged entries in the result.
491
:param extra_trees: An optional list of additional trees to use when
492
mapping the contents of specific_files (paths) to file_ids.
493
:param require_versioned: An optional boolean (defaults to False). When
494
supplied and True all the 'specific_files' must be versioned, or
495
a PathsNotVersionedError will be thrown.
497
# NB: show_status depends on being able to pass in non-versioned files and
498
# report them as unknown
499
trees = (self.source, self.target)
500
if extra_trees is not None:
501
trees = trees + tuple(extra_trees)
502
specific_file_ids = find_ids_across_trees(specific_files,
503
trees, require_versioned=require_versioned)
504
if specific_files and not specific_file_ids:
505
# All files are unversioned, so just return an empty delta
506
# _compare_trees would think we want a complete delta
507
return delta.TreeDelta()
508
return delta._compare_trees(self.source, self.target, want_unchanged,
509
specific_file_ids, include_root)
511
def _iter_changes(self, from_tree, to_tree, include_unchanged,
512
specific_file_ids, pb):
513
"""Generate an iterator of changes between trees.
516
(file_id, path, changed_content, versioned, parent, name, kind,
519
Path is relative to the to_tree. changed_content is True if the file's
520
content has changed. This includes changes to its kind, and to
523
versioned, parent, name, kind, executable are tuples of (from, to).
524
If a file is missing in a tree, its kind is None.
526
Iteration is done in parent-to-child order, relative to the to_tree.
529
from_entries_by_dir = list(from_tree.inventory.iter_entries_by_dir(
530
specific_file_ids=specific_file_ids))
531
from_data = dict((e.file_id, (p, e)) for p, e in from_entries_by_dir)
532
to_entries_by_dir = list(to_tree.inventory.iter_entries_by_dir(
533
specific_file_ids=specific_file_ids))
534
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
536
for to_path, to_entry in to_entries_by_dir:
537
file_id = to_entry.file_id
538
to_paths[file_id] = to_path
540
changed_content = False
541
from_path, from_entry = from_data.get(file_id, (None, None))
542
from_versioned = (from_entry is not None)
543
if from_entry is not None:
544
from_versioned = True
545
from_name = from_entry.name
546
from_parent = from_entry.parent_id
547
from_kind, from_executable, from_stat = \
548
from_tree._comparison_data(from_entry, from_path)
551
from_versioned = False
555
from_executable = None
556
versioned = (from_versioned, True)
557
to_kind, to_executable, to_stat = \
558
to_tree._comparison_data(to_entry, to_path)
559
kind = (from_kind, to_kind)
560
if kind[0] != kind[1]:
561
changed_content = True
562
elif from_kind == 'file':
563
from_size = from_tree._file_size(from_entry, from_stat)
564
to_size = to_tree._file_size(to_entry, to_stat)
565
if from_size != to_size:
566
changed_content = True
567
elif (from_tree.get_file_sha1(file_id, from_path, from_stat) !=
568
to_tree.get_file_sha1(file_id, to_path, to_stat)):
569
changed_content = True
570
elif from_kind == 'symlink':
571
if (from_tree.get_symlink_target(file_id) !=
572
to_tree.get_symlink_target(file_id)):
573
changed_content = True
574
parent = (from_parent, to_entry.parent_id)
575
name = (from_name, to_entry.name)
576
executable = (from_executable, to_executable)
578
pb.update('comparing files', entry_count, num_entries)
579
if (changed_content is not False or versioned[0] != versioned[1]
580
or parent[0] != parent[1] or name[0] != name[1] or
581
executable[0] != executable[1] or include_unchanged):
582
yield (file_id, to_path, changed_content, versioned, parent,
583
name, kind, executable)
585
def get_to_path(from_entry):
586
if from_entry.parent_id is None:
589
if from_entry.parent_id not in to_paths:
590
get_to_path(from_tree.inventory[from_entry.parent_id])
591
to_path = osutils.pathjoin(to_paths[from_entry.parent_id],
593
to_paths[from_entry.file_id] = to_path
596
for path, from_entry in from_entries_by_dir:
597
file_id = from_entry.file_id
598
if file_id in to_paths:
600
to_path = get_to_path(from_entry)
603
pb.update('comparing files', entry_count, num_entries)
604
versioned = (True, False)
605
parent = (from_entry.parent_id, None)
606
name = (from_entry.name, None)
607
from_kind, from_executable, stat_value = \
608
from_tree._comparison_data(from_entry, path)
609
kind = (from_kind, None)
610
executable = (from_executable, None)
611
changed_content = True
612
# the parent's path is necessarily known at this point.
613
yield(file_id, to_path, changed_content, versioned, parent,
614
name, kind, executable)
617
# This was deprecated before 0.12, but did not have an official warning
618
@symbol_versioning.deprecated_function(symbol_versioning.zero_twelve)
619
def RevisionTree(*args, **kwargs):
620
"""RevisionTree has moved to bzrlib.revisiontree.RevisionTree()
622
Accessing it as bzrlib.tree.RevisionTree has been deprecated as of
625
from bzrlib.revisiontree import RevisionTree as _RevisionTree
626
return _RevisionTree(*args, **kwargs)
added
removed
bzrlib/tree.py
