1
# Copyright (C) 2005-2011 Canonical Ltd
1
# Copyright (C) 2005 Canonical Ltd
3
3
# This program is free software; you can redistribute it and/or modify
4
4
# it under the terms of the GNU General Public License as published by
5
5
# the Free Software Foundation; either version 2 of the License, or
6
6
# (at your option) any later version.
8
8
# This program is distributed in the hope that it will be useful,
9
9
# but WITHOUT ANY WARRANTY; without even the implied warranty of
10
10
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11
11
# GNU General Public License for more details.
13
13
# You should have received a copy of the GNU General Public License
14
14
# along with this program; if not, write to the Free Software
15
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
15
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
17
17
"""Tree classes, representing directory at point in time.
22
from bzrlib.lazy_import import lazy_import
23
lazy_import(globals(), """
27
conflicts as _mod_conflicts,
34
revision as _mod_revision,
40
from bzrlib.decorators import needs_read_lock
41
from bzrlib.inter import InterObject
42
from bzrlib.symbol_versioning import (
21
from cStringIO import StringIO
24
from bzrlib.trace import mutter, note
25
from bzrlib.errors import BzrError, BzrCheckError
26
from bzrlib.inventory import Inventory
27
from bzrlib.osutils import appendpath, fingerprint_file
48
32
class Tree(object):
49
33
"""Abstract file tree.
51
35
There are several subclasses:
53
37
* `WorkingTree` exists as files on disk editable by the user.
55
39
* `RevisionTree` is a tree as recorded at some point in the past.
43
Trees contain an `Inventory` object, and also know how to retrieve
44
file texts mentioned in the inventory, either from a working
45
directory or from a store.
47
It is possible for trees to contain files that are not described
48
in their inventory or vice versa; for this use `filenames()`.
57
50
Trees can be compared, etc, regardless of whether they are working
58
51
trees or versioned trees.
61
def changes_from(self, other, want_unchanged=False, specific_files=None,
62
extra_trees=None, require_versioned=False, include_root=False,
63
want_unversioned=False):
64
"""Return a TreeDelta of the changes from other to this tree.
66
:param other: A tree to compare with.
67
:param specific_files: An optional list of file paths to restrict the
68
comparison to. When mapping filenames to ids, all matches in all
69
trees (including optional extra_trees) are used, and all children of
70
matched directories are included.
71
:param want_unchanged: An optional boolean requesting the inclusion of
72
unchanged entries in the result.
73
:param extra_trees: An optional list of additional trees to use when
74
mapping the contents of specific_files (paths) to file_ids.
75
:param require_versioned: An optional boolean (defaults to False). When
76
supplied and True all the 'specific_files' must be versioned, or
77
a PathsNotVersionedError will be thrown.
78
:param want_unversioned: Scan for unversioned paths.
80
The comparison will be performed by an InterTree object looked up on
83
# Martin observes that Tree.changes_from returns a TreeDelta and this
84
# may confuse people, because the class name of the returned object is
85
# a synonym of the object referenced in the method name.
86
return InterTree.get(other, self).compare(
87
want_unchanged=want_unchanged,
88
specific_files=specific_files,
89
extra_trees=extra_trees,
90
require_versioned=require_versioned,
91
include_root=include_root,
92
want_unversioned=want_unversioned,
95
def iter_changes(self, from_tree, include_unchanged=False,
96
specific_files=None, pb=None, extra_trees=None,
97
require_versioned=True, want_unversioned=False):
98
"""See InterTree.iter_changes"""
99
intertree = InterTree.get(from_tree, self)
100
return intertree.iter_changes(include_unchanged, specific_files, pb,
101
extra_trees, require_versioned, want_unversioned=want_unversioned)
104
"""Get a list of the conflicts in the tree.
106
Each conflict is an instance of bzrlib.conflicts.Conflict.
108
return _mod_conflicts.ConflictList()
111
"""For trees that can have unversioned files, return all such paths."""
114
def get_parent_ids(self):
115
"""Get the parent ids for this tree.
117
:return: a list of parent ids. [] is returned to indicate
118
a tree with no parents.
119
:raises: BzrError if the parents are not known.
121
raise NotImplementedError(self.get_parent_ids)
123
54
def has_filename(self, filename):
124
55
"""True if the tree has given filename."""
125
raise NotImplementedError(self.has_filename)
56
raise NotImplementedError()
127
58
def has_id(self, file_id):
128
raise NotImplementedError(self.has_id)
130
@deprecated_method(deprecated_in((2, 4, 0)))
131
def __contains__(self, file_id):
132
return self.has_id(file_id)
59
return self.inventory.has_id(file_id)
134
61
def has_or_had_id(self, file_id):
135
raise NotImplementedError(self.has_or_had_id)
137
def is_ignored(self, filename):
138
"""Check whether the filename is ignored by this tree.
140
:param filename: The relative filename within the tree.
141
:return: True if the filename is ignored.
145
def all_file_ids(self):
146
"""Iterate through all file ids, including ids for missing files."""
147
raise NotImplementedError(self.all_file_ids)
62
if file_id == self.inventory.root.file_id:
64
return self.inventory.has_id(file_id)
69
return iter(self.inventory)
149
71
def id2path(self, file_id):
150
"""Return the path for a file id.
154
raise NotImplementedError(self.id2path)
156
def iter_entries_by_dir(self, specific_file_ids=None, yield_parents=False):
157
"""Walk the tree in 'by_dir' order.
159
This will yield each entry in the tree as a (path, entry) tuple.
160
The order that they are yielded is:
162
Directories are walked in a depth-first lexicographical order,
163
however, whenever a directory is reached, all of its direct child
164
nodes are yielded in lexicographical order before yielding the
167
For example, in the tree::
177
The yield order (ignoring root) would be::
179
a, f, a/b, a/d, a/b/c, a/d/e, f/g
181
:param yield_parents: If True, yield the parents from the root leading
182
down to specific_file_ids that have been requested. This has no
183
impact if specific_file_ids is None.
185
raise NotImplementedError(self.iter_entries_by_dir)
187
def list_files(self, include_root=False, from_dir=None, recursive=True):
188
"""List all files in this tree.
190
:param include_root: Whether to include the entry for the tree root
191
:param from_dir: Directory under which to list files
192
:param recursive: Whether to list files recursively
193
:return: iterator over tuples of (path, versioned, kind, file_id,
196
raise NotImplementedError(self.list_files)
198
def iter_references(self):
199
if self.supports_tree_reference():
200
for path, entry in self.iter_entries_by_dir():
201
if entry.kind == 'tree-reference':
202
yield path, entry.file_id
204
def kind(self, file_id):
205
raise NotImplementedError("Tree subclass %s must implement kind"
206
% self.__class__.__name__)
208
def stored_kind(self, file_id):
209
"""File kind stored for this file_id.
211
May not match kind on disk for working trees. Always available
212
for versioned files, even when the file itself is missing.
214
return self.kind(file_id)
216
def path_content_summary(self, path):
217
"""Get a summary of the information about path.
219
All the attributes returned are for the canonical form, not the
220
convenient form (if content filters are in use.)
222
:param path: A relative path within the tree.
223
:return: A tuple containing kind, size, exec, sha1-or-link.
224
Kind is always present (see tree.kind()).
225
size is present if kind is file and the size of the
226
canonical form can be cheaply determined, None otherwise.
227
exec is None unless kind is file and the platform supports the 'x'
229
sha1-or-link is the link target if kind is symlink, or the sha1 if
230
it can be obtained without reading the file.
232
raise NotImplementedError(self.path_content_summary)
234
def get_reference_revision(self, file_id, path=None):
235
raise NotImplementedError("Tree subclass %s must implement "
236
"get_reference_revision"
237
% self.__class__.__name__)
239
def _comparison_data(self, entry, path):
240
"""Return a tuple of kind, executable, stat_value for a file.
242
entry may be None if there is no inventory entry for the file, but
243
path must always be supplied.
245
kind is None if there is no file present (even if an inventory id is
246
present). executable is False for non-file entries.
248
raise NotImplementedError(self._comparison_data)
250
def _file_size(self, entry, stat_value):
251
raise NotImplementedError(self._file_size)
253
def get_file(self, file_id, path=None):
254
"""Return a file object for the file file_id in the tree.
256
If both file_id and path are defined, it is implementation defined as
257
to which one is used.
259
raise NotImplementedError(self.get_file)
261
def get_file_with_stat(self, file_id, path=None):
262
"""Get a file handle and stat object for file_id.
264
The default implementation returns (self.get_file, None) for backwards
267
:param file_id: The file id to read.
268
:param path: The path of the file, if it is known.
269
:return: A tuple (file_handle, stat_value_or_None). If the tree has
270
no stat facility, or need for a stat cache feedback during commit,
271
it may return None for the second element of the tuple.
273
return (self.get_file(file_id, path), None)
275
def get_file_text(self, file_id, path=None):
276
"""Return the byte content of a file.
278
:param file_id: The file_id of the file.
279
:param path: The path of the file.
281
If both file_id and path are supplied, an implementation may use
284
:returns: A single byte string for the whole file.
286
my_file = self.get_file(file_id, path)
288
return my_file.read()
292
def get_file_lines(self, file_id, path=None):
293
"""Return the content of a file, as lines.
295
:param file_id: The file_id of the file.
296
:param path: The path of the file.
298
If both file_id and path are supplied, an implementation may use
301
return osutils.split_lines(self.get_file_text(file_id, path))
303
def get_file_verifier(self, file_id, path=None, stat_value=None):
304
"""Return a verifier for a file.
306
The default implementation returns a sha1.
308
:param file_id: The handle for this file.
309
:param path: The path that this file can be found at.
310
These must point to the same object.
311
:param stat_value: Optional stat value for the object
312
:return: Tuple with verifier name and verifier data
314
return ("SHA1", self.get_file_sha1(file_id, path=path,
315
stat_value=stat_value))
317
def get_file_sha1(self, file_id, path=None, stat_value=None):
318
"""Return the SHA1 file for a file.
320
:note: callers should use get_file_verifier instead
321
where possible, as the underlying repository implementation may
322
have quicker access to a non-sha1 verifier.
324
:param file_id: The handle for this file.
325
:param path: The path that this file can be found at.
326
These must point to the same object.
327
:param stat_value: Optional stat value for the object
329
raise NotImplementedError(self.get_file_sha1)
331
def get_file_mtime(self, file_id, path=None):
332
"""Return the modification time for a file.
334
:param file_id: The handle for this file.
335
:param path: The path that this file can be found at.
336
These must point to the same object.
338
raise NotImplementedError(self.get_file_mtime)
340
def get_file_size(self, file_id):
341
"""Return the size of a file in bytes.
343
This applies only to regular files. If invoked on directories or
344
symlinks, it will return None.
345
:param file_id: The file-id of the file
347
raise NotImplementedError(self.get_file_size)
72
return self.inventory.id2path(file_id)
74
def _get_inventory(self):
75
return self._inventory
349
77
def get_file_by_path(self, path):
350
raise NotImplementedError(self.get_file_by_path)
352
def is_executable(self, file_id, path=None):
353
"""Check if a file is executable.
355
:param file_id: The handle for this file.
356
:param path: The path that this file can be found at.
357
These must point to the same object.
359
raise NotImplementedError(self.is_executable)
361
def iter_files_bytes(self, desired_files):
362
"""Iterate through file contents.
364
Files will not necessarily be returned in the order they occur in
365
desired_files. No specific order is guaranteed.
367
Yields pairs of identifier, bytes_iterator. identifier is an opaque
368
value supplied by the caller as part of desired_files. It should
369
uniquely identify the file version in the caller's context. (Examples:
370
an index number or a TreeTransform trans_id.)
372
bytes_iterator is an iterable of bytestrings for the file. The
373
kind of iterable and length of the bytestrings are unspecified, but for
374
this implementation, it is a tuple containing a single bytestring with
375
the complete text of the file.
377
:param desired_files: a list of (file_id, identifier) pairs
379
for file_id, identifier in desired_files:
380
# We wrap the string in a tuple so that we can return an iterable
381
# of bytestrings. (Technically, a bytestring is also an iterable
382
# of bytestrings, but iterating through each character is not
384
cur_file = (self.get_file_text(file_id),)
385
yield identifier, cur_file
387
def get_symlink_target(self, file_id, path=None):
388
"""Get the target for a given file_id.
390
It is assumed that the caller already knows that file_id is referencing
392
:param file_id: Handle for the symlink entry.
393
:param path: The path of the file.
394
If both file_id and path are supplied, an implementation may use
396
:return: The path the symlink points to.
398
raise NotImplementedError(self.get_symlink_target)
400
def get_root_id(self):
401
"""Return the file_id for the root of this tree."""
402
raise NotImplementedError(self.get_root_id)
404
def annotate_iter(self, file_id,
405
default_revision=_mod_revision.CURRENT_REVISION):
406
"""Return an iterator of revision_id, line tuples.
408
For working trees (and mutable trees in general), the special
409
revision_id 'current:' will be used for lines that are new in this
410
tree, e.g. uncommitted changes.
411
:param file_id: The file to produce an annotated version from
412
:param default_revision: For lines that don't match a basis, mark them
413
with this revision id. Not all implementations will make use of
416
raise NotImplementedError(self.annotate_iter)
418
def _get_plan_merge_data(self, file_id, other, base):
419
from bzrlib import versionedfile
420
vf = versionedfile._PlanMergeVersionedFile(file_id)
421
last_revision_a = self._get_file_revision(file_id, vf, 'this:')
422
last_revision_b = other._get_file_revision(file_id, vf, 'other:')
424
last_revision_base = None
426
last_revision_base = base._get_file_revision(file_id, vf, 'base:')
427
return vf, last_revision_a, last_revision_b, last_revision_base
429
def plan_file_merge(self, file_id, other, base=None):
430
"""Generate a merge plan based on annotations.
432
If the file contains uncommitted changes in this tree, they will be
433
attributed to the 'current:' pseudo-revision. If the file contains
434
uncommitted changes in the other tree, they will be assigned to the
435
'other:' pseudo-revision.
437
data = self._get_plan_merge_data(file_id, other, base)
438
vf, last_revision_a, last_revision_b, last_revision_base = data
439
return vf.plan_merge(last_revision_a, last_revision_b,
442
def plan_file_lca_merge(self, file_id, other, base=None):
443
"""Generate a merge plan based lca-newness.
445
If the file contains uncommitted changes in this tree, they will be
446
attributed to the 'current:' pseudo-revision. If the file contains
447
uncommitted changes in the other tree, they will be assigned to the
448
'other:' pseudo-revision.
450
data = self._get_plan_merge_data(file_id, other, base)
451
vf, last_revision_a, last_revision_b, last_revision_base = data
452
return vf.plan_lca_merge(last_revision_a, last_revision_b,
455
def _iter_parent_trees(self):
456
"""Iterate through parent trees, defaulting to Tree.revision_tree."""
457
for revision_id in self.get_parent_ids():
459
yield self.revision_tree(revision_id)
460
except errors.NoSuchRevisionInTree:
461
yield self.repository.revision_tree(revision_id)
463
def _get_file_revision(self, file_id, vf, tree_revision):
464
"""Ensure that file_id, tree_revision is in vf to plan the merge."""
466
if getattr(self, '_repository', None) is None:
467
last_revision = tree_revision
468
parent_keys = [(file_id, t.get_file_revision(file_id)) for t in
469
self._iter_parent_trees()]
470
vf.add_lines((file_id, last_revision), parent_keys,
471
self.get_file_lines(file_id))
472
repo = self.branch.repository
475
last_revision = self.get_file_revision(file_id)
476
base_vf = self._repository.texts
477
if base_vf not in vf.fallback_versionedfiles:
478
vf.fallback_versionedfiles.append(base_vf)
78
return self.get_file(self._inventory.path2id(path))
80
inventory = property(_get_inventory,
81
doc="Inventory of this Tree")
481
83
def _check_retrieved(self, ie, f):
484
fp = osutils.fingerprint_file(f)
86
fp = fingerprint_file(f)
487
if ie.text_size is not None:
89
if ie.text_size != None:
488
90
if ie.text_size != fp['size']:
489
raise errors.BzrError(
490
"mismatched size for file %r in %r" %
491
(ie.file_id, self._store),
91
raise BzrError("mismatched size for file %r in %r" % (ie.file_id, self._store),
492
92
["inventory expects %d bytes" % ie.text_size,
493
93
"file is actually %d bytes" % fp['size'],
494
94
"store is probably damaged/corrupt"])
496
96
if ie.text_sha1 != fp['sha1']:
497
raise errors.BzrError("wrong SHA-1 for file %r in %r" %
498
(ie.file_id, self._store),
97
raise BzrError("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
499
98
["inventory expects %s" % ie.text_sha1,
500
99
"file is actually %s" % fp['sha1'],
501
100
"store is probably damaged/corrupt"])
503
def path2id(self, path):
504
"""Return the id for path in this tree."""
505
raise NotImplementedError(self.path2id)
507
def paths2ids(self, paths, trees=[], require_versioned=True):
508
"""Return all the ids that can be reached by walking from paths.
510
Each path is looked up in this tree and any extras provided in
511
trees, and this is repeated recursively: the children in an extra tree
512
of a directory that has been renamed under a provided path in this tree
513
are all returned, even if none exist under a provided path in this
514
tree, and vice versa.
516
:param paths: An iterable of paths to start converting to ids from.
517
Alternatively, if paths is None, no ids should be calculated and None
518
will be returned. This is offered to make calling the api unconditional
519
for code that *might* take a list of files.
520
:param trees: Additional trees to consider.
521
:param require_versioned: If False, do not raise NotVersionedError if
522
an element of paths is not versioned in this tree and all of trees.
524
return find_ids_across_trees(paths, [self] + list(trees), require_versioned)
526
def iter_children(self, file_id):
527
entry = self.iter_entries_by_dir([file_id]).next()[1]
528
for child in getattr(entry, 'children', {}).itervalues():
532
"""Lock this tree for multiple read only operations.
534
:return: A bzrlib.lock.LogicalLockResult.
538
def revision_tree(self, revision_id):
539
"""Obtain a revision tree for the revision revision_id.
541
The intention of this method is to allow access to possibly cached
542
tree data. Implementors of this method should raise NoSuchRevision if
543
the tree is not locally available, even if they could obtain the
544
tree via a repository or some other means. Callers are responsible
545
for finding the ultimate source for a revision tree.
547
:param revision_id: The revision_id of the requested tree.
549
:raises: NoSuchRevision if the tree cannot be obtained.
551
raise errors.NoSuchRevisionInTree(self, revision_id)
554
"""What files are present in this tree and unknown.
556
:return: an iterator over the unknown files.
103
def print_file(self, file_id):
104
"""Print file with id `file_id` to stdout."""
106
sys.stdout.write(self.get_file_text(file_id))
109
def export(self, dest, format='dir', root=None):
110
"""Export this tree."""
112
exporter = exporters[format]
114
from bzrlib.errors import BzrCommandError
115
raise BzrCommandError("export format %r not supported" % format)
116
exporter(self, dest, root)
120
class RevisionTree(Tree):
121
"""Tree viewing a previous revision.
123
File text can be retrieved from the text store.
125
TODO: Some kind of `__repr__` method, but a good one
126
probably means knowing the branch and revision number,
127
or at least passing a description to the constructor.
130
def __init__(self, weave_store, inv, revision_id):
131
self._weave_store = weave_store
132
self._inventory = inv
133
self._revision_id = revision_id
135
def get_weave(self, file_id):
136
# FIXME: RevisionTree should be given a branch
137
# not a store, or the store should know the branch.
138
import bzrlib.transactions as transactions
139
return self._weave_store.get_weave(file_id,
140
transactions.PassThroughTransaction())
143
def get_file_lines(self, file_id):
144
ie = self._inventory[file_id]
145
weave = self.get_weave(file_id)
146
return weave.get(ie.revision)
149
def get_file_text(self, file_id):
150
return ''.join(self.get_file_lines(file_id))
153
def get_file(self, file_id):
154
return StringIO(self.get_file_text(file_id))
156
def get_file_size(self, file_id):
157
return self._inventory[file_id].text_size
159
def get_file_sha1(self, file_id):
160
ie = self._inventory[file_id]
161
if ie.kind == "file":
164
def is_executable(self, file_id):
165
ie = self._inventory[file_id]
166
if ie.kind != "file":
168
return self._inventory[file_id].executable
170
def has_filename(self, filename):
171
return bool(self.inventory.path2id(filename))
173
def list_files(self):
174
# The only files returned by this are those from the version
175
for path, entry in self.inventory.iter_entries():
176
yield path, 'V', entry.kind, entry.file_id, entry
178
def get_symlink_target(self, file_id):
179
ie = self._inventory[file_id]
180
return ie.symlink_target;
182
def kind(self, file_id):
183
return self._inventory[file_id].kind
185
class EmptyTree(Tree):
187
self._inventory = Inventory()
189
def get_symlink_target(self, file_id):
192
def has_filename(self, filename):
195
def list_files(self):
198
def __contains__(self, file_id):
199
return file_id in self._inventory
201
def get_file_sha1(self, file_id):
202
assert self._inventory[file_id].kind == "root_directory"
206
######################################################################
209
# TODO: Merge these two functions into a single one that can operate
210
# on either a whole tree or a set of files.
212
# TODO: Return the diff in order by filename, not by category or in
213
# random order. Can probably be done by lock-stepping through the
214
# filenames from both trees.
217
def file_status(filename, old_tree, new_tree):
218
"""Return single-letter status, old and new names for a file.
220
The complexity here is in deciding how to represent renames;
221
many complex cases are possible.
223
old_inv = old_tree.inventory
224
new_inv = new_tree.inventory
225
new_id = new_inv.path2id(filename)
226
old_id = old_inv.path2id(filename)
228
if not new_id and not old_id:
229
# easy: doesn't exist in either; not versioned at all
230
if new_tree.is_ignored(filename):
231
return 'I', None, None
233
return '?', None, None
235
# There is now a file of this name, great.
563
def filter_unversioned_files(self, paths):
564
"""Filter out paths that are versioned.
566
:return: set of paths.
568
raise NotImplementedError(self.filter_unversioned_files)
570
def walkdirs(self, prefix=""):
571
"""Walk the contents of this tree from path down.
573
This yields all the data about the contents of a directory at a time.
574
After each directory has been yielded, if the caller has mutated the
575
list to exclude some directories, they are then not descended into.
577
The data yielded is of the form:
578
((directory-relpath, directory-path-from-root, directory-fileid),
579
[(relpath, basename, kind, lstat, path_from_tree_root, file_id,
580
versioned_kind), ...]),
581
- directory-relpath is the containing dirs relpath from prefix
582
- directory-path-from-root is the containing dirs path from /
583
- directory-fileid is the id of the directory if it is versioned.
584
- relpath is the relative path within the subtree being walked.
585
- basename is the basename
586
- kind is the kind of the file now. If unknonwn then the file is not
587
present within the tree - but it may be recorded as versioned. See
589
- lstat is the stat data *if* the file was statted.
590
- path_from_tree_root is the path from the root of the tree.
591
- file_id is the file_id if the entry is versioned.
592
- versioned_kind is the kind of the file as last recorded in the
593
versioning system. If 'unknown' the file is not versioned.
594
One of 'kind' and 'versioned_kind' must not be 'unknown'.
596
:param prefix: Start walking from prefix within the tree rather than
597
at the root. This allows one to walk a subtree but get paths that are
598
relative to a tree rooted higher up.
599
:return: an iterator over the directory data.
601
raise NotImplementedError(self.walkdirs)
603
def supports_content_filtering(self):
606
def _content_filter_stack(self, path=None, file_id=None):
607
"""The stack of content filters for a path if filtering is supported.
609
Readers will be applied in first-to-last order.
610
Writers will be applied in last-to-first order.
611
Either the path or the file-id needs to be provided.
613
:param path: path relative to the root of the tree
615
:param file_id: file_id or None if unknown
616
:return: the list of filters - [] if there are none
618
filter_pref_names = filters._get_registered_names()
619
if len(filter_pref_names) == 0:
622
path = self.id2path(file_id)
623
prefs = self.iter_search_rules([path], filter_pref_names).next()
624
stk = filters._get_filter_stack_for(prefs)
625
if 'filters' in debug.debug_flags:
626
trace.note("*** %s content-filter: %s => %r" % (path,prefs,stk))
629
def _content_filter_stack_provider(self):
630
"""A function that returns a stack of ContentFilters.
632
The function takes a path (relative to the top of the tree) and a
633
file-id as parameters.
635
:return: None if content filtering is not supported by this tree.
637
if self.supports_content_filtering():
638
return lambda path, file_id: \
639
self._content_filter_stack(path, file_id)
643
def iter_search_rules(self, path_names, pref_names=None,
644
_default_searcher=None):
645
"""Find the preferences for filenames in a tree.
647
:param path_names: an iterable of paths to find attributes for.
648
Paths are given relative to the root of the tree.
649
:param pref_names: the list of preferences to lookup - None for all
650
:param _default_searcher: private parameter to assist testing - don't use
651
:return: an iterator of tuple sequences, one per path-name.
652
See _RulesSearcher.get_items for details on the tuple sequence.
654
if _default_searcher is None:
655
_default_searcher = rules._per_user_searcher
656
searcher = self._get_rules_searcher(_default_searcher)
657
if searcher is not None:
658
if pref_names is not None:
659
for path in path_names:
660
yield searcher.get_selected_items(path, pref_names)
662
for path in path_names:
663
yield searcher.get_items(path)
665
def _get_rules_searcher(self, default_searcher):
666
"""Get the RulesSearcher for this tree given the default one."""
667
searcher = default_searcher
671
class InventoryTree(Tree):
672
"""A tree that relies on an inventory for its metadata.
674
Trees contain an `Inventory` object, and also know how to retrieve
675
file texts mentioned in the inventory, either from a working
676
directory or from a store.
678
It is possible for trees to contain files that are not described
679
in their inventory or vice versa; for this use `filenames()`.
681
Subclasses should set the _inventory attribute, which is considered
682
private to external API users.
685
def get_canonical_inventory_paths(self, paths):
686
"""Like get_canonical_inventory_path() but works on multiple items.
688
:param paths: A sequence of paths relative to the root of the tree.
689
:return: A list of paths, with each item the corresponding input path
690
adjusted to account for existing elements that match case
693
return list(self._yield_canonical_inventory_paths(paths))
695
def get_canonical_inventory_path(self, path):
696
"""Returns the first inventory item that case-insensitively matches path.
698
If a path matches exactly, it is returned. If no path matches exactly
699
but more than one path matches case-insensitively, it is implementation
700
defined which is returned.
702
If no path matches case-insensitively, the input path is returned, but
703
with as many path entries that do exist changed to their canonical
706
If you need to resolve many names from the same tree, you should
707
use get_canonical_inventory_paths() to avoid O(N) behaviour.
709
:param path: A paths relative to the root of the tree.
710
:return: The input path adjusted to account for existing elements
711
that match case insensitively.
713
return self._yield_canonical_inventory_paths([path]).next()
715
def _yield_canonical_inventory_paths(self, paths):
717
# First, if the path as specified exists exactly, just use it.
718
if self.path2id(path) is not None:
722
cur_id = self.get_root_id()
724
bit_iter = iter(path.split("/"))
728
for child in self.iter_children(cur_id):
730
# XXX: it seem like if the child is known to be in the
731
# tree, we shouldn't need to go from its id back to
732
# its path -- mbp 2010-02-11
734
# XXX: it seems like we could be more efficient
735
# by just directly looking up the original name and
736
# only then searching all children; also by not
737
# chopping paths so much. -- mbp 2010-02-11
738
child_base = os.path.basename(self.id2path(child))
739
if (child_base == elt):
740
# if we found an exact match, we can stop now; if
741
# we found an approximate match we need to keep
742
# searching because there might be an exact match
745
new_path = osutils.pathjoin(cur_path, child_base)
747
elif child_base.lower() == lelt:
749
new_path = osutils.pathjoin(cur_path, child_base)
750
except errors.NoSuchId:
751
# before a change is committed we can see this error...
756
# got to the end of this directory and no entries matched.
757
# Return what matched so far, plus the rest as specified.
758
cur_path = osutils.pathjoin(cur_path, elt, *list(bit_iter))
763
def _get_inventory(self):
764
return self._inventory
766
inventory = property(_get_inventory,
767
doc="Inventory of this Tree")
770
def path2id(self, path):
771
"""Return the id for path in this tree."""
772
return self._inventory.path2id(path)
774
def id2path(self, file_id):
775
"""Return the path for a file id.
779
return self.inventory.id2path(file_id)
781
def has_id(self, file_id):
782
return self.inventory.has_id(file_id)
784
def has_or_had_id(self, file_id):
785
return self.inventory.has_id(file_id)
787
def all_file_ids(self):
788
return set(self.inventory)
790
@deprecated_method(deprecated_in((2, 4, 0)))
792
return iter(self.inventory)
794
def filter_unversioned_files(self, paths):
795
"""Filter out paths that are versioned.
797
:return: set of paths.
799
# NB: we specifically *don't* call self.has_filename, because for
800
# WorkingTrees that can indicate files that exist on disk but that
802
pred = self.inventory.has_filename
803
return set((p for p in paths if not pred(p)))
806
def iter_entries_by_dir(self, specific_file_ids=None, yield_parents=False):
807
"""Walk the tree in 'by_dir' order.
809
This will yield each entry in the tree as a (path, entry) tuple.
810
The order that they are yielded is:
812
See Tree.iter_entries_by_dir for details.
814
:param yield_parents: If True, yield the parents from the root leading
815
down to specific_file_ids that have been requested. This has no
816
impact if specific_file_ids is None.
818
return self.inventory.iter_entries_by_dir(
819
specific_file_ids=specific_file_ids, yield_parents=yield_parents)
821
def get_file_by_path(self, path):
822
return self.get_file(self._inventory.path2id(path), path)
825
def find_ids_across_trees(filenames, trees, require_versioned=True):
826
"""Find the ids corresponding to specified filenames.
828
All matches in all trees will be used, and all children of matched
829
directories will be used.
831
:param filenames: The filenames to find file_ids for (if None, returns
833
:param trees: The trees to find file_ids within
834
:param require_versioned: if true, all specified filenames must occur in
836
:return: a set of file ids for the specified filenames and their children.
840
specified_path_ids = _find_ids_across_trees(filenames, trees,
842
return _find_children_across_trees(specified_path_ids, trees)
845
def _find_ids_across_trees(filenames, trees, require_versioned):
846
"""Find the ids corresponding to specified filenames.
848
All matches in all trees will be used, but subdirectories are not scanned.
850
:param filenames: The filenames to find file_ids for
851
:param trees: The trees to find file_ids within
852
:param require_versioned: if true, all specified filenames must occur in
854
:return: a set of file ids for the specified filenames
857
interesting_ids = set()
858
for tree_path in filenames:
861
file_id = tree.path2id(tree_path)
862
if file_id is not None:
863
interesting_ids.add(file_id)
866
not_versioned.append(tree_path)
867
if len(not_versioned) > 0 and require_versioned:
868
raise errors.PathsNotVersionedError(not_versioned)
869
return interesting_ids
872
def _find_children_across_trees(specified_ids, trees):
873
"""Return a set including specified ids and their children.
875
All matches in all trees will be used.
877
:param trees: The trees to find file_ids within
878
:return: a set containing all specified ids and their children
880
interesting_ids = set(specified_ids)
881
pending = interesting_ids
882
# now handle children of interesting ids
883
# we loop so that we handle all children of each id in both trees
884
while len(pending) > 0:
886
for file_id in pending:
888
if not tree.has_or_had_id(file_id):
890
for child_id in tree.iter_children(file_id):
891
if child_id not in interesting_ids:
892
new_pending.add(child_id)
893
interesting_ids.update(new_pending)
894
pending = new_pending
895
return interesting_ids
898
class InterTree(InterObject):
899
"""This class represents operations taking place between two Trees.
901
Its instances have methods like 'compare' and contain references to the
902
source and target trees these operations are to be carried out on.
904
Clients of bzrlib should not need to use InterTree directly, rather they
905
should use the convenience methods on Tree such as 'Tree.compare()' which
906
will pass through to InterTree as appropriate.
909
# Formats that will be used to test this InterTree. If both are
910
# None, this InterTree will not be tested (e.g. because a complex
912
_matching_from_tree_format = None
913
_matching_to_tree_format = None
918
def is_compatible(kls, source, target):
919
# The default implementation is naive and uses the public API, so
920
# it works for all trees.
923
def _changes_from_entries(self, source_entry, target_entry,
924
source_path=None, target_path=None):
925
"""Generate a iter_changes tuple between source_entry and target_entry.
927
:param source_entry: An inventory entry from self.source, or None.
928
:param target_entry: An inventory entry from self.target, or None.
929
:param source_path: The path of source_entry, if known. If not known
930
it will be looked up.
931
:param target_path: The path of target_entry, if known. If not known
932
it will be looked up.
933
:return: A tuple, item 0 of which is an iter_changes result tuple, and
934
item 1 is True if there are any changes in the result tuple.
936
if source_entry is None:
937
if target_entry is None:
939
file_id = target_entry.file_id
941
file_id = source_entry.file_id
942
if source_entry is not None:
943
source_versioned = True
944
source_name = source_entry.name
945
source_parent = source_entry.parent_id
946
if source_path is None:
947
source_path = self.source.id2path(file_id)
948
source_kind, source_executable, source_stat = \
949
self.source._comparison_data(source_entry, source_path)
951
source_versioned = False
955
source_executable = None
956
if target_entry is not None:
957
target_versioned = True
958
target_name = target_entry.name
959
target_parent = target_entry.parent_id
960
if target_path is None:
961
target_path = self.target.id2path(file_id)
962
target_kind, target_executable, target_stat = \
963
self.target._comparison_data(target_entry, target_path)
965
target_versioned = False
969
target_executable = None
970
versioned = (source_versioned, target_versioned)
971
kind = (source_kind, target_kind)
972
changed_content = False
973
if source_kind != target_kind:
974
changed_content = True
975
elif source_kind == 'file':
976
if not self.file_content_matches(file_id, file_id, source_path,
977
target_path, source_stat, target_stat):
978
changed_content = True
979
elif source_kind == 'symlink':
980
if (self.source.get_symlink_target(file_id) !=
981
self.target.get_symlink_target(file_id)):
982
changed_content = True
983
# XXX: Yes, the indentation below is wrong. But fixing it broke
984
# test_merge.TestMergerEntriesLCAOnDisk.
985
# test_nested_tree_subtree_renamed_and_modified. We'll wait for
986
# the fix from bzr.dev -- vila 2009026
987
elif source_kind == 'tree-reference':
988
if (self.source.get_reference_revision(file_id, source_path)
989
!= self.target.get_reference_revision(file_id, target_path)):
990
changed_content = True
991
parent = (source_parent, target_parent)
992
name = (source_name, target_name)
993
executable = (source_executable, target_executable)
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]):
1000
return (file_id, (source_path, target_path), changed_content,
1001
versioned, parent, name, kind, executable), changes
1004
def compare(self, want_unchanged=False, specific_files=None,
1005
extra_trees=None, require_versioned=False, include_root=False,
1006
want_unversioned=False):
1007
"""Return the changes from source to target.
1009
:return: A TreeDelta.
1010
:param specific_files: An optional list of file paths to restrict the
1011
comparison to. When mapping filenames to ids, all matches in all
1012
trees (including optional extra_trees) are used, and all children of
1013
matched directories are included.
1014
:param want_unchanged: An optional boolean requesting the inclusion of
1015
unchanged entries in the result.
1016
:param extra_trees: An optional list of additional trees to use when
1017
mapping the contents of specific_files (paths) to file_ids.
1018
:param require_versioned: An optional boolean (defaults to False). When
1019
supplied and True all the 'specific_files' must be versioned, or
1020
a PathsNotVersionedError will be thrown.
1021
:param want_unversioned: Scan for unversioned paths.
1023
trees = (self.source,)
1024
if extra_trees is not None:
1025
trees = trees + tuple(extra_trees)
1026
# target is usually the newer tree:
1027
specific_file_ids = self.target.paths2ids(specific_files, trees,
1028
require_versioned=require_versioned)
1029
if specific_files and not specific_file_ids:
1030
# All files are unversioned, so just return an empty delta
1031
# _compare_trees would think we want a complete delta
1032
result = delta.TreeDelta()
1033
fake_entry = inventory.InventoryFile('unused', 'unused', 'unused')
1034
result.unversioned = [(path, None,
1035
self.target._comparison_data(fake_entry, path)[0]) for path in
1038
return delta._compare_trees(self.source, self.target, want_unchanged,
1039
specific_files, include_root, extra_trees=extra_trees,
1040
require_versioned=require_versioned,
1041
want_unversioned=want_unversioned)
1043
def iter_changes(self, include_unchanged=False,
1044
specific_files=None, pb=None, extra_trees=[],
1045
require_versioned=True, want_unversioned=False):
1046
"""Generate an iterator of changes between trees.
1048
A tuple is returned:
1049
(file_id, (path_in_source, path_in_target),
1050
changed_content, versioned, parent, name, kind,
1053
Changed_content is True if the file's content has changed. This
1054
includes changes to its kind, and to a symlink's target.
1056
versioned, parent, name, kind, executable are tuples of (from, to).
1057
If a file is missing in a tree, its kind is None.
1059
Iteration is done in parent-to-child order, relative to the target
1062
There is no guarantee that all paths are in sorted order: the
1063
requirement to expand the search due to renames may result in children
1064
that should be found early being found late in the search, after
1065
lexically later results have been returned.
1066
:param require_versioned: Raise errors.PathsNotVersionedError if a
1067
path in the specific_files list is not versioned in one of
1068
source, target or extra_trees.
1069
:param specific_files: An optional list of file paths to restrict the
1070
comparison to. When mapping filenames to ids, all matches in all
1071
trees (including optional extra_trees) are used, and all children
1072
of matched directories are included. The parents in the target tree
1073
of the specific files up to and including the root of the tree are
1074
always evaluated for changes too.
1075
:param want_unversioned: Should unversioned files be returned in the
1076
output. An unversioned file is defined as one with (False, False)
1077
for the versioned pair.
1079
lookup_trees = [self.source]
1081
lookup_trees.extend(extra_trees)
1082
# The ids of items we need to examine to insure delta consistency.
1083
precise_file_ids = set()
1084
changed_file_ids = []
1085
if specific_files == []:
1086
specific_file_ids = []
1088
specific_file_ids = self.target.paths2ids(specific_files,
1089
lookup_trees, require_versioned=require_versioned)
1090
if specific_files is not None:
1091
# reparented or added entries must have their parents included
1092
# so that valid deltas can be created. The seen_parents set
1093
# tracks the parents that we need to have.
1094
# The seen_dirs set tracks directory entries we've yielded.
1095
# After outputting version object in to_entries we set difference
1096
# the two seen sets and start checking parents.
1097
seen_parents = set()
1099
if want_unversioned:
1100
all_unversioned = sorted([(p.split('/'), p) for p in
1101
self.target.extras()
1102
if specific_files is None or
1103
osutils.is_inside_any(specific_files, p)])
1104
all_unversioned = collections.deque(all_unversioned)
1106
all_unversioned = collections.deque()
1108
from_entries_by_dir = list(self.source.iter_entries_by_dir(
1109
specific_file_ids=specific_file_ids))
1110
from_data = dict((e.file_id, (p, e)) for p, e in from_entries_by_dir)
1111
to_entries_by_dir = list(self.target.iter_entries_by_dir(
1112
specific_file_ids=specific_file_ids))
1113
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
1115
# the unversioned path lookup only occurs on real trees - where there
1116
# can be extras. So the fake_entry is solely used to look up
1117
# executable it values when execute is not supported.
1118
fake_entry = inventory.InventoryFile('unused', 'unused', 'unused')
1119
for target_path, target_entry in to_entries_by_dir:
1120
while (all_unversioned and
1121
all_unversioned[0][0] < target_path.split('/')):
1122
unversioned_path = all_unversioned.popleft()
1123
target_kind, target_executable, target_stat = \
1124
self.target._comparison_data(fake_entry, unversioned_path[1])
1125
yield (None, (None, unversioned_path[1]), True, (False, False),
1127
(None, unversioned_path[0][-1]),
1128
(None, target_kind),
1129
(None, target_executable))
1130
source_path, source_entry = from_data.get(target_entry.file_id,
1132
result, changes = self._changes_from_entries(source_entry,
1133
target_entry, source_path=source_path, target_path=target_path)
1134
to_paths[result[0]] = result[1][1]
1139
pb.update('comparing files', entry_count, num_entries)
1140
if changes or include_unchanged:
1141
if specific_file_ids is not None:
1142
new_parent_id = result[4][1]
1143
precise_file_ids.add(new_parent_id)
1144
changed_file_ids.append(result[0])
1146
# Ensure correct behaviour for reparented/added specific files.
1147
if specific_files is not None:
1148
# Record output dirs
1149
if result[6][1] == 'directory':
1150
seen_dirs.add(result[0])
1151
# Record parents of reparented/added entries.
1152
versioned = result[3]
1154
if not versioned[0] or parents[0] != parents[1]:
1155
seen_parents.add(parents[1])
1156
while all_unversioned:
1157
# yield any trailing unversioned paths
1158
unversioned_path = all_unversioned.popleft()
1159
to_kind, to_executable, to_stat = \
1160
self.target._comparison_data(fake_entry, unversioned_path[1])
1161
yield (None, (None, unversioned_path[1]), True, (False, False),
1163
(None, unversioned_path[0][-1]),
1165
(None, to_executable))
1166
# Yield all remaining source paths
1167
for path, from_entry in from_entries_by_dir:
1168
file_id = from_entry.file_id
1169
if file_id in to_paths:
1172
if not self.target.has_id(file_id):
1173
# common case - paths we have not emitted are not present in
1177
to_path = self.target.id2path(file_id)
1180
pb.update('comparing files', entry_count, num_entries)
1181
versioned = (True, False)
1182
parent = (from_entry.parent_id, None)
1183
name = (from_entry.name, None)
1184
from_kind, from_executable, stat_value = \
1185
self.source._comparison_data(from_entry, path)
1186
kind = (from_kind, None)
1187
executable = (from_executable, None)
1188
changed_content = from_kind is not None
1189
# the parent's path is necessarily known at this point.
1190
changed_file_ids.append(file_id)
1191
yield(file_id, (path, to_path), changed_content, versioned, parent,
1192
name, kind, executable)
1193
changed_file_ids = set(changed_file_ids)
1194
if specific_file_ids is not None:
1195
for result in self._handle_precise_ids(precise_file_ids,
1199
def _get_entry(self, tree, file_id):
1200
"""Get an inventory entry from a tree, with missing entries as None.
1202
If the tree raises NotImplementedError on accessing .inventory, then
1203
this is worked around using iter_entries_by_dir on just the file id
1206
:param tree: The tree to lookup the entry in.
1207
:param file_id: The file_id to lookup.
1210
inventory = tree.inventory
1211
except NotImplementedError:
1212
# No inventory available.
1214
iterator = tree.iter_entries_by_dir(specific_file_ids=[file_id])
1215
return iterator.next()[1]
1216
except StopIteration:
1220
return inventory[file_id]
1221
except errors.NoSuchId:
1224
def _handle_precise_ids(self, precise_file_ids, changed_file_ids,
1225
discarded_changes=None):
1226
"""Fill out a partial iter_changes to be consistent.
1228
:param precise_file_ids: The file ids of parents that were seen during
1230
:param changed_file_ids: The file ids of already emitted items.
1231
:param discarded_changes: An optional dict of precalculated
1232
iter_changes items which the partial iter_changes had not output
1234
:return: A generator of iter_changes items to output.
1236
# process parents of things that had changed under the users
1237
# requested paths to prevent incorrect paths or parent ids which
1238
# aren't in the tree.
1239
while precise_file_ids:
1240
precise_file_ids.discard(None)
1241
# Don't emit file_ids twice
1242
precise_file_ids.difference_update(changed_file_ids)
1243
if not precise_file_ids:
1245
# If the there was something at a given output path in source, we
1246
# have to include the entry from source in the delta, or we would
1247
# be putting this entry into a used path.
1249
for parent_id in precise_file_ids:
1251
paths.append(self.target.id2path(parent_id))
1252
except errors.NoSuchId:
1253
# This id has been dragged in from the source by delta
1254
# expansion and isn't present in target at all: we don't
1255
# need to check for path collisions on it.
1258
old_id = self.source.path2id(path)
1259
precise_file_ids.add(old_id)
1260
precise_file_ids.discard(None)
1261
current_ids = precise_file_ids
1262
precise_file_ids = set()
1263
# We have to emit all of precise_file_ids that have been altered.
1264
# We may have to output the children of some of those ids if any
1265
# directories have stopped being directories.
1266
for file_id in current_ids:
1268
if discarded_changes:
1269
result = discarded_changes.get(file_id)
1274
old_entry = self._get_entry(self.source, file_id)
1275
new_entry = self._get_entry(self.target, file_id)
1276
result, changes = self._changes_from_entries(
1277
old_entry, new_entry)
1280
# Get this parents parent to examine.
1281
new_parent_id = result[4][1]
1282
precise_file_ids.add(new_parent_id)
1284
if (result[6][0] == 'directory' and
1285
result[6][1] != 'directory'):
1286
# This stopped being a directory, the old children have
1288
if old_entry is None:
1289
# Reusing a discarded change.
1290
old_entry = self._get_entry(self.source, file_id)
1291
for child in old_entry.children.values():
1292
precise_file_ids.add(child.file_id)
1293
changed_file_ids.add(result[0])
1297
def file_content_matches(self, source_file_id, target_file_id,
1298
source_path=None, target_path=None, source_stat=None, target_stat=None):
1299
"""Check if two files are the same in the source and target trees.
1301
This only checks that the contents of the files are the same,
1302
it does not touch anything else.
1304
:param source_file_id: File id of the file in the source tree
1305
:param target_file_id: File id of the file in the target tree
1306
:param source_path: Path of the file in the source tree
1307
:param target_path: Path of the file in the target tree
1308
:param source_stat: Optional stat value of the file in the source tree
1309
:param target_stat: Optional stat value of the file in the target tree
1310
:return: Boolean indicating whether the files have the same contents
1312
source_verifier_kind, source_verifier_data = self.source.get_file_verifier(
1313
source_file_id, source_path, source_stat)
1314
target_verifier_kind, target_verifier_data = self.target.get_file_verifier(
1315
target_file_id, target_path, target_stat)
1316
if source_verifier_kind == target_verifier_kind:
1317
return (source_verifier_data == target_verifier_data)
1318
# Fall back to SHA1 for now
1319
if source_verifier_kind != "SHA1":
1320
source_sha1 = self.source.get_file_sha1(source_file_id,
1321
source_path, source_stat)
1323
source_sha1 = source_verifier_data
1324
if target_verifier_kind != "SHA1":
1325
target_sha1 = self.target.get_file_sha1(target_file_id,
1326
target_path, target_stat)
1328
target_sha1 = target_verifier_data
1329
return (source_sha1 == target_sha1)
1331
InterTree.register_optimiser(InterTree)
1334
class MultiWalker(object):
1335
"""Walk multiple trees simultaneously, getting combined results."""
1337
# Note: This could be written to not assume you can do out-of-order
1338
# lookups. Instead any nodes that don't match in all trees could be
1339
# marked as 'deferred', and then returned in the final cleanup loop.
1340
# For now, I think it is "nicer" to return things as close to the
1341
# "master_tree" order as we can.
1343
def __init__(self, master_tree, other_trees):
1344
"""Create a new MultiWalker.
1346
All trees being walked must implement "iter_entries_by_dir()", such
1347
that they yield (path, object) tuples, where that object will have a
1348
'.file_id' member, that can be used to check equality.
1350
:param master_tree: All trees will be 'slaved' to the master_tree such
1351
that nodes in master_tree will be used as 'first-pass' sync points.
1352
Any nodes that aren't in master_tree will be merged in a second
1354
:param other_trees: A list of other trees to walk simultaneously.
1356
self._master_tree = master_tree
1357
self._other_trees = other_trees
1359
# Keep track of any nodes that were properly processed just out of
1360
# order, that way we don't return them at the end, we don't have to
1361
# track *all* processed file_ids, just the out-of-order ones
1362
self._out_of_order_processed = set()
1365
def _step_one(iterator):
1366
"""Step an iter_entries_by_dir iterator.
1368
:return: (has_more, path, ie)
1369
If has_more is False, path and ie will be None.
1372
path, ie = iterator.next()
1373
except StopIteration:
1374
return False, None, None
1376
return True, path, ie
1379
def _cmp_path_by_dirblock(path1, path2):
1380
"""Compare two paths based on what directory they are in.
1382
This generates a sort order, such that all children of a directory are
1383
sorted together, and grandchildren are in the same order as the
1384
children appear. But all grandchildren come after all children.
1386
:param path1: first path
1387
:param path2: the second path
1388
:return: negative number if ``path1`` comes first,
1389
0 if paths are equal
1390
and a positive number if ``path2`` sorts first
1392
# Shortcut this special case
1395
# This is stolen from _dirstate_helpers_py.py, only switching it to
1396
# Unicode objects. Consider using encode_utf8() and then using the
1397
# optimized versions, or maybe writing optimized unicode versions.
1398
if not isinstance(path1, unicode):
1399
raise TypeError("'path1' must be a unicode string, not %s: %r"
1400
% (type(path1), path1))
1401
if not isinstance(path2, unicode):
1402
raise TypeError("'path2' must be a unicode string, not %s: %r"
1403
% (type(path2), path2))
1404
return cmp(MultiWalker._path_to_key(path1),
1405
MultiWalker._path_to_key(path2))
1408
def _path_to_key(path):
1409
dirname, basename = osutils.split(path)
1410
return (dirname.split(u'/'), basename)
1412
def _lookup_by_file_id(self, extra_entries, other_tree, file_id):
1413
"""Lookup an inventory entry by file_id.
1415
This is called when an entry is missing in the normal order.
1416
Generally this is because a file was either renamed, or it was
1417
deleted/added. If the entry was found in the inventory and not in
1418
extra_entries, it will be added to self._out_of_order_processed
1420
:param extra_entries: A dictionary of {file_id: (path, ie)}. This
1421
should be filled with entries that were found before they were
1422
used. If file_id is present, it will be removed from the
1424
:param other_tree: The Tree to search, in case we didn't find the entry
1426
:param file_id: The file_id to look for
1427
:return: (path, ie) if found or (None, None) if not present.
1429
if file_id in extra_entries:
1430
return extra_entries.pop(file_id)
1431
# TODO: Is id2path better as the first call, or is
1432
# inventory[file_id] better as a first check?
1434
cur_path = other_tree.id2path(file_id)
1435
except errors.NoSuchId:
1437
if cur_path is None:
1440
self._out_of_order_processed.add(file_id)
1441
cur_ie = other_tree.inventory[file_id]
1442
return (cur_path, cur_ie)
1445
"""Match up the values in the different trees."""
1446
for result in self._walk_master_tree():
1448
self._finish_others()
1449
for result in self._walk_others():
1452
def _walk_master_tree(self):
1453
"""First pass, walk all trees in lock-step.
1455
When we are done, all nodes in the master_tree will have been
1456
processed. _other_walkers, _other_entries, and _others_extra will be
1457
set on 'self' for future processing.
1459
# This iterator has the most "inlining" done, because it tends to touch
1460
# every file in the tree, while the others only hit nodes that don't
1462
master_iterator = self._master_tree.iter_entries_by_dir()
1464
other_walkers = [other.iter_entries_by_dir()
1465
for other in self._other_trees]
1466
other_entries = [self._step_one(walker) for walker in other_walkers]
1467
# Track extra nodes in the other trees
1468
others_extra = [{} for i in xrange(len(self._other_trees))]
1470
master_has_more = True
1471
step_one = self._step_one
1472
lookup_by_file_id = self._lookup_by_file_id
1473
out_of_order_processed = self._out_of_order_processed
1475
while master_has_more:
1476
(master_has_more, path, master_ie) = step_one(master_iterator)
1477
if not master_has_more:
1480
file_id = master_ie.file_id
1482
other_values_append = other_values.append
1483
next_other_entries = []
1484
next_other_entries_append = next_other_entries.append
1485
for idx, (other_has_more, other_path, other_ie) in enumerate(other_entries):
1486
if not other_has_more:
1487
other_values_append(lookup_by_file_id(
1488
others_extra[idx], self._other_trees[idx], file_id))
1489
next_other_entries_append((False, None, None))
1490
elif file_id == other_ie.file_id:
1491
# This is the critical code path, as most of the entries
1492
# should match between most trees.
1493
other_values_append((other_path, other_ie))
1494
next_other_entries_append(step_one(other_walkers[idx]))
1496
# This walker did not match, step it until it either
1497
# matches, or we know we are past the current walker.
1498
other_walker = other_walkers[idx]
1499
other_extra = others_extra[idx]
1500
while (other_has_more and
1501
self._cmp_path_by_dirblock(other_path, path) < 0):
1502
other_file_id = other_ie.file_id
1503
if other_file_id not in out_of_order_processed:
1504
other_extra[other_file_id] = (other_path, other_ie)
1505
other_has_more, other_path, other_ie = \
1506
step_one(other_walker)
1507
if other_has_more and other_ie.file_id == file_id:
1508
# We ended up walking to this point, match and step
1510
other_values_append((other_path, other_ie))
1511
other_has_more, other_path, other_ie = \
1512
step_one(other_walker)
1514
# This record isn't in the normal order, see if it
1516
other_values_append(lookup_by_file_id(
1517
other_extra, self._other_trees[idx], file_id))
1518
next_other_entries_append((other_has_more, other_path,
1520
other_entries = next_other_entries
1522
# We've matched all the walkers, yield this datapoint
1523
yield path, file_id, master_ie, other_values
1524
self._other_walkers = other_walkers
1525
self._other_entries = other_entries
1526
self._others_extra = others_extra
1528
def _finish_others(self):
1529
"""Finish walking the other iterators, so we get all entries."""
1530
for idx, info in enumerate(self._other_entries):
1531
other_extra = self._others_extra[idx]
1532
(other_has_more, other_path, other_ie) = info
1533
while other_has_more:
1534
other_file_id = other_ie.file_id
1535
if other_file_id not in self._out_of_order_processed:
1536
other_extra[other_file_id] = (other_path, other_ie)
1537
other_has_more, other_path, other_ie = \
1538
self._step_one(self._other_walkers[idx])
1539
del self._other_entries
1541
def _walk_others(self):
1542
"""Finish up by walking all the 'deferred' nodes."""
1543
# TODO: One alternative would be to grab all possible unprocessed
1544
# file_ids, and then sort by path, and then yield them. That
1545
# might ensure better ordering, in case a caller strictly
1546
# requires parents before children.
1547
for idx, other_extra in enumerate(self._others_extra):
1548
others = sorted(other_extra.itervalues(),
1549
key=lambda x: self._path_to_key(x[0]))
1550
for other_path, other_ie in others:
1551
file_id = other_ie.file_id
1552
# We don't need to check out_of_order_processed here, because
1553
# the lookup_by_file_id will be removing anything processed
1554
# from the extras cache
1555
other_extra.pop(file_id)
1556
other_values = [(None, None) for i in xrange(idx)]
1557
other_values.append((other_path, other_ie))
1558
for alt_idx, alt_extra in enumerate(self._others_extra[idx+1:]):
1559
alt_idx = alt_idx + idx + 1
1560
alt_extra = self._others_extra[alt_idx]
1561
alt_tree = self._other_trees[alt_idx]
1562
other_values.append(self._lookup_by_file_id(
1563
alt_extra, alt_tree, file_id))
1564
yield other_path, file_id, None, other_values
238
# There is no longer a file of this name, but we can describe
239
# what happened to the file that used to have
240
# this name. There are two possibilities: either it was
241
# deleted entirely, or renamed.
243
if new_inv.has_id(old_id):
244
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
246
return 'D', old_inv.id2path(old_id), None
248
# if the file_id is new in this revision, it is added
249
if new_id and not old_inv.has_id(new_id):
252
# if there used to be a file of this name, but that ID has now
253
# disappeared, it is deleted
254
if old_id and not new_inv.has_id(old_id):
261
def find_renames(old_inv, new_inv):
262
for file_id in old_inv:
263
if file_id not in new_inv:
265
old_name = old_inv.id2path(file_id)
266
new_name = new_inv.id2path(file_id)
267
if old_name != new_name:
268
yield (old_name, new_name)
272
######################################################################
275
def dir_exporter(tree, dest, root):
276
"""Export this tree to a new directory.
278
`dest` should not exist, and will be created holding the
279
contents of this tree.
281
TODO: To handle subdirectories we need to create the
284
:note: If the export fails, the destination directory will be
285
left in a half-assed state.
289
mutter('export version %r' % tree)
291
for dp, ie in inv.iter_entries():
292
ie.put_on_disk(dest, dp, tree)
294
exporters['dir'] = dir_exporter
301
def get_root_name(dest):
302
"""Get just the root name for a tarball.
304
>>> get_root_name('mytar.tar')
306
>>> get_root_name('mytar.tar.bz2')
308
>>> get_root_name('tar.tar.tar.tgz')
310
>>> get_root_name('bzr-0.0.5.tar.gz')
312
>>> get_root_name('a/long/path/mytar.tgz')
314
>>> get_root_name('../parent/../dir/other.tbz2')
317
endings = ['.tar', '.tar.gz', '.tgz', '.tar.bz2', '.tbz2']
318
dest = os.path.basename(dest)
320
if dest.endswith(end):
321
return dest[:-len(end)]
323
def tar_exporter(tree, dest, root, compression=None):
324
"""Export this tree to a new tar file.
326
`dest` will be created holding the contents of this tree; if it
327
already exists, it will be clobbered, like with "tar -c".
329
from time import time
331
compression = str(compression or '')
333
root = get_root_name(dest)
335
ball = tarfile.open(dest, 'w:' + compression)
336
except tarfile.CompressionError, e:
337
raise BzrError(str(e))
338
mutter('export version %r' % tree)
340
for dp, ie in inv.iter_entries():
341
mutter(" export {%s} kind %s to %s" % (ie.file_id, ie.kind, dest))
342
item, fileobj = ie.get_tar_item(root, dp, now, tree)
343
ball.addfile(item, fileobj)
346
exporters['tar'] = tar_exporter
348
def tgz_exporter(tree, dest, root):
349
tar_exporter(tree, dest, root, compression='gz')
350
exporters['tgz'] = tgz_exporter
352
def tbz_exporter(tree, dest, root):
353
tar_exporter(tree, dest, root, compression='bz2')
354
exporters['tbz2'] = tbz_exporter
added
removed
bzrlib/tree.py
