53
60
Trees can be compared, etc, regardless of whether they are working
54
61
trees or versioned trees.
64
def changes_from(self, other, want_unchanged=False, specific_files=None,
65
extra_trees=None, require_versioned=False, include_root=False,
66
want_unversioned=False):
67
"""Return a TreeDelta of the changes from other to this tree.
69
:param other: A tree to compare with.
70
:param specific_files: An optional list of file paths to restrict the
71
comparison to. When mapping filenames to ids, all matches in all
72
trees (including optional extra_trees) are used, and all children of
73
matched directories are included.
74
:param want_unchanged: An optional boolean requesting the inclusion of
75
unchanged entries in the result.
76
:param extra_trees: An optional list of additional trees to use when
77
mapping the contents of specific_files (paths) to file_ids.
78
:param require_versioned: An optional boolean (defaults to False). When
79
supplied and True all the 'specific_files' must be versioned, or
80
a PathsNotVersionedError will be thrown.
81
:param want_unversioned: Scan for unversioned paths.
83
The comparison will be performed by an InterTree object looked up on
86
# Martin observes that Tree.changes_from returns a TreeDelta and this
87
# may confuse people, because the class name of the returned object is
88
# a synonym of the object referenced in the method name.
89
return InterTree.get(other, self).compare(
90
want_unchanged=want_unchanged,
91
specific_files=specific_files,
92
extra_trees=extra_trees,
93
require_versioned=require_versioned,
94
include_root=include_root,
95
want_unversioned=want_unversioned,
98
def iter_changes(self, from_tree, include_unchanged=False,
99
specific_files=None, pb=None, extra_trees=None,
100
require_versioned=True, want_unversioned=False):
101
"""See InterTree.iter_changes"""
102
intertree = InterTree.get(from_tree, self)
103
return intertree.iter_changes(include_unchanged, specific_files, pb,
104
extra_trees, require_versioned, want_unversioned=want_unversioned)
107
"""Get a list of the conflicts in the tree.
109
Each conflict is an instance of bzrlib.conflicts.Conflict.
111
return _mod_conflicts.ConflictList()
114
"""For trees that can have unversioned files, return all such paths."""
117
def get_parent_ids(self):
118
"""Get the parent ids for this tree.
120
:return: a list of parent ids. [] is returned to indicate
121
a tree with no parents.
122
:raises: BzrError if the parents are not known.
124
raise NotImplementedError(self.get_parent_ids)
57
126
def has_filename(self, filename):
58
127
"""True if the tree has given filename."""
59
raise NotImplementedError()
128
raise NotImplementedError(self.has_filename)
61
130
def has_id(self, file_id):
62
131
return self.inventory.has_id(file_id)
65
"""Return set of all ids in this tree."""
66
return self.inventory.id_set()
133
def __contains__(self, file_id):
134
return self.has_id(file_id)
136
def has_or_had_id(self, file_id):
137
return self.inventory.has_id(file_id)
139
def is_ignored(self, filename):
140
"""Check whether the filename is ignored by this tree.
142
:param filename: The relative filename within the tree.
143
:return: True if the filename is ignored.
148
return iter(self.inventory)
150
def all_file_ids(self):
151
"""Iterate through all file ids, including ids for missing files."""
152
return set(self.inventory)
68
154
def id2path(self, file_id):
155
"""Return the path for a file id.
69
159
return self.inventory.id2path(file_id)
161
def is_control_filename(self, filename):
162
"""True if filename is the name of a control file in this tree.
164
:param filename: A filename within the tree. This is a relative path
165
from the root of this tree.
167
This is true IF and ONLY IF the filename is part of the meta data
168
that bzr controls in this tree. I.E. a random .bzr directory placed
169
on disk will not be a control file for this tree.
171
return self.bzrdir.is_control_filename(filename)
174
def iter_entries_by_dir(self, specific_file_ids=None, yield_parents=False):
175
"""Walk the tree in 'by_dir' order.
177
This will yield each entry in the tree as a (path, entry) tuple.
178
The order that they are yielded is:
180
Directories are walked in a depth-first lexicographical order,
181
however, whenever a directory is reached, all of its direct child
182
nodes are yielded in lexicographical order before yielding the
185
For example, in the tree::
195
The yield order (ignoring root) would be::
196
a, f, a/b, a/d, a/b/c, a/d/e, f/g
198
:param yield_parents: If True, yield the parents from the root leading
199
down to specific_file_ids that have been requested. This has no
200
impact if specific_file_ids is None.
202
return self.inventory.iter_entries_by_dir(
203
specific_file_ids=specific_file_ids, yield_parents=yield_parents)
205
def iter_references(self):
206
if self.supports_tree_reference():
207
for path, entry in self.iter_entries_by_dir():
208
if entry.kind == 'tree-reference':
209
yield path, entry.file_id
211
def kind(self, file_id):
212
raise NotImplementedError("Tree subclass %s must implement kind"
213
% self.__class__.__name__)
215
def stored_kind(self, file_id):
216
"""File kind stored for this file_id.
218
May not match kind on disk for working trees. Always available
219
for versioned files, even when the file itself is missing.
221
return self.kind(file_id)
223
def path_content_summary(self, path):
224
"""Get a summary of the information about path.
226
All the attributes returned are for the canonical form, not the
227
convenient form (if content filters are in use.)
229
:param path: A relative path within the tree.
230
:return: A tuple containing kind, size, exec, sha1-or-link.
231
Kind is always present (see tree.kind()).
232
size is present if kind is file and the size of the
233
canonical form can be cheaply determined, None otherwise.
234
exec is None unless kind is file and the platform supports the 'x'
236
sha1-or-link is the link target if kind is symlink, or the sha1 if
237
it can be obtained without reading the file.
239
raise NotImplementedError(self.path_content_summary)
241
def get_reference_revision(self, file_id, path=None):
242
raise NotImplementedError("Tree subclass %s must implement "
243
"get_reference_revision"
244
% self.__class__.__name__)
246
def _comparison_data(self, entry, path):
247
"""Return a tuple of kind, executable, stat_value for a file.
249
entry may be None if there is no inventory entry for the file, but
250
path must always be supplied.
252
kind is None if there is no file present (even if an inventory id is
253
present). executable is False for non-file entries.
255
raise NotImplementedError(self._comparison_data)
257
def _file_size(self, entry, stat_value):
258
raise NotImplementedError(self._file_size)
71
260
def _get_inventory(self):
72
261
return self._inventory
263
def get_file(self, file_id, path=None):
264
"""Return a file object for the file file_id in the tree.
266
If both file_id and path are defined, it is implementation defined as
267
to which one is used.
269
raise NotImplementedError(self.get_file)
271
def get_file_with_stat(self, file_id, path=None):
272
"""Get a file handle and stat object for file_id.
274
The default implementation returns (self.get_file, None) for backwards
277
:param file_id: The file id to read.
278
:param path: The path of the file, if it is known.
279
:return: A tuple (file_handle, stat_value_or_None). If the tree has
280
no stat facility, or need for a stat cache feedback during commit,
281
it may return None for the second element of the tuple.
283
return (self.get_file(file_id, path), None)
285
def get_file_text(self, file_id, path=None):
286
"""Return the byte content of a file.
288
:param file_id: The file_id of the file.
289
:param path: The path of the file.
290
If both file_id and path are supplied, an implementation may use
293
my_file = self.get_file(file_id, path)
295
return my_file.read()
299
def get_file_lines(self, file_id, path=None):
300
"""Return the content of a file, as lines.
302
:param file_id: The file_id of the file.
303
:param path: The path of the file.
304
If both file_id and path are supplied, an implementation may use
307
return osutils.split_lines(self.get_file_text(file_id, path))
309
def get_file_mtime(self, file_id, path=None):
310
"""Return the modification time for a file.
312
:param file_id: The handle for this file.
313
:param path: The path that this file can be found at.
314
These must point to the same object.
316
raise NotImplementedError(self.get_file_mtime)
318
def get_file_size(self, file_id):
319
"""Return the size of a file in bytes.
321
This applies only to regular files. If invoked on directories or
322
symlinks, it will return None.
323
:param file_id: The file-id of the file
325
raise NotImplementedError(self.get_file_size)
327
def get_file_by_path(self, path):
328
return self.get_file(self._inventory.path2id(path), path)
330
def iter_files_bytes(self, desired_files):
331
"""Iterate through file contents.
333
Files will not necessarily be returned in the order they occur in
334
desired_files. No specific order is guaranteed.
336
Yields pairs of identifier, bytes_iterator. identifier is an opaque
337
value supplied by the caller as part of desired_files. It should
338
uniquely identify the file version in the caller's context. (Examples:
339
an index number or a TreeTransform trans_id.)
341
bytes_iterator is an iterable of bytestrings for the file. The
342
kind of iterable and length of the bytestrings are unspecified, but for
343
this implementation, it is a tuple containing a single bytestring with
344
the complete text of the file.
346
:param desired_files: a list of (file_id, identifier) pairs
348
for file_id, identifier in desired_files:
349
# We wrap the string in a tuple so that we can return an iterable
350
# of bytestrings. (Technically, a bytestring is also an iterable
351
# of bytestrings, but iterating through each character is not
353
cur_file = (self.get_file_text(file_id),)
354
yield identifier, cur_file
356
def get_symlink_target(self, file_id):
357
"""Get the target for a given file_id.
359
It is assumed that the caller already knows that file_id is referencing
361
:param file_id: Handle for the symlink entry.
362
:return: The path the symlink points to.
364
raise NotImplementedError(self.get_symlink_target)
366
def get_canonical_inventory_paths(self, paths):
367
"""Like get_canonical_inventory_path() but works on multiple items.
369
:param paths: A sequence of paths relative to the root of the tree.
370
:return: A list of paths, with each item the corresponding input path
371
adjusted to account for existing elements that match case
374
return list(self._yield_canonical_inventory_paths(paths))
376
def get_canonical_inventory_path(self, path):
377
"""Returns the first inventory item that case-insensitively matches path.
379
If a path matches exactly, it is returned. If no path matches exactly
380
but more than one path matches case-insensitively, it is implementation
381
defined which is returned.
383
If no path matches case-insensitively, the input path is returned, but
384
with as many path entries that do exist changed to their canonical
387
If you need to resolve many names from the same tree, you should
388
use get_canonical_inventory_paths() to avoid O(N) behaviour.
390
:param path: A paths relative to the root of the tree.
391
:return: The input path adjusted to account for existing elements
392
that match case insensitively.
394
return self._yield_canonical_inventory_paths([path]).next()
396
def _yield_canonical_inventory_paths(self, paths):
398
# First, if the path as specified exists exactly, just use it.
399
if self.path2id(path) is not None:
403
cur_id = self.get_root_id()
405
bit_iter = iter(path.split("/"))
408
for child in self.iter_children(cur_id):
410
child_base = os.path.basename(self.id2path(child))
411
if child_base.lower() == lelt:
413
cur_path = osutils.pathjoin(cur_path, child_base)
416
# before a change is committed we can see this error...
419
# got to the end of this directory and no entries matched.
420
# Return what matched so far, plus the rest as specified.
421
cur_path = osutils.pathjoin(cur_path, elt, *list(bit_iter))
426
def get_root_id(self):
427
"""Return the file_id for the root of this tree."""
428
raise NotImplementedError(self.get_root_id)
430
def annotate_iter(self, file_id,
431
default_revision=_mod_revision.CURRENT_REVISION):
432
"""Return an iterator of revision_id, line tuples.
434
For working trees (and mutable trees in general), the special
435
revision_id 'current:' will be used for lines that are new in this
436
tree, e.g. uncommitted changes.
437
:param file_id: The file to produce an annotated version from
438
:param default_revision: For lines that don't match a basis, mark them
439
with this revision id. Not all implementations will make use of
442
raise NotImplementedError(self.annotate_iter)
444
def _get_plan_merge_data(self, file_id, other, base):
445
from bzrlib import versionedfile
446
vf = versionedfile._PlanMergeVersionedFile(file_id)
447
last_revision_a = self._get_file_revision(file_id, vf, 'this:')
448
last_revision_b = other._get_file_revision(file_id, vf, 'other:')
450
last_revision_base = None
452
last_revision_base = base._get_file_revision(file_id, vf, 'base:')
453
return vf, last_revision_a, last_revision_b, last_revision_base
455
def plan_file_merge(self, file_id, other, base=None):
456
"""Generate a merge plan based on annotations.
458
If the file contains uncommitted changes in this tree, they will be
459
attributed to the 'current:' pseudo-revision. If the file contains
460
uncommitted changes in the other tree, they will be assigned to the
461
'other:' pseudo-revision.
463
data = self._get_plan_merge_data(file_id, other, base)
464
vf, last_revision_a, last_revision_b, last_revision_base = data
465
return vf.plan_merge(last_revision_a, last_revision_b,
468
def plan_file_lca_merge(self, file_id, other, base=None):
469
"""Generate a merge plan based lca-newness.
471
If the file contains uncommitted changes in this tree, they will be
472
attributed to the 'current:' pseudo-revision. If the file contains
473
uncommitted changes in the other tree, they will be assigned to the
474
'other:' pseudo-revision.
476
data = self._get_plan_merge_data(file_id, other, base)
477
vf, last_revision_a, last_revision_b, last_revision_base = data
478
return vf.plan_lca_merge(last_revision_a, last_revision_b,
481
def _iter_parent_trees(self):
482
"""Iterate through parent trees, defaulting to Tree.revision_tree."""
483
for revision_id in self.get_parent_ids():
485
yield self.revision_tree(revision_id)
486
except errors.NoSuchRevisionInTree:
487
yield self.repository.revision_tree(revision_id)
490
def _file_revision(revision_tree, file_id):
491
"""Determine the revision associated with a file in a given tree."""
492
revision_tree.lock_read()
494
return revision_tree.inventory[file_id].revision
496
revision_tree.unlock()
498
def _get_file_revision(self, file_id, vf, tree_revision):
499
"""Ensure that file_id, tree_revision is in vf to plan the merge."""
501
if getattr(self, '_repository', None) is None:
502
last_revision = tree_revision
503
parent_keys = [(file_id, self._file_revision(t, file_id)) for t in
504
self._iter_parent_trees()]
505
vf.add_lines((file_id, last_revision), parent_keys,
506
self.get_file(file_id).readlines())
507
repo = self.branch.repository
510
last_revision = self._file_revision(self, file_id)
511
base_vf = self._repository.texts
512
if base_vf not in vf.fallback_versionedfiles:
513
vf.fallback_versionedfiles.append(base_vf)
74
516
inventory = property(_get_inventory,
75
517
doc="Inventory of this Tree")
77
519
def _check_retrieved(self, ie, f):
78
522
fp = fingerprint_file(f)
81
525
if ie.text_size is not None:
82
526
if ie.text_size != fp['size']:
83
bailout("mismatched size for file %r in %r" % (ie.file_id, self._store),
527
raise BzrError("mismatched size for file %r in %r" % (ie.file_id, self._store),
84
528
["inventory expects %d bytes" % ie.text_size,
85
529
"file is actually %d bytes" % fp['size'],
86
530
"store is probably damaged/corrupt"])
88
532
if ie.text_sha1 != fp['sha1']:
89
bailout("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
533
raise BzrError("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
90
534
["inventory expects %s" % ie.text_sha1,
91
535
"file is actually %s" % fp['sha1'],
92
536
"store is probably damaged/corrupt"])
95
def export(self, dest):
96
"""Export this tree to a new directory.
98
`dest` should not exist, and will be created holding the
99
contents of this tree.
101
:todo: To handle subdirectories we need to create the
104
:note: If the export fails, the destination directory will be
105
left in a half-assed state.
108
mutter('export version %r' % self)
110
for dp, ie in inv.iter_entries():
112
fullpath = appendpath(dest, dp)
113
if kind == 'directory':
116
pumpfile(self.get_file(ie.file_id), file(fullpath, 'wb'))
118
bailout("don't know how to export {%s} of kind %r", fid, kind)
119
mutter(" export {%s} kind %s to %s" % (ie.file_id, kind, fullpath))
123
class WorkingTree(Tree):
124
"""Working copy tree.
126
The inventory is held in the `Branch` working-inventory, and the
127
files are in a directory on disk.
129
It is possible for a `WorkingTree` to have a filename which is
130
not listed in the Inventory and vice versa.
132
def __init__(self, basedir, inv):
133
self._inventory = inv
134
self.basedir = basedir
135
self.path2id = inv.path2id
138
return "<%s of %s>" % (self.__class__.__name__,
141
def abspath(self, filename):
142
return os.path.join(self.basedir, filename)
144
def has_filename(self, filename):
145
return os.path.exists(self.abspath(filename))
147
def get_file(self, file_id):
148
return self.get_file_byname(self.id2path(file_id))
150
def get_file_byname(self, filename):
151
return file(self.abspath(filename), 'rb')
153
def _get_store_filename(self, file_id):
154
return self.abspath(self.id2path(file_id))
156
def has_id(self, file_id):
157
# files that have been deleted are excluded
158
if not self.inventory.has_id(file_id):
160
return os.access(self.abspath(self.inventory.id2path(file_id)), os.F_OK)
162
def get_file_size(self, file_id):
163
return os.stat(self._get_store_filename(file_id))[ST_SIZE]
165
def get_file_sha1(self, file_id):
166
f = self.get_file(file_id)
170
def file_class(self, filename):
171
if self.path2id(filename):
173
elif self.is_ignored(filename):
179
def list_files(self):
180
"""Recursively list all files as (path, class, kind, id).
182
Lists, but does not descend into unversioned directories.
184
This does not include files that have been deleted in this
187
Skips the control directory.
191
def descend(from_dir, from_dir_id, dp):
195
## TODO: If we find a subdirectory with its own .bzr
196
## directory, then that is a separate tree and we
197
## should exclude it.
198
if bzrlib.BZRDIR == f:
202
fp = appendpath(from_dir, f)
205
fap = appendpath(dp, f)
207
f_ie = inv.get_child(from_dir_id, f)
210
elif self.is_ignored(fp):
219
bailout("file %r entered as kind %r id %r, now of kind %r"
220
% (fap, f_ie.kind, f_ie.file_id, fk))
222
yield fp, c, fk, (f_ie and f_ie.file_id)
224
if fk != 'directory':
228
# don't descend unversioned directories
231
for ff in descend(fp, f_ie.file_id, fap):
234
for f in descend('', None, self.basedir):
539
def path2id(self, path):
540
"""Return the id for path in this tree."""
541
return self._inventory.path2id(path)
543
def paths2ids(self, paths, trees=[], require_versioned=True):
544
"""Return all the ids that can be reached by walking from paths.
546
Each path is looked up in this tree and any extras provided in
547
trees, and this is repeated recursively: the children in an extra tree
548
of a directory that has been renamed under a provided path in this tree
549
are all returned, even if none exist under a provided path in this
550
tree, and vice versa.
552
:param paths: An iterable of paths to start converting to ids from.
553
Alternatively, if paths is None, no ids should be calculated and None
554
will be returned. This is offered to make calling the api unconditional
555
for code that *might* take a list of files.
556
:param trees: Additional trees to consider.
557
:param require_versioned: If False, do not raise NotVersionedError if
558
an element of paths is not versioned in this tree and all of trees.
560
return find_ids_across_trees(paths, [self] + list(trees), require_versioned)
562
def iter_children(self, file_id):
563
entry = self.iter_entries_by_dir([file_id]).next()[1]
564
for child in getattr(entry, 'children', {}).itervalues():
570
def revision_tree(self, revision_id):
571
"""Obtain a revision tree for the revision revision_id.
573
The intention of this method is to allow access to possibly cached
574
tree data. Implementors of this method should raise NoSuchRevision if
575
the tree is not locally available, even if they could obtain the
576
tree via a repository or some other means. Callers are responsible
577
for finding the ultimate source for a revision tree.
579
:param revision_id: The revision_id of the requested tree.
581
:raises: NoSuchRevision if the tree cannot be obtained.
583
raise errors.NoSuchRevisionInTree(self, revision_id)
239
585
def unknowns(self):
240
for subp in self.extras():
241
if not self.is_ignored(subp):
246
"""Yield all unknown files in this WorkingTree.
248
If there are any unknown directories then only the directory is
249
returned, not all its children. But if there are unknown files
250
under a versioned subdirectory, they are returned.
252
Currently returned depth-first, sorted by name within directories.
254
## TODO: Work from given directory downwards
256
for path, dir_entry in self.inventory.directories():
257
mutter("search for unknowns in %r" % path)
258
dirabs = self.abspath(path)
259
if not isdir(dirabs):
260
# e.g. directory deleted
264
for subf in os.listdir(dirabs):
266
and (subf not in dir_entry.children)):
271
subp = appendpath(path, subf)
275
def ignored_files(self):
276
"""Yield list of PATH, IGNORE_PATTERN"""
277
for subp in self.extras():
278
pat = self.is_ignored(subp)
283
def get_ignore_list(self):
284
"""Return list of ignore patterns.
286
Cached in the Tree object after the first call.
288
if hasattr(self, '_ignorelist'):
289
return self._ignorelist
291
l = bzrlib.DEFAULT_IGNORE[:]
292
if self.has_filename(bzrlib.IGNORE_FILENAME):
293
f = self.get_file_byname(bzrlib.IGNORE_FILENAME)
294
l.extend([line.rstrip("\n\r") for line in f.readlines()])
299
def is_ignored(self, filename):
300
"""Check whether the filename matches an ignore pattern.
302
Patterns containing '/' need to match the whole path; others
303
match against only the last component.
305
If the file is ignored, returns the pattern which caused it to
306
be ignored, otherwise None. So this can simply be used as a
307
boolean if desired."""
309
## TODO: Use '**' to match directories, and other extended globbing stuff from cvs/rsync.
311
for pat in self.get_ignore_list():
313
# as a special case, you can put ./ at the start of a pattern;
314
# this is good to match in the top-level only;
319
if fnmatch.fnmatchcase(filename, newpat):
586
"""What files are present in this tree and unknown.
588
:return: an iterator over the unknown files.
595
def filter_unversioned_files(self, paths):
596
"""Filter out paths that are versioned.
598
:return: set of paths.
600
# NB: we specifically *don't* call self.has_filename, because for
601
# WorkingTrees that can indicate files that exist on disk but that
603
pred = self.inventory.has_filename
604
return set((p for p in paths if not pred(p)))
606
def walkdirs(self, prefix=""):
607
"""Walk the contents of this tree from path down.
609
This yields all the data about the contents of a directory at a time.
610
After each directory has been yielded, if the caller has mutated the
611
list to exclude some directories, they are then not descended into.
613
The data yielded is of the form:
614
((directory-relpath, directory-path-from-root, directory-fileid),
615
[(relpath, basename, kind, lstat, path_from_tree_root, file_id,
616
versioned_kind), ...]),
617
- directory-relpath is the containing dirs relpath from prefix
618
- directory-path-from-root is the containing dirs path from /
619
- directory-fileid is the id of the directory if it is versioned.
620
- relpath is the relative path within the subtree being walked.
621
- basename is the basename
622
- kind is the kind of the file now. If unknonwn then the file is not
623
present within the tree - but it may be recorded as versioned. See
625
- lstat is the stat data *if* the file was statted.
626
- path_from_tree_root is the path from the root of the tree.
627
- file_id is the file_id if the entry is versioned.
628
- versioned_kind is the kind of the file as last recorded in the
629
versioning system. If 'unknown' the file is not versioned.
630
One of 'kind' and 'versioned_kind' must not be 'unknown'.
632
:param prefix: Start walking from prefix within the tree rather than
633
at the root. This allows one to walk a subtree but get paths that are
634
relative to a tree rooted higher up.
635
:return: an iterator over the directory data.
637
raise NotImplementedError(self.walkdirs)
639
def supports_content_filtering(self):
642
def _content_filter_stack(self, path=None, file_id=None):
643
"""The stack of content filters for a path if filtering is supported.
645
Readers will be applied in first-to-last order.
646
Writers will be applied in last-to-first order.
647
Either the path or the file-id needs to be provided.
649
:param path: path relative to the root of the tree
651
:param file_id: file_id or None if unknown
652
:return: the list of filters - [] if there are none
654
filter_pref_names = filters._get_registered_names()
655
if len(filter_pref_names) == 0:
658
path = self.id2path(file_id)
659
prefs = self.iter_search_rules([path], filter_pref_names).next()
660
stk = filters._get_filter_stack_for(prefs)
661
if 'filters' in debug.debug_flags:
662
note("*** %s content-filter: %s => %r" % (path,prefs,stk))
665
def _content_filter_stack_provider(self):
666
"""A function that returns a stack of ContentFilters.
668
The function takes a path (relative to the top of the tree) and a
669
file-id as parameters.
671
:return: None if content filtering is not supported by this tree.
673
if self.supports_content_filtering():
674
return lambda path, file_id: \
675
self._content_filter_stack(path, file_id)
679
def iter_search_rules(self, path_names, pref_names=None,
680
_default_searcher=None):
681
"""Find the preferences for filenames in a tree.
683
:param path_names: an iterable of paths to find attributes for.
684
Paths are given relative to the root of the tree.
685
:param pref_names: the list of preferences to lookup - None for all
686
:param _default_searcher: private parameter to assist testing - don't use
687
:return: an iterator of tuple sequences, one per path-name.
688
See _RulesSearcher.get_items for details on the tuple sequence.
690
if _default_searcher is None:
691
_default_searcher = rules._per_user_searcher
692
searcher = self._get_rules_searcher(_default_searcher)
693
if searcher is not None:
694
if pref_names is not None:
695
for path in path_names:
696
yield searcher.get_selected_items(path, pref_names)
322
if fnmatch.fnmatchcase(splitpath(filename)[-1], pat):
330
class RevisionTree(Tree):
331
"""Tree viewing a previous revision.
333
File text can be retrieved from the text store.
335
:todo: Some kind of `__repr__` method, but a good one
336
probably means knowing the branch and revision number,
337
or at least passing a description to the constructor.
340
def __init__(self, store, inv):
342
self._inventory = inv
344
def get_file(self, file_id):
345
ie = self._inventory[file_id]
346
f = self._store[ie.text_id]
347
mutter(" get fileid{%s} from %r" % (file_id, self))
348
self._check_retrieved(ie, f)
351
def get_file_size(self, file_id):
352
return self._inventory[file_id].text_size
354
def get_file_sha1(self, file_id):
355
ie = self._inventory[file_id]
358
def has_filename(self, filename):
359
return bool(self.inventory.path2id(filename))
361
def list_files(self):
362
# The only files returned by this are those from the version
363
for path, entry in self.inventory.iter_entries():
364
yield path, 'V', entry.kind, entry.file_id
367
class EmptyTree(Tree):
369
self._inventory = Inventory()
371
def has_filename(self, filename):
374
def list_files(self):
375
if False: # just to make it a generator
698
for path in path_names:
699
yield searcher.get_items(path)
701
def _get_rules_searcher(self, default_searcher):
702
"""Get the RulesSearcher for this tree given the default one."""
703
searcher = default_searcher
380
707
######################################################################
440
766
new_name = new_inv.id2path(file_id)
441
767
if old_name != new_name:
442
768
yield (old_name, new_name)
771
def find_ids_across_trees(filenames, trees, require_versioned=True):
772
"""Find the ids corresponding to specified filenames.
774
All matches in all trees will be used, and all children of matched
775
directories will be used.
777
:param filenames: The filenames to find file_ids for (if None, returns
779
:param trees: The trees to find file_ids within
780
:param require_versioned: if true, all specified filenames must occur in
782
:return: a set of file ids for the specified filenames and their children.
786
specified_path_ids = _find_ids_across_trees(filenames, trees,
788
return _find_children_across_trees(specified_path_ids, trees)
791
def _find_ids_across_trees(filenames, trees, require_versioned):
792
"""Find the ids corresponding to specified filenames.
794
All matches in all trees will be used, but subdirectories are not scanned.
796
:param filenames: The filenames to find file_ids for
797
:param trees: The trees to find file_ids within
798
:param require_versioned: if true, all specified filenames must occur in
800
:return: a set of file ids for the specified filenames
803
interesting_ids = set()
804
for tree_path in filenames:
807
file_id = tree.path2id(tree_path)
808
if file_id is not None:
809
interesting_ids.add(file_id)
812
not_versioned.append(tree_path)
813
if len(not_versioned) > 0 and require_versioned:
814
raise errors.PathsNotVersionedError(not_versioned)
815
return interesting_ids
818
def _find_children_across_trees(specified_ids, trees):
819
"""Return a set including specified ids and their children.
821
All matches in all trees will be used.
823
:param trees: The trees to find file_ids within
824
:return: a set containing all specified ids and their children
826
interesting_ids = set(specified_ids)
827
pending = interesting_ids
828
# now handle children of interesting ids
829
# we loop so that we handle all children of each id in both trees
830
while len(pending) > 0:
832
for file_id in pending:
834
if not tree.has_or_had_id(file_id):
836
for child_id in tree.iter_children(file_id):
837
if child_id not in interesting_ids:
838
new_pending.add(child_id)
839
interesting_ids.update(new_pending)
840
pending = new_pending
841
return interesting_ids
844
class InterTree(InterObject):
845
"""This class represents operations taking place between two Trees.
847
Its instances have methods like 'compare' and contain references to the
848
source and target trees these operations are to be carried out on.
850
Clients of bzrlib should not need to use InterTree directly, rather they
851
should use the convenience methods on Tree such as 'Tree.compare()' which
852
will pass through to InterTree as appropriate.
855
# Formats that will be used to test this InterTree. If both are
856
# None, this InterTree will not be tested (e.g. because a complex
858
_matching_from_tree_format = None
859
_matching_to_tree_format = None
863
def _changes_from_entries(self, source_entry, target_entry,
864
source_path=None, target_path=None):
865
"""Generate a iter_changes tuple between source_entry and target_entry.
867
:param source_entry: An inventory entry from self.source, or None.
868
:param target_entry: An inventory entry from self.target, or None.
869
:param source_path: The path of source_entry, if known. If not known
870
it will be looked up.
871
:param target_path: The path of target_entry, if known. If not known
872
it will be looked up.
873
:return: A tuple, item 0 of which is an iter_changes result tuple, and
874
item 1 is True if there are any changes in the result tuple.
876
if source_entry is None:
877
if target_entry is None:
879
file_id = target_entry.file_id
881
file_id = source_entry.file_id
882
if source_entry is not None:
883
source_versioned = True
884
source_name = source_entry.name
885
source_parent = source_entry.parent_id
886
if source_path is None:
887
source_path = self.source.id2path(file_id)
888
source_kind, source_executable, source_stat = \
889
self.source._comparison_data(source_entry, source_path)
891
source_versioned = False
895
source_executable = None
896
if target_entry is not None:
897
target_versioned = True
898
target_name = target_entry.name
899
target_parent = target_entry.parent_id
900
if target_path is None:
901
target_path = self.target.id2path(file_id)
902
target_kind, target_executable, target_stat = \
903
self.target._comparison_data(target_entry, target_path)
905
target_versioned = False
909
target_executable = None
910
versioned = (source_versioned, target_versioned)
911
kind = (source_kind, target_kind)
912
changed_content = False
913
if source_kind != target_kind:
914
changed_content = True
915
elif source_kind == 'file':
916
if (self.source.get_file_sha1(file_id, source_path, source_stat) !=
917
self.target.get_file_sha1(file_id, target_path, target_stat)):
918
changed_content = True
919
elif source_kind == 'symlink':
920
if (self.source.get_symlink_target(file_id) !=
921
self.target.get_symlink_target(file_id)):
922
changed_content = True
923
# XXX: Yes, the indentation below is wrong. But fixing it broke
924
# test_merge.TestMergerEntriesLCAOnDisk.
925
# test_nested_tree_subtree_renamed_and_modified. We'll wait for
926
# the fix from bzr.dev -- vila 2009026
927
elif source_kind == 'tree-reference':
928
if (self.source.get_reference_revision(file_id, source_path)
929
!= self.target.get_reference_revision(file_id, target_path)):
930
changed_content = True
931
parent = (source_parent, target_parent)
932
name = (source_name, target_name)
933
executable = (source_executable, target_executable)
934
if (changed_content is not False or versioned[0] != versioned[1]
935
or parent[0] != parent[1] or name[0] != name[1] or
936
executable[0] != executable[1]):
940
return (file_id, (source_path, target_path), changed_content,
941
versioned, parent, name, kind, executable), changes
944
def compare(self, want_unchanged=False, specific_files=None,
945
extra_trees=None, require_versioned=False, include_root=False,
946
want_unversioned=False):
947
"""Return the changes from source to target.
949
:return: A TreeDelta.
950
:param specific_files: An optional list of file paths to restrict the
951
comparison to. When mapping filenames to ids, all matches in all
952
trees (including optional extra_trees) are used, and all children of
953
matched directories are included.
954
:param want_unchanged: An optional boolean requesting the inclusion of
955
unchanged entries in the result.
956
:param extra_trees: An optional list of additional trees to use when
957
mapping the contents of specific_files (paths) to file_ids.
958
:param require_versioned: An optional boolean (defaults to False). When
959
supplied and True all the 'specific_files' must be versioned, or
960
a PathsNotVersionedError will be thrown.
961
:param want_unversioned: Scan for unversioned paths.
963
trees = (self.source,)
964
if extra_trees is not None:
965
trees = trees + tuple(extra_trees)
966
# target is usually the newer tree:
967
specific_file_ids = self.target.paths2ids(specific_files, trees,
968
require_versioned=require_versioned)
969
if specific_files and not specific_file_ids:
970
# All files are unversioned, so just return an empty delta
971
# _compare_trees would think we want a complete delta
972
result = delta.TreeDelta()
973
fake_entry = InventoryFile('unused', 'unused', 'unused')
974
result.unversioned = [(path, None,
975
self.target._comparison_data(fake_entry, path)[0]) for path in
978
return delta._compare_trees(self.source, self.target, want_unchanged,
979
specific_files, include_root, extra_trees=extra_trees,
980
require_versioned=require_versioned,
981
want_unversioned=want_unversioned)
983
def iter_changes(self, include_unchanged=False,
984
specific_files=None, pb=None, extra_trees=[],
985
require_versioned=True, want_unversioned=False):
986
"""Generate an iterator of changes between trees.
989
(file_id, (path_in_source, path_in_target),
990
changed_content, versioned, parent, name, kind,
993
Changed_content is True if the file's content has changed. This
994
includes changes to its kind, and to a symlink's target.
996
versioned, parent, name, kind, executable are tuples of (from, to).
997
If a file is missing in a tree, its kind is None.
999
Iteration is done in parent-to-child order, relative to the target
1002
There is no guarantee that all paths are in sorted order: the
1003
requirement to expand the search due to renames may result in children
1004
that should be found early being found late in the search, after
1005
lexically later results have been returned.
1006
:param require_versioned: Raise errors.PathsNotVersionedError if a
1007
path in the specific_files list is not versioned in one of
1008
source, target or extra_trees.
1009
:param specific_files: An optional list of file paths to restrict the
1010
comparison to. When mapping filenames to ids, all matches in all
1011
trees (including optional extra_trees) are used, and all children
1012
of matched directories are included. The parents in the target tree
1013
of the specific files up to and including the root of the tree are
1014
always evaluated for changes too.
1015
:param want_unversioned: Should unversioned files be returned in the
1016
output. An unversioned file is defined as one with (False, False)
1017
for the versioned pair.
1019
lookup_trees = [self.source]
1021
lookup_trees.extend(extra_trees)
1022
# The ids of items we need to examine to insure delta consistency.
1023
precise_file_ids = set()
1024
changed_file_ids = []
1025
if specific_files == []:
1026
specific_file_ids = []
1028
specific_file_ids = self.target.paths2ids(specific_files,
1029
lookup_trees, require_versioned=require_versioned)
1030
if specific_files is not None:
1031
# reparented or added entries must have their parents included
1032
# so that valid deltas can be created. The seen_parents set
1033
# tracks the parents that we need to have.
1034
# The seen_dirs set tracks directory entries we've yielded.
1035
# After outputting version object in to_entries we set difference
1036
# the two seen sets and start checking parents.
1037
seen_parents = set()
1039
if want_unversioned:
1040
all_unversioned = sorted([(p.split('/'), p) for p in
1041
self.target.extras()
1042
if specific_files is None or
1043
osutils.is_inside_any(specific_files, p)])
1044
all_unversioned = deque(all_unversioned)
1046
all_unversioned = deque()
1048
from_entries_by_dir = list(self.source.iter_entries_by_dir(
1049
specific_file_ids=specific_file_ids))
1050
from_data = dict((e.file_id, (p, e)) for p, e in from_entries_by_dir)
1051
to_entries_by_dir = list(self.target.iter_entries_by_dir(
1052
specific_file_ids=specific_file_ids))
1053
num_entries = len(from_entries_by_dir) + len(to_entries_by_dir)
1055
# the unversioned path lookup only occurs on real trees - where there
1056
# can be extras. So the fake_entry is solely used to look up
1057
# executable it values when execute is not supported.
1058
fake_entry = InventoryFile('unused', 'unused', 'unused')
1059
for target_path, target_entry in to_entries_by_dir:
1060
while (all_unversioned and
1061
all_unversioned[0][0] < target_path.split('/')):
1062
unversioned_path = all_unversioned.popleft()
1063
target_kind, target_executable, target_stat = \
1064
self.target._comparison_data(fake_entry, unversioned_path[1])
1065
yield (None, (None, unversioned_path[1]), True, (False, False),
1067
(None, unversioned_path[0][-1]),
1068
(None, target_kind),
1069
(None, target_executable))
1070
source_path, source_entry = from_data.get(target_entry.file_id,
1072
result, changes = self._changes_from_entries(source_entry,
1073
target_entry, source_path=source_path, target_path=target_path)
1074
to_paths[result[0]] = result[1][1]
1079
pb.update('comparing files', entry_count, num_entries)
1080
if changes or include_unchanged:
1081
if specific_file_ids is not None:
1082
new_parent_id = result[4][1]
1083
precise_file_ids.add(new_parent_id)
1084
changed_file_ids.append(result[0])
1086
# Ensure correct behaviour for reparented/added specific files.
1087
if specific_files is not None:
1088
# Record output dirs
1089
if result[6][1] == 'directory':
1090
seen_dirs.add(result[0])
1091
# Record parents of reparented/added entries.
1092
versioned = result[3]
1094
if not versioned[0] or parents[0] != parents[1]:
1095
seen_parents.add(parents[1])
1096
while all_unversioned:
1097
# yield any trailing unversioned paths
1098
unversioned_path = all_unversioned.popleft()
1099
to_kind, to_executable, to_stat = \
1100
self.target._comparison_data(fake_entry, unversioned_path[1])
1101
yield (None, (None, unversioned_path[1]), True, (False, False),
1103
(None, unversioned_path[0][-1]),
1105
(None, to_executable))
1106
# Yield all remaining source paths
1107
for path, from_entry in from_entries_by_dir:
1108
file_id = from_entry.file_id
1109
if file_id in to_paths:
1112
if file_id not in self.target.all_file_ids():
1113
# common case - paths we have not emitted are not present in
1117
to_path = self.target.id2path(file_id)
1120
pb.update('comparing files', entry_count, num_entries)
1121
versioned = (True, False)
1122
parent = (from_entry.parent_id, None)
1123
name = (from_entry.name, None)
1124
from_kind, from_executable, stat_value = \
1125
self.source._comparison_data(from_entry, path)
1126
kind = (from_kind, None)
1127
executable = (from_executable, None)
1128
changed_content = from_kind is not None
1129
# the parent's path is necessarily known at this point.
1130
changed_file_ids.append(file_id)
1131
yield(file_id, (path, to_path), changed_content, versioned, parent,
1132
name, kind, executable)
1133
changed_file_ids = set(changed_file_ids)
1134
if specific_file_ids is not None:
1135
for result in self._handle_precise_ids(precise_file_ids,
1139
def _get_entry(self, tree, file_id):
1140
"""Get an inventory entry from a tree, with missing entries as None.
1142
If the tree raises NotImplementedError on accessing .inventory, then
1143
this is worked around using iter_entries_by_dir on just the file id
1146
:param tree: The tree to lookup the entry in.
1147
:param file_id: The file_id to lookup.
1150
inventory = tree.inventory
1151
except NotImplementedError:
1152
# No inventory available.
1154
iterator = tree.iter_entries_by_dir(specific_file_ids=[file_id])
1155
return iterator.next()[1]
1156
except StopIteration:
1160
return inventory[file_id]
1161
except errors.NoSuchId:
1164
def _handle_precise_ids(self, precise_file_ids, changed_file_ids,
1165
discarded_changes=None):
1166
"""Fill out a partial iter_changes to be consistent.
1168
:param precise_file_ids: The file ids of parents that were seen during
1170
:param changed_file_ids: The file ids of already emitted items.
1171
:param discarded_changes: An optional dict of precalculated
1172
iter_changes items which the partial iter_changes had not output
1174
:return: A generator of iter_changes items to output.
1176
# process parents of things that had changed under the users
1177
# requested paths to prevent incorrect paths or parent ids which
1178
# aren't in the tree.
1179
while precise_file_ids:
1180
precise_file_ids.discard(None)
1181
# Don't emit file_ids twice
1182
precise_file_ids.difference_update(changed_file_ids)
1183
if not precise_file_ids:
1185
# If the there was something at a given output path in source, we
1186
# have to include the entry from source in the delta, or we would
1187
# be putting this entry into a used path.
1189
for parent_id in precise_file_ids:
1191
paths.append(self.target.id2path(parent_id))
1192
except errors.NoSuchId:
1193
# This id has been dragged in from the source by delta
1194
# expansion and isn't present in target at all: we don't
1195
# need to check for path collisions on it.
1198
old_id = self.source.path2id(path)
1199
precise_file_ids.add(old_id)
1200
precise_file_ids.discard(None)
1201
current_ids = precise_file_ids
1202
precise_file_ids = set()
1203
# We have to emit all of precise_file_ids that have been altered.
1204
# We may have to output the children of some of those ids if any
1205
# directories have stopped being directories.
1206
for file_id in current_ids:
1208
if discarded_changes:
1209
result = discarded_changes.get(file_id)
1214
old_entry = self._get_entry(self.source, file_id)
1215
new_entry = self._get_entry(self.target, file_id)
1216
result, changes = self._changes_from_entries(
1217
old_entry, new_entry)
1220
# Get this parents parent to examine.
1221
new_parent_id = result[4][1]
1222
precise_file_ids.add(new_parent_id)
1224
if (result[6][0] == 'directory' and
1225
result[6][1] != 'directory'):
1226
# This stopped being a directory, the old children have
1228
if old_entry is None:
1229
# Reusing a discarded change.
1230
old_entry = self._get_entry(self.source, file_id)
1231
for child in old_entry.children.values():
1232
precise_file_ids.add(child.file_id)
1233
changed_file_ids.add(result[0])
1237
class MultiWalker(object):
1238
"""Walk multiple trees simultaneously, getting combined results."""
1240
# Note: This could be written to not assume you can do out-of-order
1241
# lookups. Instead any nodes that don't match in all trees could be
1242
# marked as 'deferred', and then returned in the final cleanup loop.
1243
# For now, I think it is "nicer" to return things as close to the
1244
# "master_tree" order as we can.
1246
def __init__(self, master_tree, other_trees):
1247
"""Create a new MultiWalker.
1249
All trees being walked must implement "iter_entries_by_dir()", such
1250
that they yield (path, object) tuples, where that object will have a
1251
'.file_id' member, that can be used to check equality.
1253
:param master_tree: All trees will be 'slaved' to the master_tree such
1254
that nodes in master_tree will be used as 'first-pass' sync points.
1255
Any nodes that aren't in master_tree will be merged in a second
1257
:param other_trees: A list of other trees to walk simultaneously.
1259
self._master_tree = master_tree
1260
self._other_trees = other_trees
1262
# Keep track of any nodes that were properly processed just out of
1263
# order, that way we don't return them at the end, we don't have to
1264
# track *all* processed file_ids, just the out-of-order ones
1265
self._out_of_order_processed = set()
1268
def _step_one(iterator):
1269
"""Step an iter_entries_by_dir iterator.
1271
:return: (has_more, path, ie)
1272
If has_more is False, path and ie will be None.
1275
path, ie = iterator.next()
1276
except StopIteration:
1277
return False, None, None
1279
return True, path, ie
1282
def _cmp_path_by_dirblock(path1, path2):
1283
"""Compare two paths based on what directory they are in.
1285
This generates a sort order, such that all children of a directory are
1286
sorted together, and grandchildren are in the same order as the
1287
children appear. But all grandchildren come after all children.
1289
:param path1: first path
1290
:param path2: the second path
1291
:return: negative number if ``path1`` comes first,
1292
0 if paths are equal
1293
and a positive number if ``path2`` sorts first
1295
# Shortcut this special case
1298
# This is stolen from _dirstate_helpers_py.py, only switching it to
1299
# Unicode objects. Consider using encode_utf8() and then using the
1300
# optimized versions, or maybe writing optimized unicode versions.
1301
if not isinstance(path1, unicode):
1302
raise TypeError("'path1' must be a unicode string, not %s: %r"
1303
% (type(path1), path1))
1304
if not isinstance(path2, unicode):
1305
raise TypeError("'path2' must be a unicode string, not %s: %r"
1306
% (type(path2), path2))
1307
return cmp(MultiWalker._path_to_key(path1),
1308
MultiWalker._path_to_key(path2))
1311
def _path_to_key(path):
1312
dirname, basename = osutils.split(path)
1313
return (dirname.split(u'/'), basename)
1315
def _lookup_by_file_id(self, extra_entries, other_tree, file_id):
1316
"""Lookup an inventory entry by file_id.
1318
This is called when an entry is missing in the normal order.
1319
Generally this is because a file was either renamed, or it was
1320
deleted/added. If the entry was found in the inventory and not in
1321
extra_entries, it will be added to self._out_of_order_processed
1323
:param extra_entries: A dictionary of {file_id: (path, ie)}. This
1324
should be filled with entries that were found before they were
1325
used. If file_id is present, it will be removed from the
1327
:param other_tree: The Tree to search, in case we didn't find the entry
1329
:param file_id: The file_id to look for
1330
:return: (path, ie) if found or (None, None) if not present.
1332
if file_id in extra_entries:
1333
return extra_entries.pop(file_id)
1334
# TODO: Is id2path better as the first call, or is
1335
# inventory[file_id] better as a first check?
1337
cur_path = other_tree.id2path(file_id)
1338
except errors.NoSuchId:
1340
if cur_path is None:
1343
self._out_of_order_processed.add(file_id)
1344
cur_ie = other_tree.inventory[file_id]
1345
return (cur_path, cur_ie)
1348
"""Match up the values in the different trees."""
1349
for result in self._walk_master_tree():
1351
self._finish_others()
1352
for result in self._walk_others():
1355
def _walk_master_tree(self):
1356
"""First pass, walk all trees in lock-step.
1358
When we are done, all nodes in the master_tree will have been
1359
processed. _other_walkers, _other_entries, and _others_extra will be
1360
set on 'self' for future processing.
1362
# This iterator has the most "inlining" done, because it tends to touch
1363
# every file in the tree, while the others only hit nodes that don't
1365
master_iterator = self._master_tree.iter_entries_by_dir()
1367
other_walkers = [other.iter_entries_by_dir()
1368
for other in self._other_trees]
1369
other_entries = [self._step_one(walker) for walker in other_walkers]
1370
# Track extra nodes in the other trees
1371
others_extra = [{} for i in xrange(len(self._other_trees))]
1373
master_has_more = True
1374
step_one = self._step_one
1375
lookup_by_file_id = self._lookup_by_file_id
1376
out_of_order_processed = self._out_of_order_processed
1378
while master_has_more:
1379
(master_has_more, path, master_ie) = step_one(master_iterator)
1380
if not master_has_more:
1383
file_id = master_ie.file_id
1385
other_values_append = other_values.append
1386
next_other_entries = []
1387
next_other_entries_append = next_other_entries.append
1388
for idx, (other_has_more, other_path, other_ie) in enumerate(other_entries):
1389
if not other_has_more:
1390
other_values_append(lookup_by_file_id(
1391
others_extra[idx], self._other_trees[idx], file_id))
1392
next_other_entries_append((False, None, None))
1393
elif file_id == other_ie.file_id:
1394
# This is the critical code path, as most of the entries
1395
# should match between most trees.
1396
other_values_append((other_path, other_ie))
1397
next_other_entries_append(step_one(other_walkers[idx]))
1399
# This walker did not match, step it until it either
1400
# matches, or we know we are past the current walker.
1401
other_walker = other_walkers[idx]
1402
other_extra = others_extra[idx]
1403
while (other_has_more and
1404
self._cmp_path_by_dirblock(other_path, path) < 0):
1405
other_file_id = other_ie.file_id
1406
if other_file_id not in out_of_order_processed:
1407
other_extra[other_file_id] = (other_path, other_ie)
1408
other_has_more, other_path, other_ie = \
1409
step_one(other_walker)
1410
if other_has_more and other_ie.file_id == file_id:
1411
# We ended up walking to this point, match and step
1413
other_values_append((other_path, other_ie))
1414
other_has_more, other_path, other_ie = \
1415
step_one(other_walker)
1417
# This record isn't in the normal order, see if it
1419
other_values_append(lookup_by_file_id(
1420
other_extra, self._other_trees[idx], file_id))
1421
next_other_entries_append((other_has_more, other_path,
1423
other_entries = next_other_entries
1425
# We've matched all the walkers, yield this datapoint
1426
yield path, file_id, master_ie, other_values
1427
self._other_walkers = other_walkers
1428
self._other_entries = other_entries
1429
self._others_extra = others_extra
1431
def _finish_others(self):
1432
"""Finish walking the other iterators, so we get all entries."""
1433
for idx, info in enumerate(self._other_entries):
1434
other_extra = self._others_extra[idx]
1435
(other_has_more, other_path, other_ie) = info
1436
while other_has_more:
1437
other_file_id = other_ie.file_id
1438
if other_file_id not in self._out_of_order_processed:
1439
other_extra[other_file_id] = (other_path, other_ie)
1440
other_has_more, other_path, other_ie = \
1441
self._step_one(self._other_walkers[idx])
1442
del self._other_entries
1444
def _walk_others(self):
1445
"""Finish up by walking all the 'deferred' nodes."""
1446
# TODO: One alternative would be to grab all possible unprocessed
1447
# file_ids, and then sort by path, and then yield them. That
1448
# might ensure better ordering, in case a caller strictly
1449
# requires parents before children.
1450
for idx, other_extra in enumerate(self._others_extra):
1451
others = sorted(other_extra.itervalues(),
1452
key=lambda x: self._path_to_key(x[0]))
1453
for other_path, other_ie in others:
1454
file_id = other_ie.file_id
1455
# We don't need to check out_of_order_processed here, because
1456
# the lookup_by_file_id will be removing anything processed
1457
# from the extras cache
1458
other_extra.pop(file_id)
1459
other_values = [(None, None) for i in xrange(idx)]
1460
other_values.append((other_path, other_ie))
1461
for alt_idx, alt_extra in enumerate(self._others_extra[idx+1:]):
1462
alt_idx = alt_idx + idx + 1
1463
alt_extra = self._others_extra[alt_idx]
1464
alt_tree = self._other_trees[alt_idx]
1465
other_values.append(self._lookup_by_file_id(
1466
alt_extra, alt_tree, file_id))
1467
yield other_path, file_id, None, other_values