301
304
:param file_id: The file_id of the file.
302
305
:param path: The path of the file.
303
307
If both file_id and path are supplied, an implementation may use
306
310
return osutils.split_lines(self.get_file_text(file_id, path))
312
def get_file_verifier(self, file_id, path=None, stat_value=None):
313
"""Return a verifier for a file.
315
The default implementation returns a sha1.
317
:param file_id: The handle for this file.
318
:param path: The path that this file can be found at.
319
These must point to the same object.
320
:param stat_value: Optional stat value for the object
321
:return: Tuple with verifier name and verifier data
323
return ("SHA1", self.get_file_sha1(file_id, path=path,
324
stat_value=stat_value))
326
def get_file_sha1(self, file_id, path=None, stat_value=None):
327
"""Return the SHA1 file for a file.
329
:note: callers should use get_file_verifier instead
330
where possible, as the underlying repository implementation may
331
have quicker access to a non-sha1 verifier.
333
:param file_id: The handle for this file.
334
:param path: The path that this file can be found at.
335
These must point to the same object.
336
:param stat_value: Optional stat value for the object
338
raise NotImplementedError(self.get_file_sha1)
308
340
def get_file_mtime(self, file_id, path=None):
309
341
"""Return the modification time for a file.
352
390
cur_file = (self.get_file_text(file_id),)
353
391
yield identifier, cur_file
355
def get_symlink_target(self, file_id):
393
def get_symlink_target(self, file_id, path=None):
356
394
"""Get the target for a given file_id.
358
396
It is assumed that the caller already knows that file_id is referencing
360
398
:param file_id: Handle for the symlink entry.
399
:param path: The path of the file.
400
If both file_id and path are supplied, an implementation may use
361
402
:return: The path the symlink points to.
363
404
raise NotImplementedError(self.get_symlink_target)
365
def get_canonical_inventory_paths(self, paths):
366
"""Like get_canonical_inventory_path() but works on multiple items.
368
:param paths: A sequence of paths relative to the root of the tree.
369
:return: A list of paths, with each item the corresponding input path
370
adjusted to account for existing elements that match case
373
return list(self._yield_canonical_inventory_paths(paths))
375
def get_canonical_inventory_path(self, path):
376
"""Returns the first inventory item that case-insensitively matches path.
378
If a path matches exactly, it is returned. If no path matches exactly
379
but more than one path matches case-insensitively, it is implementation
380
defined which is returned.
382
If no path matches case-insensitively, the input path is returned, but
383
with as many path entries that do exist changed to their canonical
386
If you need to resolve many names from the same tree, you should
387
use get_canonical_inventory_paths() to avoid O(N) behaviour.
389
:param path: A paths relative to the root of the tree.
390
:return: The input path adjusted to account for existing elements
391
that match case insensitively.
393
return self._yield_canonical_inventory_paths([path]).next()
395
def _yield_canonical_inventory_paths(self, paths):
397
# First, if the path as specified exists exactly, just use it.
398
if self.path2id(path) is not None:
402
cur_id = self.get_root_id()
404
bit_iter = iter(path.split("/"))
408
for child in self.iter_children(cur_id):
410
# XXX: it seem like if the child is known to be in the
411
# tree, we shouldn't need to go from its id back to
412
# its path -- mbp 2010-02-11
414
# XXX: it seems like we could be more efficient
415
# by just directly looking up the original name and
416
# only then searching all children; also by not
417
# chopping paths so much. -- mbp 2010-02-11
418
child_base = os.path.basename(self.id2path(child))
419
if (child_base == elt):
420
# if we found an exact match, we can stop now; if
421
# we found an approximate match we need to keep
422
# searching because there might be an exact match
425
new_path = osutils.pathjoin(cur_path, child_base)
427
elif child_base.lower() == lelt:
429
new_path = osutils.pathjoin(cur_path, child_base)
431
# before a change is committed we can see this error...
436
# got to the end of this directory and no entries matched.
437
# Return what matched so far, plus the rest as specified.
438
cur_path = osutils.pathjoin(cur_path, elt, *list(bit_iter))
443
406
def get_root_id(self):
444
407
"""Return the file_id for the root of this tree."""
445
408
raise NotImplementedError(self.get_root_id)
503
466
except errors.NoSuchRevisionInTree:
504
467
yield self.repository.revision_tree(revision_id)
507
def _file_revision(revision_tree, file_id):
508
"""Determine the revision associated with a file in a given tree."""
509
revision_tree.lock_read()
511
return revision_tree.inventory[file_id].revision
513
revision_tree.unlock()
515
469
def _get_file_revision(self, file_id, vf, tree_revision):
516
470
"""Ensure that file_id, tree_revision is in vf to plan the merge."""
518
472
if getattr(self, '_repository', None) is None:
519
473
last_revision = tree_revision
520
parent_keys = [(file_id, self._file_revision(t, file_id)) for t in
474
parent_keys = [(file_id, t.get_file_revision(file_id)) for t in
521
475
self._iter_parent_trees()]
522
476
vf.add_lines((file_id, last_revision), parent_keys,
523
477
self.get_file_lines(file_id))
524
478
repo = self.branch.repository
525
479
base_vf = repo.texts
527
last_revision = self._file_revision(self, file_id)
481
last_revision = self.get_file_revision(file_id)
528
482
base_vf = self._repository.texts
529
483
if base_vf not in vf.fallback_versionedfiles:
530
484
vf.fallback_versionedfiles.append(base_vf)
531
485
return last_revision
533
inventory = property(_get_inventory,
534
doc="Inventory of this Tree")
536
487
def _check_retrieved(self, ie, f):
537
488
if not __debug__:
539
fp = fingerprint_file(f)
490
fp = osutils.fingerprint_file(f)
542
493
if ie.text_size is not None:
543
494
if ie.text_size != fp['size']:
544
raise BzrError("mismatched size for file %r in %r" % (ie.file_id, self._store),
495
raise errors.BzrError(
496
"mismatched size for file %r in %r" %
497
(ie.file_id, self._store),
545
498
["inventory expects %d bytes" % ie.text_size,
546
499
"file is actually %d bytes" % fp['size'],
547
500
"store is probably damaged/corrupt"])
549
502
if ie.text_sha1 != fp['sha1']:
550
raise BzrError("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
503
raise errors.BzrError("wrong SHA-1 for file %r in %r" %
504
(ie.file_id, self._store),
551
505
["inventory expects %s" % ie.text_sha1,
552
506
"file is actually %s" % fp['sha1'],
553
507
"store is probably damaged/corrupt"])
556
509
def path2id(self, path):
557
510
"""Return the id for path in this tree."""
558
return self._inventory.path2id(path)
511
raise NotImplementedError(self.path2id)
560
513
def paths2ids(self, paths, trees=[], require_versioned=True):
561
514
"""Return all the ids that can be reached by walking from paths.
728
######################################################################
731
# TODO: Merge these two functions into a single one that can operate
732
# on either a whole tree or a set of files.
734
# TODO: Return the diff in order by filename, not by category or in
735
# random order. Can probably be done by lock-stepping through the
736
# filenames from both trees.
739
def file_status(filename, old_tree, new_tree):
740
"""Return single-letter status, old and new names for a file.
742
The complexity here is in deciding how to represent renames;
743
many complex cases are possible.
677
class InventoryTree(Tree):
678
"""A tree that relies on an inventory for its metadata.
680
Trees contain an `Inventory` object, and also know how to retrieve
681
file texts mentioned in the inventory, either from a working
682
directory or from a store.
684
It is possible for trees to contain files that are not described
685
in their inventory or vice versa; for this use `filenames()`.
687
Subclasses should set the _inventory attribute, which is considered
688
private to external API users.
745
old_inv = old_tree.inventory
746
new_inv = new_tree.inventory
747
new_id = new_inv.path2id(filename)
748
old_id = old_inv.path2id(filename)
750
if not new_id and not old_id:
751
# easy: doesn't exist in either; not versioned at all
752
if new_tree.is_ignored(filename):
753
return 'I', None, None
755
return '?', None, None
757
# There is now a file of this name, great.
760
# There is no longer a file of this name, but we can describe
761
# what happened to the file that used to have
762
# this name. There are two possibilities: either it was
763
# deleted entirely, or renamed.
764
if new_inv.has_id(old_id):
765
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
767
return 'D', old_inv.id2path(old_id), None
769
# if the file_id is new in this revision, it is added
770
if new_id and not old_inv.has_id(new_id):
773
# if there used to be a file of this name, but that ID has now
774
# disappeared, it is deleted
775
if old_id and not new_inv.has_id(old_id):
781
@deprecated_function(deprecated_in((1, 9, 0)))
782
def find_renames(old_inv, new_inv):
783
for file_id in old_inv:
784
if file_id not in new_inv:
786
old_name = old_inv.id2path(file_id)
787
new_name = new_inv.id2path(file_id)
788
if old_name != new_name:
789
yield (old_name, new_name)
691
def get_canonical_inventory_paths(self, paths):
692
"""Like get_canonical_inventory_path() but works on multiple items.
694
:param paths: A sequence of paths relative to the root of the tree.
695
:return: A list of paths, with each item the corresponding input path
696
adjusted to account for existing elements that match case
699
return list(self._yield_canonical_inventory_paths(paths))
701
def get_canonical_inventory_path(self, path):
702
"""Returns the first inventory item that case-insensitively matches path.
704
If a path matches exactly, it is returned. If no path matches exactly
705
but more than one path matches case-insensitively, it is implementation
706
defined which is returned.
708
If no path matches case-insensitively, the input path is returned, but
709
with as many path entries that do exist changed to their canonical
712
If you need to resolve many names from the same tree, you should
713
use get_canonical_inventory_paths() to avoid O(N) behaviour.
715
:param path: A paths relative to the root of the tree.
716
:return: The input path adjusted to account for existing elements
717
that match case insensitively.
719
return self._yield_canonical_inventory_paths([path]).next()
721
def _yield_canonical_inventory_paths(self, paths):
723
# First, if the path as specified exists exactly, just use it.
724
if self.path2id(path) is not None:
728
cur_id = self.get_root_id()
730
bit_iter = iter(path.split("/"))
734
for child in self.iter_children(cur_id):
736
# XXX: it seem like if the child is known to be in the
737
# tree, we shouldn't need to go from its id back to
738
# its path -- mbp 2010-02-11
740
# XXX: it seems like we could be more efficient
741
# by just directly looking up the original name and
742
# only then searching all children; also by not
743
# chopping paths so much. -- mbp 2010-02-11
744
child_base = os.path.basename(self.id2path(child))
745
if (child_base == elt):
746
# if we found an exact match, we can stop now; if
747
# we found an approximate match we need to keep
748
# searching because there might be an exact match
751
new_path = osutils.pathjoin(cur_path, child_base)
753
elif child_base.lower() == lelt:
755
new_path = osutils.pathjoin(cur_path, child_base)
756
except errors.NoSuchId:
757
# before a change is committed we can see this error...
762
# got to the end of this directory and no entries matched.
763
# Return what matched so far, plus the rest as specified.
764
cur_path = osutils.pathjoin(cur_path, elt, *list(bit_iter))
769
def _get_inventory(self):
770
return self._inventory
772
inventory = property(_get_inventory,
773
doc="Inventory of this Tree")
776
def path2id(self, path):
777
"""Return the id for path in this tree."""
778
return self._inventory.path2id(path)
780
def id2path(self, file_id):
781
"""Return the path for a file id.
785
return self.inventory.id2path(file_id)
787
def has_id(self, file_id):
788
return self.inventory.has_id(file_id)
790
def has_or_had_id(self, file_id):
791
return self.inventory.has_id(file_id)
793
def all_file_ids(self):
794
return set(self.inventory)
796
@deprecated_method(deprecated_in((2, 4, 0)))
798
return iter(self.inventory)
800
def filter_unversioned_files(self, paths):
801
"""Filter out paths that are versioned.
803
:return: set of paths.
805
# NB: we specifically *don't* call self.has_filename, because for
806
# WorkingTrees that can indicate files that exist on disk but that
808
pred = self.inventory.has_filename
809
return set((p for p in paths if not pred(p)))
812
def iter_entries_by_dir(self, specific_file_ids=None, yield_parents=False):
813
"""Walk the tree in 'by_dir' order.
815
This will yield each entry in the tree as a (path, entry) tuple.
816
The order that they are yielded is:
818
See Tree.iter_entries_by_dir for details.
820
:param yield_parents: If True, yield the parents from the root leading
821
down to specific_file_ids that have been requested. This has no
822
impact if specific_file_ids is None.
824
return self.inventory.iter_entries_by_dir(
825
specific_file_ids=specific_file_ids, yield_parents=yield_parents)
827
@deprecated_method(deprecated_in((2, 5, 0)))
828
def get_file_by_path(self, path):
829
return self.get_file(self.path2id(path), path)
792
832
def find_ids_across_trees(filenames, trees, require_versioned=True):
1254
1300
changed_file_ids.add(result[0])
1304
def file_content_matches(self, source_file_id, target_file_id,
1305
source_path=None, target_path=None, source_stat=None, target_stat=None):
1306
"""Check if two files are the same in the source and target trees.
1308
This only checks that the contents of the files are the same,
1309
it does not touch anything else.
1311
:param source_file_id: File id of the file in the source tree
1312
:param target_file_id: File id of the file in the target tree
1313
:param source_path: Path of the file in the source tree
1314
:param target_path: Path of the file in the target tree
1315
:param source_stat: Optional stat value of the file in the source tree
1316
:param target_stat: Optional stat value of the file in the target tree
1317
:return: Boolean indicating whether the files have the same contents
1319
source_verifier_kind, source_verifier_data = self.source.get_file_verifier(
1320
source_file_id, source_path, source_stat)
1321
target_verifier_kind, target_verifier_data = self.target.get_file_verifier(
1322
target_file_id, target_path, target_stat)
1323
if source_verifier_kind == target_verifier_kind:
1324
return (source_verifier_data == target_verifier_data)
1325
# Fall back to SHA1 for now
1326
if source_verifier_kind != "SHA1":
1327
source_sha1 = self.source.get_file_sha1(source_file_id,
1328
source_path, source_stat)
1330
source_sha1 = source_verifier_data
1331
if target_verifier_kind != "SHA1":
1332
target_sha1 = self.target.get_file_sha1(target_file_id,
1333
target_path, target_stat)
1335
target_sha1 = target_verifier_data
1336
return (source_sha1 == target_sha1)
1338
InterTree.register_optimiser(InterTree)
1258
1341
class MultiWalker(object):
1259
1342
"""Walk multiple trees simultaneously, getting combined results."""
added
removed
bzrlib/tree.py
