1
# Copyright (C) 2005 Canonical Ltd
3
# This program is free software; you can redistribute it and/or modify
4
# it under the terms of the GNU General Public License as published by
5
# the Free Software Foundation; either version 2 of the License, or
6
# (at your option) any later version.
8
# This program is distributed in the hope that it will be useful,
9
# but WITHOUT ANY WARRANTY; without even the implied warranty of
10
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11
# GNU General Public License for more details.
13
# You should have received a copy of the GNU General Public License
14
# along with this program; if not, write to the Free Software
15
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
17
"""Tree classes, representing directory at point in time.
21
from cStringIO import StringIO
24
from bzrlib.errors import BzrError, BzrCheckError
25
from bzrlib import errors
26
from bzrlib.inventory import Inventory
27
from bzrlib.osutils import fingerprint_file
28
import bzrlib.revision
29
from bzrlib.trace import mutter, note
32
"""Abstract file tree.
34
There are several subclasses:
36
* `WorkingTree` exists as files on disk editable by the user.
38
* `RevisionTree` is a tree as recorded at some point in the past.
42
Trees contain an `Inventory` object, and also know how to retrieve
43
file texts mentioned in the inventory, either from a working
44
directory or from a store.
46
It is possible for trees to contain files that are not described
47
in their inventory or vice versa; for this use `filenames()`.
49
Trees can be compared, etc, regardless of whether they are working
50
trees or versioned trees.
54
"""Get a list of the conflicts in the tree.
56
Each conflict is an instance of bzrlib.conflicts.Conflict.
60
def get_parent_ids(self):
61
"""Get the parent ids for this tree.
63
:return: a list of parent ids. [] is returned to indicate
64
a tree with no parents.
65
:raises: BzrError if the parents are not known.
67
raise NotImplementedError(self.get_parent_ids)
69
def has_filename(self, filename):
70
"""True if the tree has given filename."""
71
raise NotImplementedError()
73
def has_id(self, file_id):
74
return self.inventory.has_id(file_id)
76
def has_or_had_id(self, file_id):
77
if file_id == self.inventory.root.file_id:
79
return self.inventory.has_id(file_id)
84
return iter(self.inventory)
86
def id2path(self, file_id):
87
return self.inventory.id2path(file_id)
89
def kind(self, file_id):
90
raise NotImplementedError("subclasses must implement kind")
92
def _get_inventory(self):
93
return self._inventory
95
def get_file_by_path(self, path):
96
return self.get_file(self._inventory.path2id(path))
98
inventory = property(_get_inventory,
99
doc="Inventory of this Tree")
101
def _check_retrieved(self, ie, f):
104
fp = fingerprint_file(f)
107
if ie.text_size != None:
108
if ie.text_size != fp['size']:
109
raise BzrError("mismatched size for file %r in %r" % (ie.file_id, self._store),
110
["inventory expects %d bytes" % ie.text_size,
111
"file is actually %d bytes" % fp['size'],
112
"store is probably damaged/corrupt"])
114
if ie.text_sha1 != fp['sha1']:
115
raise BzrError("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
116
["inventory expects %s" % ie.text_sha1,
117
"file is actually %s" % fp['sha1'],
118
"store is probably damaged/corrupt"])
121
def print_file(self, file_id):
122
"""Print file with id `file_id` to stdout."""
124
sys.stdout.write(self.get_file_text(file_id))
130
"""What files are present in this tree and unknown.
132
:return: an iterator over the unknown files.
139
def filter_unversioned_files(self, paths):
140
"""Filter out paths that are not versioned.
142
:return: set of paths.
144
# NB: we specifically *don't* call self.has_filename, because for
145
# WorkingTrees that can indicate files that exist on disk but that
147
pred = self.inventory.has_filename
148
return set((p for p in paths if not pred(p)))
151
class RevisionTree(Tree):
152
"""Tree viewing a previous revision.
154
File text can be retrieved from the text store.
156
TODO: Some kind of `__repr__` method, but a good one
157
probably means knowing the branch and revision number,
158
or at least passing a description to the constructor.
161
def __init__(self, branch, inv, revision_id):
162
# for compatability the 'branch' parameter has not been renamed to
163
# repository at this point. However, we should change RevisionTree's
164
# construction to always be via Repository and not via direct
165
# construction - this will mean that we can change the constructor
166
# with much less chance of breaking client code.
167
self._repository = branch
168
self._weave_store = branch.weave_store
169
self._inventory = inv
170
self._revision_id = revision_id
172
def get_parent_ids(self):
173
"""See Tree.get_parent_ids.
175
A RevisionTree's parents match the revision graph.
177
parent_ids = self._repository.get_revision(self._revision_id).parent_ids
180
def get_revision_id(self):
181
"""Return the revision id associated with this tree."""
182
return self._revision_id
184
def get_weave(self, file_id):
185
return self._weave_store.get_weave(file_id,
186
self._repository.get_transaction())
188
def get_file_lines(self, file_id):
189
ie = self._inventory[file_id]
190
weave = self.get_weave(file_id)
191
return weave.get_lines(ie.revision)
193
def get_file_text(self, file_id):
194
return ''.join(self.get_file_lines(file_id))
196
def get_file(self, file_id):
197
return StringIO(self.get_file_text(file_id))
199
def get_file_size(self, file_id):
200
return self._inventory[file_id].text_size
202
def get_file_sha1(self, file_id, path=None):
203
ie = self._inventory[file_id]
204
if ie.kind == "file":
208
def get_file_mtime(self, file_id, path=None):
209
ie = self._inventory[file_id]
210
revision = self._repository.get_revision(ie.revision)
211
return revision.timestamp
213
def is_executable(self, file_id, path=None):
214
ie = self._inventory[file_id]
215
if ie.kind != "file":
217
return self._inventory[file_id].executable
219
def has_filename(self, filename):
220
return bool(self.inventory.path2id(filename))
222
def list_files(self):
223
# The only files returned by this are those from the version
224
for path, entry in self.inventory.iter_entries():
225
yield path, 'V', entry.kind, entry.file_id, entry
227
def get_symlink_target(self, file_id):
228
ie = self._inventory[file_id]
229
return ie.symlink_target;
231
def kind(self, file_id):
232
return self._inventory[file_id].kind
235
self._repository.lock_read()
238
self._repository.unlock()
241
class EmptyTree(Tree):
244
self._inventory = Inventory()
246
def get_parent_ids(self):
247
"""See Tree.get_parent_ids.
249
An EmptyTree always has NULL_REVISION as the only parent.
253
def get_symlink_target(self, file_id):
256
def has_filename(self, filename):
259
def kind(self, file_id):
260
assert self._inventory[file_id].kind == "root_directory"
261
return "root_directory"
263
def list_files(self):
266
def __contains__(self, file_id):
267
return file_id in self._inventory
269
def get_file_sha1(self, file_id, path=None):
270
assert self._inventory[file_id].kind == "root_directory"
274
######################################################################
277
# TODO: Merge these two functions into a single one that can operate
278
# on either a whole tree or a set of files.
280
# TODO: Return the diff in order by filename, not by category or in
281
# random order. Can probably be done by lock-stepping through the
282
# filenames from both trees.
285
def file_status(filename, old_tree, new_tree):
286
"""Return single-letter status, old and new names for a file.
288
The complexity here is in deciding how to represent renames;
289
many complex cases are possible.
291
old_inv = old_tree.inventory
292
new_inv = new_tree.inventory
293
new_id = new_inv.path2id(filename)
294
old_id = old_inv.path2id(filename)
296
if not new_id and not old_id:
297
# easy: doesn't exist in either; not versioned at all
298
if new_tree.is_ignored(filename):
299
return 'I', None, None
301
return '?', None, None
303
# There is now a file of this name, great.
306
# There is no longer a file of this name, but we can describe
307
# what happened to the file that used to have
308
# this name. There are two possibilities: either it was
309
# deleted entirely, or renamed.
311
if new_inv.has_id(old_id):
312
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
314
return 'D', old_inv.id2path(old_id), None
316
# if the file_id is new in this revision, it is added
317
if new_id and not old_inv.has_id(new_id):
320
# if there used to be a file of this name, but that ID has now
321
# disappeared, it is deleted
322
if old_id and not new_inv.has_id(old_id):
329
def find_renames(old_inv, new_inv):
330
for file_id in old_inv:
331
if file_id not in new_inv:
333
old_name = old_inv.id2path(file_id)
334
new_name = new_inv.id2path(file_id)
335
if old_name != new_name:
336
yield (old_name, new_name)
339
def find_ids_across_trees(filenames, trees, require_versioned=True):
340
"""Find the ids corresponding to specified filenames.
342
All matches in all trees will be used, and all children of matched
343
directories will be used.
345
:param filenames: The filenames to find file_ids for
346
:param trees: The trees to find file_ids within
347
:param require_versioned: if true, all specified filenames must occur in
349
:return: a set of file ids for the specified filenames and their children.
353
specified_ids = _find_filename_ids_across_trees(filenames, trees,
355
return _find_children_across_trees(specified_ids, trees)
358
def _find_filename_ids_across_trees(filenames, trees, require_versioned):
359
"""Find the ids corresponding to specified filenames.
361
All matches in all trees will be used.
363
:param filenames: The filenames to find file_ids for
364
:param trees: The trees to find file_ids within
365
:param require_versioned: if true, all specified filenames must occur in
367
:return: a set of file ids for the specified filenames
370
interesting_ids = set()
371
for tree_path in filenames:
374
file_id = tree.inventory.path2id(tree_path)
375
if file_id is not None:
376
interesting_ids.add(file_id)
379
not_versioned.append(tree_path)
380
if len(not_versioned) > 0 and require_versioned:
381
raise errors.PathsNotVersionedError(not_versioned)
382
return interesting_ids
385
def _find_children_across_trees(specified_ids, trees):
386
"""Return a set including specified ids and their children
388
All matches in all trees will be used.
390
:param trees: The trees to find file_ids within
391
:return: a set containing all specified ids and their children
393
interesting_ids = set(specified_ids)
394
pending = interesting_ids
395
# now handle children of interesting ids
396
# we loop so that we handle all children of each id in both trees
397
while len(pending) > 0:
399
for file_id in pending:
401
if file_id not in tree:
403
entry = tree.inventory[file_id]
404
for child in getattr(entry, 'children', {}).itervalues():
405
if child.file_id not in interesting_ids:
406
new_pending.add(child.file_id)
407
interesting_ids.update(new_pending)
408
pending = new_pending
409
return interesting_ids
added
removed
bzrlib/tree.py
