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.trace import mutter, note
25
from bzrlib.errors import BzrError, BzrCheckError
26
from bzrlib.inventory import Inventory
27
from bzrlib.osutils import pumpfile, appendpath, fingerprint_file
33
"""Abstract file tree.
35
There are several subclasses:
37
* `WorkingTree` exists as files on disk editable by the user.
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()`.
50
Trees can be compared, etc, regardless of whether they are working
51
trees or versioned trees.
54
def has_filename(self, filename):
55
"""True if the tree has given filename."""
56
raise NotImplementedError()
58
def has_id(self, file_id):
59
return self.inventory.has_id(file_id)
64
return iter(self.inventory)
66
def id2path(self, file_id):
67
return self.inventory.id2path(file_id)
69
def _get_inventory(self):
70
return self._inventory
72
def get_file_by_path(self, path):
73
return self.get_file(self._inventory.path2id(path))
75
inventory = property(_get_inventory,
76
doc="Inventory of this Tree")
78
def _check_retrieved(self, ie, f):
79
fp = fingerprint_file(f)
82
if ie.text_size != None:
83
if ie.text_size != fp['size']:
84
raise BzrError("mismatched size for file %r in %r" % (ie.file_id, self._store),
85
["inventory expects %d bytes" % ie.text_size,
86
"file is actually %d bytes" % fp['size'],
87
"store is probably damaged/corrupt"])
89
if ie.text_sha1 != fp['sha1']:
90
raise BzrError("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
91
["inventory expects %s" % ie.text_sha1,
92
"file is actually %s" % fp['sha1'],
93
"store is probably damaged/corrupt"])
96
def print_file(self, file_id):
97
"""Print file with id `file_id` to stdout."""
99
sys.stdout.write(self.get_file_text(file_id))
102
def export(self, dest, format='dir', root=None):
103
"""Export this tree."""
105
exporter = exporters[format]
107
from bzrlib.errors import BzrCommandError
108
raise BzrCommandError("export format %r not supported" % format)
109
exporter(self, dest, root)
113
class RevisionTree(Tree):
114
"""Tree viewing a previous revision.
116
File text can be retrieved from the text store.
118
TODO: Some kind of `__repr__` method, but a good one
119
probably means knowing the branch and revision number,
120
or at least passing a description to the constructor.
123
def __init__(self, weave_store, inv, revision_id):
124
self._weave_store = weave_store
125
self._inventory = inv
126
self._revision_id = revision_id
128
def get_weave(self, file_id):
129
return self._weave_store.get_weave(file_id)
132
def get_file_text(self, file_id):
133
ie = self._inventory[file_id]
134
weave = self.get_weave(file_id)
135
idx = weave.lookup(ie.text_version)
136
content = weave.get_text(idx)
137
if len(content) != ie.text_size:
138
raise BzrCheckError('mismatched size on revision %s of file %s: '
140
% (self._revision_id, file_id, len(content),
144
def get_file(self, file_id):
145
return StringIO(self.get_file_text(file_id))
147
def get_file_size(self, file_id):
148
return self._inventory[file_id].text_size
150
def get_file_sha1(self, file_id):
151
ie = self._inventory[file_id]
152
if ie.kind == "file":
155
def has_filename(self, filename):
156
return bool(self.inventory.path2id(filename))
158
def list_files(self):
159
# The only files returned by this are those from the version
160
for path, entry in self.inventory.iter_entries():
161
yield path, 'V', entry.kind, entry.file_id
164
class EmptyTree(Tree):
166
self._inventory = Inventory()
168
def has_filename(self, filename):
171
def list_files(self):
172
if False: # just to make it a generator
175
def __contains__(self, file_id):
176
return file_id in self._inventory
178
def get_file_sha1(self, file_id):
179
assert self._inventory[file_id].kind == "root_directory"
185
######################################################################
188
# TODO: Merge these two functions into a single one that can operate
189
# on either a whole tree or a set of files.
191
# TODO: Return the diff in order by filename, not by category or in
192
# random order. Can probably be done by lock-stepping through the
193
# filenames from both trees.
196
def file_status(filename, old_tree, new_tree):
197
"""Return single-letter status, old and new names for a file.
199
The complexity here is in deciding how to represent renames;
200
many complex cases are possible.
202
old_inv = old_tree.inventory
203
new_inv = new_tree.inventory
204
new_id = new_inv.path2id(filename)
205
old_id = old_inv.path2id(filename)
207
if not new_id and not old_id:
208
# easy: doesn't exist in either; not versioned at all
209
if new_tree.is_ignored(filename):
210
return 'I', None, None
212
return '?', None, None
214
# There is now a file of this name, great.
217
# There is no longer a file of this name, but we can describe
218
# what happened to the file that used to have
219
# this name. There are two possibilities: either it was
220
# deleted entirely, or renamed.
222
if new_inv.has_id(old_id):
223
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
225
return 'D', old_inv.id2path(old_id), None
227
# if the file_id is new in this revision, it is added
228
if new_id and not old_inv.has_id(new_id):
231
# if there used to be a file of this name, but that ID has now
232
# disappeared, it is deleted
233
if old_id and not new_inv.has_id(old_id):
240
def find_renames(old_inv, new_inv):
241
for file_id in old_inv:
242
if file_id not in new_inv:
244
old_name = old_inv.id2path(file_id)
245
new_name = new_inv.id2path(file_id)
246
if old_name != new_name:
247
yield (old_name, new_name)
251
######################################################################
254
def dir_exporter(tree, dest, root):
255
"""Export this tree to a new directory.
257
`dest` should not exist, and will be created holding the
258
contents of this tree.
260
TODO: To handle subdirectories we need to create the
263
:note: If the export fails, the destination directory will be
264
left in a half-assed state.
268
mutter('export version %r' % tree)
270
for dp, ie in inv.iter_entries():
272
fullpath = appendpath(dest, dp)
273
if kind == 'directory':
276
pumpfile(tree.get_file(ie.file_id), file(fullpath, 'wb'))
278
raise BzrError("don't know how to export {%s} of kind %r" % (ie.file_id, kind))
279
mutter(" export {%s} kind %s to %s" % (ie.file_id, kind, fullpath))
280
exporters['dir'] = dir_exporter
287
def get_root_name(dest):
288
"""Get just the root name for a tarball.
290
>>> get_root_name('mytar.tar')
292
>>> get_root_name('mytar.tar.bz2')
294
>>> get_root_name('tar.tar.tar.tgz')
296
>>> get_root_name('bzr-0.0.5.tar.gz')
298
>>> get_root_name('a/long/path/mytar.tgz')
300
>>> get_root_name('../parent/../dir/other.tbz2')
303
endings = ['.tar', '.tar.gz', '.tgz', '.tar.bz2', '.tbz2']
304
dest = os.path.basename(dest)
306
if dest.endswith(end):
307
return dest[:-len(end)]
309
def tar_exporter(tree, dest, root, compression=None):
310
"""Export this tree to a new tar file.
312
`dest` will be created holding the contents of this tree; if it
313
already exists, it will be clobbered, like with "tar -c".
315
from time import time
317
compression = str(compression or '')
319
root = get_root_name(dest)
321
ball = tarfile.open(dest, 'w:' + compression)
322
except tarfile.CompressionError, e:
323
raise BzrError(str(e))
324
mutter('export version %r' % tree)
326
for dp, ie in inv.iter_entries():
327
mutter(" export {%s} kind %s to %s" % (ie.file_id, ie.kind, dest))
328
item = tarfile.TarInfo(os.path.join(root, dp))
329
# TODO: would be cool to actually set it to the timestamp of the
330
# revision it was last changed
332
if ie.kind == 'directory':
333
item.type = tarfile.DIRTYPE
338
elif ie.kind == 'file':
339
item.type = tarfile.REGTYPE
340
fileobj = tree.get_file(ie.file_id)
341
item.size = _find_file_size(fileobj)
344
raise BzrError("don't know how to export {%s} of kind %r" %
345
(ie.file_id, ie.kind))
347
ball.addfile(item, fileobj)
349
exporters['tar'] = tar_exporter
351
def tgz_exporter(tree, dest, root):
352
tar_exporter(tree, dest, root, compression='gz')
353
exporters['tgz'] = tgz_exporter
355
def tbz_exporter(tree, dest, root):
356
tar_exporter(tree, dest, root, compression='bz2')
357
exporters['tbz2'] = tbz_exporter
360
def _find_file_size(fileobj):
361
offset = fileobj.tell()
364
size = fileobj.tell()
366
# gzip doesn't accept second argument to seek()
370
nread = len(fileobj.read())
added
removed
bzrlib/tree.py
