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):
81
fp = fingerprint_file(f)
84
if ie.text_size != None:
85
if ie.text_size != fp['size']:
86
raise BzrError("mismatched size for file %r in %r" % (ie.file_id, self._store),
87
["inventory expects %d bytes" % ie.text_size,
88
"file is actually %d bytes" % fp['size'],
89
"store is probably damaged/corrupt"])
91
if ie.text_sha1 != fp['sha1']:
92
raise BzrError("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
93
["inventory expects %s" % ie.text_sha1,
94
"file is actually %s" % fp['sha1'],
95
"store is probably damaged/corrupt"])
98
def print_file(self, file_id):
99
"""Print file with id `file_id` to stdout."""
101
sys.stdout.write(self.get_file_text(file_id))
104
def export(self, dest, format='dir', root=None):
105
"""Export this tree."""
107
exporter = exporters[format]
109
from bzrlib.errors import BzrCommandError
110
raise BzrCommandError("export format %r not supported" % format)
111
exporter(self, dest, root)
115
class RevisionTree(Tree):
116
"""Tree viewing a previous revision.
118
File text can be retrieved from the text store.
120
TODO: Some kind of `__repr__` method, but a good one
121
probably means knowing the branch and revision number,
122
or at least passing a description to the constructor.
125
def __init__(self, weave_store, inv, revision_id):
126
self._weave_store = weave_store
127
self._inventory = inv
128
self._revision_id = revision_id
130
def get_weave(self, file_id):
131
return self._weave_store.get_weave(file_id)
134
def get_file_lines(self, file_id):
135
ie = self._inventory[file_id]
136
weave = self.get_weave(file_id)
137
return weave.get(ie.revision)
140
def get_file_text(self, file_id):
141
return ''.join(self.get_file_lines(file_id))
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
163
def get_symlink_target(self, file_id):
164
ie = self._inventory[file_id]
165
return ie.symlink_target;
167
class EmptyTree(Tree):
169
self._inventory = Inventory()
171
def get_symlink_target(self, file_id):
174
def has_filename(self, filename):
177
def list_files(self):
178
if False: # just to make it a generator
181
def __contains__(self, file_id):
182
return file_id in self._inventory
184
def get_file_sha1(self, file_id):
185
assert self._inventory[file_id].kind == "root_directory"
191
######################################################################
194
# TODO: Merge these two functions into a single one that can operate
195
# on either a whole tree or a set of files.
197
# TODO: Return the diff in order by filename, not by category or in
198
# random order. Can probably be done by lock-stepping through the
199
# filenames from both trees.
202
def file_status(filename, old_tree, new_tree):
203
"""Return single-letter status, old and new names for a file.
205
The complexity here is in deciding how to represent renames;
206
many complex cases are possible.
208
old_inv = old_tree.inventory
209
new_inv = new_tree.inventory
210
new_id = new_inv.path2id(filename)
211
old_id = old_inv.path2id(filename)
213
if not new_id and not old_id:
214
# easy: doesn't exist in either; not versioned at all
215
if new_tree.is_ignored(filename):
216
return 'I', None, None
218
return '?', None, None
220
# There is now a file of this name, great.
223
# There is no longer a file of this name, but we can describe
224
# what happened to the file that used to have
225
# this name. There are two possibilities: either it was
226
# deleted entirely, or renamed.
228
if new_inv.has_id(old_id):
229
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
231
return 'D', old_inv.id2path(old_id), None
233
# if the file_id is new in this revision, it is added
234
if new_id and not old_inv.has_id(new_id):
237
# if there used to be a file of this name, but that ID has now
238
# disappeared, it is deleted
239
if old_id and not new_inv.has_id(old_id):
246
def find_renames(old_inv, new_inv):
247
for file_id in old_inv:
248
if file_id not in new_inv:
250
old_name = old_inv.id2path(file_id)
251
new_name = new_inv.id2path(file_id)
252
if old_name != new_name:
253
yield (old_name, new_name)
257
######################################################################
260
def dir_exporter(tree, dest, root):
261
"""Export this tree to a new directory.
263
`dest` should not exist, and will be created holding the
264
contents of this tree.
266
TODO: To handle subdirectories we need to create the
269
:note: If the export fails, the destination directory will be
270
left in a half-assed state.
274
mutter('export version %r' % tree)
276
for dp, ie in inv.iter_entries():
278
fullpath = appendpath(dest, dp)
279
if kind == 'directory':
282
pumpfile(tree.get_file(ie.file_id), file(fullpath, 'wb'))
283
elif kind == 'symlink':
285
os.symlink(ie.symlink_target, fullpath)
287
raise BzrError("Failed to create symlink %r -> %r, error: %s" % (fullpath, ie.symlink_target, e))
289
raise BzrError("don't know how to export {%s} of kind %r" % (ie.file_id, kind))
290
mutter(" export {%s} kind %s to %s" % (ie.file_id, kind, fullpath))
291
exporters['dir'] = dir_exporter
298
def get_root_name(dest):
299
"""Get just the root name for a tarball.
301
>>> get_root_name('mytar.tar')
303
>>> get_root_name('mytar.tar.bz2')
305
>>> get_root_name('tar.tar.tar.tgz')
307
>>> get_root_name('bzr-0.0.5.tar.gz')
309
>>> get_root_name('a/long/path/mytar.tgz')
311
>>> get_root_name('../parent/../dir/other.tbz2')
314
endings = ['.tar', '.tar.gz', '.tgz', '.tar.bz2', '.tbz2']
315
dest = os.path.basename(dest)
317
if dest.endswith(end):
318
return dest[:-len(end)]
320
def tar_exporter(tree, dest, root, compression=None):
321
"""Export this tree to a new tar file.
323
`dest` will be created holding the contents of this tree; if it
324
already exists, it will be clobbered, like with "tar -c".
326
from time import time
328
compression = str(compression or '')
330
root = get_root_name(dest)
332
ball = tarfile.open(dest, 'w:' + compression)
333
except tarfile.CompressionError, e:
334
raise BzrError(str(e))
335
mutter('export version %r' % tree)
337
for dp, ie in inv.iter_entries():
338
mutter(" export {%s} kind %s to %s" % (ie.file_id, ie.kind, dest))
339
item = tarfile.TarInfo(os.path.join(root, dp))
340
# TODO: would be cool to actually set it to the timestamp of the
341
# revision it was last changed
343
if ie.kind == 'directory':
344
item.type = tarfile.DIRTYPE
349
elif ie.kind == 'file':
350
item.type = tarfile.REGTYPE
351
fileobj = tree.get_file(ie.file_id)
352
item.size = _find_file_size(fileobj)
355
raise BzrError("don't know how to export {%s} of kind %r" %
356
(ie.file_id, ie.kind))
358
ball.addfile(item, fileobj)
360
exporters['tar'] = tar_exporter
362
def tgz_exporter(tree, dest, root):
363
tar_exporter(tree, dest, root, compression='gz')
364
exporters['tgz'] = tgz_exporter
366
def tbz_exporter(tree, dest, root):
367
tar_exporter(tree, dest, root, compression='bz2')
368
exporters['tbz2'] = tbz_exporter
371
def _find_file_size(fileobj):
372
offset = fileobj.tell()
375
size = fileobj.tell()
377
# gzip doesn't accept second argument to seek()
381
nread = len(fileobj.read())