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
import os.path, os, fnmatch
23
from osutils import pumpfile, compare_files, filesize, quotefn, sha_file, \
24
joinpath, splitpath, appendpath, isdir, isfile, file_kind, fingerprint_file
26
from stat import S_ISREG, S_ISDIR, ST_MODE, ST_SIZE
28
from inventory import Inventory
29
from trace import mutter, note
30
from errors import bailout
36
"""Abstract file tree.
38
There are several subclasses:
40
* `WorkingTree` exists as files on disk editable by the user.
42
* `RevisionTree` is a tree as recorded at some point in the past.
46
Trees contain an `Inventory` object, and also know how to retrieve
47
file texts mentioned in the inventory, either from a working
48
directory or from a store.
50
It is possible for trees to contain files that are not described
51
in their inventory or vice versa; for this use `filenames()`.
53
Trees can be compared, etc, regardless of whether they are working
54
trees or versioned trees.
57
def has_filename(self, filename):
58
"""True if the tree has given filename."""
59
raise NotImplementedError()
61
def has_id(self, file_id):
62
return self.inventory.has_id(file_id)
65
"""Return set of all ids in this tree."""
66
return self.inventory.id_set()
68
def id2path(self, file_id):
69
return self.inventory.id2path(file_id)
71
def _get_inventory(self):
72
return self._inventory
74
inventory = property(_get_inventory,
75
doc="Inventory of this Tree")
77
def _check_retrieved(self, ie, f):
78
fp = fingerprint_file(f)
81
if ie.text_size is not None:
82
if ie.text_size != fp['size']:
83
bailout("mismatched size for file %r in %r" % (ie.file_id, self._store),
84
["inventory expects %d bytes" % ie.text_size,
85
"file is actually %d bytes" % fp['size'],
86
"store is probably damaged/corrupt"])
88
if ie.text_sha1 != fp['sha1']:
89
bailout("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
90
["inventory expects %s" % ie.text_sha1,
91
"file is actually %s" % fp['sha1'],
92
"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):
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):
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
380
######################################################################
383
# TODO: Merge these two functions into a single one that can operate
384
# on either a whole tree or a set of files.
386
# TODO: Return the diff in order by filename, not by category or in
387
# random order. Can probably be done by lock-stepping through the
388
# filenames from both trees.
391
def file_status(filename, old_tree, new_tree):
392
"""Return single-letter status, old and new names for a file.
394
The complexity here is in deciding how to represent renames;
395
many complex cases are possible.
397
old_inv = old_tree.inventory
398
new_inv = new_tree.inventory
399
new_id = new_inv.path2id(filename)
400
old_id = old_inv.path2id(filename)
402
if not new_id and not old_id:
403
# easy: doesn't exist in either; not versioned at all
404
if new_tree.is_ignored(filename):
405
return 'I', None, None
407
return '?', None, None
409
# There is now a file of this name, great.
412
# There is no longer a file of this name, but we can describe
413
# what happened to the file that used to have
414
# this name. There are two possibilities: either it was
415
# deleted entirely, or renamed.
417
if new_inv.has_id(old_id):
418
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
420
return 'D', old_inv.id2path(old_id), None
422
# if the file_id is new in this revision, it is added
423
if new_id and not old_inv.has_id(new_id):
426
# if there used to be a file of this name, but that ID has now
427
# disappeared, it is deleted
428
if old_id and not new_inv.has_id(old_id):
435
def find_renames(old_inv, new_inv):
436
for file_id in old_inv:
437
if file_id not in new_inv:
439
old_name = old_inv.id2path(file_id)
440
new_name = new_inv.id2path(file_id)
441
if old_name != new_name:
442
yield (old_name, new_name)