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 inventory import Inventory
24
from trace import mutter, note
25
from osutils import pumpfile, compare_files, filesize, quotefn, sha_file, \
26
joinpath, splitpath, appendpath, isdir, isfile, file_kind, fingerprint_file
27
from errors import bailout
29
from stat import S_ISREG, S_ISDIR, ST_MODE, ST_SIZE
34
"""Abstract file tree.
36
There are several subclasses:
38
* `WorkingTree` exists as files on disk editable by the user.
40
* `RevisionTree` is a tree as recorded at some point in the past.
44
Trees contain an `Inventory` object, and also know how to retrieve
45
file texts mentioned in the inventory, either from a working
46
directory or from a store.
48
It is possible for trees to contain files that are not described
49
in their inventory or vice versa; for this use `filenames()`.
51
Trees can be compared, etc, regardless of whether they are working
52
trees or versioned trees.
55
def has_filename(self, filename):
56
"""True if the tree has given filename."""
57
raise NotImplementedError()
59
def has_id(self, file_id):
60
return self.inventory.has_id(file_id)
63
"""Return set of all ids in this tree."""
64
return self.inventory.id_set()
66
def id2path(self, file_id):
67
return self.inventory.id2path(file_id)
69
def _get_inventory(self):
70
return self._inventory
72
inventory = property(_get_inventory,
73
doc="Inventory of this Tree")
75
def _check_retrieved(self, ie, f):
76
fp = fingerprint_file(f)
79
if ie.text_size is not None:
81
bailout("mismatched size for file %r in %r" % (ie.file_id, self._store),
82
["inventory expects %d bytes" % ie.text_size,
83
"file is actually %d bytes" % fp['size'],
84
"store is probably damaged/corrupt"])
86
if ie.text_sha1 != fp['sha1']:
87
bailout("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
88
["inventory expects %s" % ie.text_sha1,
89
"file is actually %s" % fp['sha1'],
90
"store is probably damaged/corrupt"])
93
def export(self, dest):
94
"""Export this tree to a new directory.
96
`dest` should not exist, and will be created holding the
97
contents of this tree.
99
:todo: To handle subdirectories we need to create the
102
:note: If the export fails, the destination directory will be
103
left in a half-assed state.
106
mutter('export version %r' % self)
108
for dp, ie in inv.iter_entries():
110
fullpath = appendpath(dest, dp)
111
if kind == 'directory':
114
pumpfile(self.get_file(ie.file_id), file(fullpath, 'wb'))
116
bailout("don't know how to export {%s} of kind %r", fid, kind)
117
mutter(" export {%s} kind %s to %s" % (ie.file_id, kind, fullpath))
121
class WorkingTree(Tree):
122
"""Working copy tree.
124
The inventory is held in the `Branch` working-inventory, and the
125
files are in a directory on disk.
127
It is possible for a `WorkingTree` to have a filename which is
128
not listed in the Inventory and vice versa.
130
def __init__(self, basedir, inv):
131
self._inventory = inv
132
self.basedir = basedir
133
self.path2id = inv.path2id
136
return "<%s of %s>" % (self.__class__.__name__,
139
def abspath(self, filename):
140
return os.path.join(self.basedir, filename)
142
def has_filename(self, filename):
143
return os.path.exists(self.abspath(filename))
145
def get_file(self, file_id):
146
return self.get_file_byname(self.id2path(file_id))
148
def get_file_byname(self, filename):
149
return file(self.abspath(filename), 'rb')
151
def _get_store_filename(self, file_id):
152
return self.abspath(self.id2path(file_id))
154
def has_id(self, file_id):
155
# files that have been deleted are excluded
156
if not self.inventory.has_id(file_id):
158
return os.access(self.abspath(self.inventory.id2path(file_id)), os.F_OK)
160
def get_file_size(self, file_id):
161
return os.stat(self._get_store_filename(file_id))[ST_SIZE]
163
def get_file_sha1(self, file_id):
164
f = self.get_file(file_id)
168
def file_class(self, filename):
169
if self.path2id(filename):
171
elif self.is_ignored(filename):
177
def file_kind(self, filename):
178
if isfile(self.abspath(filename)):
180
elif isdir(self.abspath(filename)):
186
def list_files(self):
187
"""Recursively list all files as (path, class, kind, id).
189
Lists, but does not descend into unversioned directories.
191
This does not include files that have been deleted in this
194
Skips the control directory.
198
def descend(from_dir, from_dir_id, dp):
202
if bzrlib.BZRDIR == f:
206
fp = appendpath(from_dir, f)
209
fap = appendpath(dp, f)
211
f_ie = inv.get_child(from_dir_id, f)
214
elif self.is_ignored(fp):
223
bailout("file %r entered as kind %r id %r, now of kind %r"
224
% (fap, f_ie.kind, f_ie.file_id, fk))
226
yield fp, c, fk, (f_ie and f_ie.file_id)
228
if fk != 'directory':
232
# don't descend unversioned directories
235
for ff in descend(fp, f_ie.file_id, fap):
238
for f in descend('', None, self.basedir):
243
def unknowns(self, path='', dir_id=None):
244
"""Yield names of unknown files in this WorkingTree.
246
If there are any unknown directories then only the directory is
247
returned, not all its children. But if there are unknown files
248
under a versioned subdirectory, they are returned.
250
Currently returned depth-first, sorted by name within directories.
252
for fpath, fclass, fkind, fid in self.list_files():
257
def ignored_files(self):
258
for fpath, fclass, fkind, fid in self.list_files():
263
def get_ignore_list(self):
264
"""Return list of ignore patterns.
266
Cached in the Tree object after the first call.
268
if hasattr(self, '_ignorelist'):
269
return self._ignorelist
271
l = bzrlib.DEFAULT_IGNORE[:]
272
if self.has_filename(bzrlib.IGNORE_FILENAME):
273
f = self.get_file_byname(bzrlib.IGNORE_FILENAME)
274
l.extend([line.rstrip("\n\r") for line in f.readlines()])
279
def is_ignored(self, filename):
280
"""Check whether the filename matches an ignore pattern.
282
Patterns containing '/' need to match the whole path; others
283
match against only the last component."""
284
## TODO: Use extended zsh-style globs maybe?
285
## TODO: Use '**' to match directories?
286
for pat in self.get_ignore_list():
288
if fnmatch.fnmatchcase(filename, pat):
291
if fnmatch.fnmatchcase(splitpath(filename)[-1], pat):
299
class RevisionTree(Tree):
300
"""Tree viewing a previous revision.
302
File text can be retrieved from the text store.
304
:todo: Some kind of `__repr__` method, but a good one
305
probably means knowing the branch and revision number,
306
or at least passing a description to the constructor.
309
def __init__(self, store, inv):
311
self._inventory = inv
313
def get_file(self, file_id):
314
ie = self._inventory[file_id]
315
f = self._store[ie.text_id]
316
mutter(" get fileid{%s} from %r" % (file_id, self))
317
## self._check_retrieved(ie, f)
320
def get_file_size(self, file_id):
321
return self._inventory[file_id].text_size
323
def get_file_sha1(self, file_id):
324
ie = self._inventory[file_id]
327
def has_filename(self, filename):
328
return bool(self.inventory.path2id(filename))
330
def list_files(self):
331
# The only files returned by this are those from the version
332
for path, entry in self.inventory.iter_entries():
333
yield path, 'V', entry.kind, entry.file_id
336
class EmptyTree(Tree):
338
self._inventory = Inventory()
340
def has_filename(self, filename):
343
def list_files(self):
344
if False: # just to make it a generator
349
######################################################################
352
# TODO: Merge these two functions into a single one that can operate
353
# on either a whole tree or a set of files.
355
# TODO: Return the diff in order by filename, not by category or in
356
# random order. Can probably be done by lock-stepping through the
357
# filenames from both trees.
360
def file_status(filename, old_tree, new_tree):
361
"""Return single-letter status, old and new names for a file.
363
The complexity here is in deciding how to represent renames;
364
many complex cases are possible.
366
old_inv = old_tree.inventory
367
new_inv = new_tree.inventory
368
new_id = new_inv.path2id(filename)
369
old_id = old_inv.path2id(filename)
371
if not new_id and not old_id:
372
# easy: doesn't exist in either; not versioned at all
373
if new_tree.is_ignored(filename):
374
return 'I', None, None
376
return '?', None, None
378
# There is now a file of this name, great.
381
# There is no longer a file of this name, but we can describe
382
# what happened to the file that used to have
383
# this name. There are two possibilities: either it was
384
# deleted entirely, or renamed.
386
if new_inv.has_id(old_id):
387
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
389
return 'D', old_inv.id2path(old_id), None
391
# if the file_id is new in this revision, it is added
392
if new_id and not old_inv.has_id(new_id):
395
# if there used to be a file of this name, but that ID has now
396
# disappeared, it is deleted
397
if old_id and not new_inv.has_id(old_id):