2
# -*- coding: UTF-8 -*-
4
# This program is free software; you can redistribute it and/or modify
5
# it under the terms of the GNU General Public License as published by
6
# the Free Software Foundation; either version 2 of the License, or
7
# (at your option) any later version.
9
# This program is distributed in the hope that it will be useful,
10
# but WITHOUT ANY WARRANTY; without even the implied warranty of
11
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12
# GNU General Public License for more details.
14
# You should have received a copy of the GNU General Public License
15
# along with this program; if not, write to the Free Software
16
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
18
"""Tree classes, representing directory at point in time.
22
import os.path, os, fnmatch
24
from inventory import Inventory
25
from trace import mutter, note
26
from osutils import pumpfile, compare_files, filesize, quotefn, sha_file, \
27
joinpath, splitpath, appendpath, isdir, isfile, file_kind
28
from errors import bailout
30
from stat import S_ISREG, S_ISDIR, ST_MODE, ST_SIZE
35
"""Abstract file tree.
37
There are several subclasses:
39
* `WorkingTree` exists as files on disk editable by the user.
41
* `RevisionTree` is a tree as recorded at some point in the past.
45
Trees contain an `Inventory` object, and also know how to retrieve
46
file texts mentioned in the inventory, either from a working
47
directory or from a store.
49
It is possible for trees to contain files that are not described
50
in their inventory or vice versa; for this use `filenames()`.
52
Trees can be compared, etc, regardless of whether they are working
53
trees or versioned trees.
56
def has_filename(self, filename):
57
"""True if the tree has given filename."""
58
raise NotImplementedError()
60
def has_id(self, file_id):
61
return self.inventory.has_id(file_id)
64
"""Return set of all ids in this tree."""
65
return self.inventory.id_set()
67
def id2path(self, file_id):
68
return self.inventory.id2path(file_id)
70
def _get_inventory(self):
71
return self._inventory
73
inventory = property(_get_inventory,
74
doc="Inventory of this Tree")
76
def _check_retrieved(self, ie, f):
77
# TODO: Test this check by damaging the store?
78
if ie.text_size is not None:
80
if fs != ie.text_size:
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" % fs,
84
"store is probably damaged/corrupt"])
88
if ie.text_sha1 != f_hash:
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" % f_hash,
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 _rel(self, filename):
142
return os.path.join(self.basedir, filename)
144
def has_filename(self, filename):
145
return os.path.exists(self._rel(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._rel(filename), 'rb')
153
def _get_store_filename(self, file_id):
154
return self._rel(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._rel(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 file_kind(self, filename):
180
if isfile(self._rel(filename)):
182
elif isdir(self._rel(filename)):
188
def list_files(self):
189
"""Recursively list all files as (path, class, kind, id).
191
Lists, but does not descend into unversioned directories.
193
This does not include files that have been deleted in this
196
Skips the control directory.
200
def descend(from_dir, from_dir_id, dp):
204
if bzrlib.BZRDIR == f:
208
fp = appendpath(from_dir, f)
211
fap = appendpath(dp, f)
213
f_ie = inv.get_child(from_dir_id, f)
216
elif self.is_ignored(fp):
225
bailout("file %r entered as kind %r id %r, now of kind %r"
226
% (fap, f_ie.kind, f_ie.file_id, fk))
228
yield fp, c, fk, (f_ie and f_ie.file_id)
230
if fk != 'directory':
234
# don't descend unversioned directories
237
for ff in descend(fp, f_ie.file_id, fap):
240
for f in descend('', None, self.basedir):
245
def unknowns(self, path='', dir_id=None):
246
"""Yield names of 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
for fpath, fclass, fkind, fid in self.list_files():
259
def ignored_files(self):
260
for fpath, fclass, fkind, fid in self.list_files():
265
def get_ignore_list(self):
266
"""Return list of ignore patterns."""
267
if self.has_filename(bzrlib.IGNORE_FILENAME):
268
f = self.get_file_byname(bzrlib.IGNORE_FILENAME)
269
return [line.rstrip("\n\r") for line in f.readlines()]
271
return bzrlib.DEFAULT_IGNORE
274
def is_ignored(self, filename):
275
"""Check whether the filename matches an ignore pattern.
277
Patterns containing '/' need to match the whole path; others
278
match against only the last component."""
279
## TODO: Take them from a file, not hardcoded
280
## TODO: Use extended zsh-style globs maybe?
281
## TODO: Use '**' to match directories?
282
for pat in self.get_ignore_list():
284
if fnmatch.fnmatchcase(filename, pat):
287
if fnmatch.fnmatchcase(splitpath(filename)[-1], pat):
295
class RevisionTree(Tree):
296
"""Tree viewing a previous revision.
298
File text can be retrieved from the text store.
300
:todo: Some kind of `__repr__` method, but a good one
301
probably means knowing the branch and revision number,
302
or at least passing a description to the constructor.
305
def __init__(self, store, inv):
307
self._inventory = inv
309
def get_file(self, file_id):
310
ie = self._inventory[file_id]
311
f = self._store[ie.text_id]
312
mutter(" get fileid{%s} from %r" % (file_id, self))
314
if ie.text_size is None:
315
note("warning: no text size recorded on %r" % ie)
316
self._check_retrieved(ie, f)
319
def get_file_size(self, file_id):
320
return self._inventory[file_id].text_size
322
def get_file_sha1(self, file_id):
323
ie = self._inventory[file_id]
326
def has_filename(self, filename):
327
return bool(self.inventory.path2id(filename))
329
def list_files(self):
330
# The only files returned by this are those from the version
331
for path, entry in self.inventory.iter_entries():
332
yield path, 'V', entry.kind, entry.file_id
335
class EmptyTree(Tree):
337
self._inventory = Inventory()
339
def has_filename(self, filename):
342
def list_files(self):
343
if False: # just to make it a generator
348
######################################################################
351
# TODO: Merge these two functions into a single one that can operate
352
# on either a whole tree or a set of files.
354
# TODO: Return the diff in order by filename, not by category or in
355
# random order. Can probably be done by lock-stepping through the
356
# filenames from both trees.
359
def file_status(filename, old_tree, new_tree):
360
"""Return single-letter status, old and new names for a file.
362
The complexity here is in deciding how to represent renames;
363
many complex cases are possible.
365
old_inv = old_tree.inventory
366
new_inv = new_tree.inventory
367
new_id = new_inv.path2id(filename)
368
old_id = old_inv.path2id(filename)
370
if not new_id and not old_id:
371
# easy: doesn't exist in either; not versioned at all
372
if new_tree.is_ignored(filename):
373
return 'I', None, None
375
return '?', None, None
377
# There is now a file of this name, great.
380
# There is no longer a file of this name, but we can describe
381
# what happened to the file that used to have
382
# this name. There are two possibilities: either it was
383
# deleted entirely, or renamed.
385
if new_inv.has_id(old_id):
386
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
388
return 'D', old_inv.id2path(old_id), None
390
# if the file_id is new in this revision, it is added
391
if new_id and not old_inv.has_id(new_id):
394
# if there used to be a file of this name, but that ID has now
395
# disappeared, it is deleted
396
if old_id and not new_inv.has_id(old_id):
added
removed
bzrlib/tree.py
