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
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
# TODO: Test this check by damaging the store?
77
if ie.text_size is not None:
79
if fs != ie.text_size:
80
bailout("mismatched size for file %r in %r" % (ie.file_id, self._store),
81
["inventory expects %d bytes" % ie.text_size,
82
"file is actually %d bytes" % fs,
83
"store is probably damaged/corrupt"])
87
if ie.text_sha1 != f_hash:
88
bailout("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
89
["inventory expects %s" % ie.text_sha1,
90
"file is actually %s" % f_hash,
91
"store is probably damaged/corrupt"])
94
def export(self, dest):
95
"""Export this tree to a new directory.
97
`dest` should not exist, and will be created holding the
98
contents of this tree.
100
:todo: To handle subdirectories we need to create the
103
:note: If the export fails, the destination directory will be
104
left in a half-assed state.
107
mutter('export version %r' % self)
109
for dp, ie in inv.iter_entries():
111
fullpath = appendpath(dest, dp)
112
if kind == 'directory':
115
pumpfile(self.get_file(ie.file_id), file(fullpath, 'wb'))
117
bailout("don't know how to export {%s} of kind %r", fid, kind)
118
mutter(" export {%s} kind %s to %s" % (ie.file_id, kind, fullpath))
122
class WorkingTree(Tree):
123
"""Working copy tree.
125
The inventory is held in the `Branch` working-inventory, and the
126
files are in a directory on disk.
128
It is possible for a `WorkingTree` to have a filename which is
129
not listed in the Inventory and vice versa.
131
def __init__(self, basedir, inv):
132
self._inventory = inv
133
self.basedir = basedir
134
self.path2id = inv.path2id
137
return "<%s of %s>" % (self.__class__.__name__,
140
def abspath(self, filename):
141
return os.path.join(self.basedir, filename)
143
def has_filename(self, filename):
144
return os.path.exists(self.abspath(filename))
146
def get_file(self, file_id):
147
return self.get_file_byname(self.id2path(file_id))
149
def get_file_byname(self, filename):
150
return file(self.abspath(filename), 'rb')
152
def _get_store_filename(self, file_id):
153
return self.abspath(self.id2path(file_id))
155
def has_id(self, file_id):
156
# files that have been deleted are excluded
157
if not self.inventory.has_id(file_id):
159
return os.access(self.abspath(self.inventory.id2path(file_id)), os.F_OK)
161
def get_file_size(self, file_id):
162
return os.stat(self._get_store_filename(file_id))[ST_SIZE]
164
def get_file_sha1(self, file_id):
165
f = self.get_file(file_id)
169
def file_class(self, filename):
170
if self.path2id(filename):
172
elif self.is_ignored(filename):
178
def file_kind(self, filename):
179
if isfile(self.abspath(filename)):
181
elif isdir(self.abspath(filename)):
187
def list_files(self):
188
"""Recursively list all files as (path, class, kind, id).
190
Lists, but does not descend into unversioned directories.
192
This does not include files that have been deleted in this
195
Skips the control directory.
199
def descend(from_dir, from_dir_id, dp):
203
if bzrlib.BZRDIR == f:
207
fp = appendpath(from_dir, f)
210
fap = appendpath(dp, f)
212
f_ie = inv.get_child(from_dir_id, f)
215
elif self.is_ignored(fp):
224
bailout("file %r entered as kind %r id %r, now of kind %r"
225
% (fap, f_ie.kind, f_ie.file_id, fk))
227
yield fp, c, fk, (f_ie and f_ie.file_id)
229
if fk != 'directory':
233
# don't descend unversioned directories
236
for ff in descend(fp, f_ie.file_id, fap):
239
for f in descend('', None, self.basedir):
244
def unknowns(self, path='', dir_id=None):
245
"""Yield names of unknown files in this WorkingTree.
247
If there are any unknown directories then only the directory is
248
returned, not all its children. But if there are unknown files
249
under a versioned subdirectory, they are returned.
251
Currently returned depth-first, sorted by name within directories.
253
for fpath, fclass, fkind, fid in self.list_files():
258
def ignored_files(self):
259
for fpath, fclass, fkind, fid in self.list_files():
264
def get_ignore_list(self):
265
"""Return list of ignore patterns."""
266
if self.has_filename(bzrlib.IGNORE_FILENAME):
267
f = self.get_file_byname(bzrlib.IGNORE_FILENAME)
268
return [line.rstrip("\n\r") for line in f.readlines()]
270
return bzrlib.DEFAULT_IGNORE
273
def is_ignored(self, filename):
274
"""Check whether the filename matches an ignore pattern.
276
Patterns containing '/' need to match the whole path; others
277
match against only the last component."""
278
## TODO: Take them from a file, not hardcoded
279
## TODO: Use extended zsh-style globs maybe?
280
## TODO: Use '**' to match directories?
281
for pat in self.get_ignore_list():
283
if fnmatch.fnmatchcase(filename, pat):
286
if fnmatch.fnmatchcase(splitpath(filename)[-1], pat):
294
class RevisionTree(Tree):
295
"""Tree viewing a previous revision.
297
File text can be retrieved from the text store.
299
:todo: Some kind of `__repr__` method, but a good one
300
probably means knowing the branch and revision number,
301
or at least passing a description to the constructor.
304
def __init__(self, store, inv):
306
self._inventory = inv
308
def get_file(self, file_id):
309
ie = self._inventory[file_id]
310
f = self._store[ie.text_id]
311
mutter(" get fileid{%s} from %r" % (file_id, self))
313
if ie.text_size is None:
314
note("warning: no text size recorded on %r" % ie)
315
self._check_retrieved(ie, f)
318
def get_file_size(self, file_id):
319
return self._inventory[file_id].text_size
321
def get_file_sha1(self, file_id):
322
ie = self._inventory[file_id]
325
def has_filename(self, filename):
326
return bool(self.inventory.path2id(filename))
328
def list_files(self):
329
# The only files returned by this are those from the version
330
for path, entry in self.inventory.iter_entries():
331
yield path, 'V', entry.kind, entry.file_id
334
class EmptyTree(Tree):
336
self._inventory = Inventory()
338
def has_filename(self, filename):
341
def list_files(self):
342
if False: # just to make it a generator
347
######################################################################
350
# TODO: Merge these two functions into a single one that can operate
351
# on either a whole tree or a set of files.
353
# TODO: Return the diff in order by filename, not by category or in
354
# random order. Can probably be done by lock-stepping through the
355
# filenames from both trees.
358
def file_status(filename, old_tree, new_tree):
359
"""Return single-letter status, old and new names for a file.
361
The complexity here is in deciding how to represent renames;
362
many complex cases are possible.
364
old_inv = old_tree.inventory
365
new_inv = new_tree.inventory
366
new_id = new_inv.path2id(filename)
367
old_id = old_inv.path2id(filename)
369
if not new_id and not old_id:
370
# easy: doesn't exist in either; not versioned at all
371
if new_tree.is_ignored(filename):
372
return 'I', None, None
374
return '?', None, None
376
# There is now a file of this name, great.
379
# There is no longer a file of this name, but we can describe
380
# what happened to the file that used to have
381
# this name. There are two possibilities: either it was
382
# deleted entirely, or renamed.
384
if new_inv.has_id(old_id):
385
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
387
return 'D', old_inv.id2path(old_id), None
389
# if the file_id is new in this revision, it is added
390
if new_id and not old_inv.has_id(new_id):
393
# if there used to be a file of this name, but that ID has now
394
# disappeared, it is deleted
395
if old_id and not new_inv.has_id(old_id):
added
removed
bzrlib/tree.py
