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:
80
if ie.text_size != fp['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" % 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 list_files(self):
178
"""Recursively list all files as (path, class, kind, id).
180
Lists, but does not descend into unversioned directories.
182
This does not include files that have been deleted in this
185
Skips the control directory.
189
def descend(from_dir, from_dir_id, dp):
193
## TODO: If we find a subdirectory with its own .bzr
194
## directory, then that is a separate tree and we
195
## should exclude it.
196
if bzrlib.BZRDIR == f:
200
fp = appendpath(from_dir, f)
203
fap = appendpath(dp, f)
205
f_ie = inv.get_child(from_dir_id, f)
208
elif self.is_ignored(fp):
217
bailout("file %r entered as kind %r id %r, now of kind %r"
218
% (fap, f_ie.kind, f_ie.file_id, fk))
220
yield fp, c, fk, (f_ie and f_ie.file_id)
222
if fk != 'directory':
226
# don't descend unversioned directories
229
for ff in descend(fp, f_ie.file_id, fap):
232
for f in descend('', None, self.basedir):
238
for subp in self.extras():
239
if not self.is_ignored(subp):
244
"""Yield all 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
## TODO: Work from given directory downwards
254
for path, dir_entry in self.inventory.directories():
255
mutter("search for unknowns in %r" % path)
256
dirabs = self.abspath(path)
257
if not isdir(dirabs):
258
# e.g. directory deleted
262
for subf in os.listdir(dirabs):
264
and (subf not in dir_entry.children)):
269
subp = appendpath(path, subf)
273
def ignored_files(self):
274
"""Yield list of PATH, IGNORE_PATTERN"""
275
for subp in self.extras():
276
pat = self.is_ignored(subp)
281
def get_ignore_list(self):
282
"""Return list of ignore patterns.
284
Cached in the Tree object after the first call.
286
if hasattr(self, '_ignorelist'):
287
return self._ignorelist
289
l = bzrlib.DEFAULT_IGNORE[:]
290
if self.has_filename(bzrlib.IGNORE_FILENAME):
291
f = self.get_file_byname(bzrlib.IGNORE_FILENAME)
292
l.extend([line.rstrip("\n\r") for line in f.readlines()])
297
def is_ignored(self, filename):
298
"""Check whether the filename matches an ignore pattern.
300
Patterns containing '/' need to match the whole path; others
301
match against only the last component.
303
If the file is ignored, returns the pattern which caused it to
304
be ignored, otherwise None. So this can simply be used as a
305
boolean if desired."""
307
## TODO: Use '**' to match directories, and other extended globbing stuff from cvs/rsync.
309
for pat in self.get_ignore_list():
311
# as a special case, you can put ./ at the start of a pattern;
312
# this is good to match in the top-level only;
317
if fnmatch.fnmatchcase(filename, newpat):
320
if fnmatch.fnmatchcase(splitpath(filename)[-1], pat):
328
class RevisionTree(Tree):
329
"""Tree viewing a previous revision.
331
File text can be retrieved from the text store.
333
:todo: Some kind of `__repr__` method, but a good one
334
probably means knowing the branch and revision number,
335
or at least passing a description to the constructor.
338
def __init__(self, store, inv):
340
self._inventory = inv
342
def get_file(self, file_id):
343
ie = self._inventory[file_id]
344
f = self._store[ie.text_id]
345
mutter(" get fileid{%s} from %r" % (file_id, self))
346
self._check_retrieved(ie, f)
349
def get_file_size(self, file_id):
350
return self._inventory[file_id].text_size
352
def get_file_sha1(self, file_id):
353
ie = self._inventory[file_id]
356
def has_filename(self, filename):
357
return bool(self.inventory.path2id(filename))
359
def list_files(self):
360
# The only files returned by this are those from the version
361
for path, entry in self.inventory.iter_entries():
362
yield path, 'V', entry.kind, entry.file_id
365
class EmptyTree(Tree):
367
self._inventory = Inventory()
369
def has_filename(self, filename):
372
def list_files(self):
373
if False: # just to make it a generator
378
######################################################################
381
# TODO: Merge these two functions into a single one that can operate
382
# on either a whole tree or a set of files.
384
# TODO: Return the diff in order by filename, not by category or in
385
# random order. Can probably be done by lock-stepping through the
386
# filenames from both trees.
389
def file_status(filename, old_tree, new_tree):
390
"""Return single-letter status, old and new names for a file.
392
The complexity here is in deciding how to represent renames;
393
many complex cases are possible.
395
old_inv = old_tree.inventory
396
new_inv = new_tree.inventory
397
new_id = new_inv.path2id(filename)
398
old_id = old_inv.path2id(filename)
400
if not new_id and not old_id:
401
# easy: doesn't exist in either; not versioned at all
402
if new_tree.is_ignored(filename):
403
return 'I', None, None
405
return '?', None, None
407
# There is now a file of this name, great.
410
# There is no longer a file of this name, but we can describe
411
# what happened to the file that used to have
412
# this name. There are two possibilities: either it was
413
# deleted entirely, or renamed.
415
if new_inv.has_id(old_id):
416
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
418
return 'D', old_inv.id2path(old_id), None
420
# if the file_id is new in this revision, it is added
421
if new_id and not old_inv.has_id(new_id):
424
# if there used to be a file of this name, but that ID has now
425
# disappeared, it is deleted
426
if old_id and not new_inv.has_id(old_id):