~bzr-pqm/bzr/bzr.dev

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
# Copyright (C) 2005 by Canonical Development Ltd

# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.

# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.

# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA

# TODO: Could remember a bias towards whether a particular store is typically
# compressed or not.

"""
Stores are the main data-storage mechanism for Bazaar-NG.

A store is a simple write-once container indexed by a universally
unique ID.
"""

import os
from cStringIO import StringIO
import urllib
from zlib import adler32

import bzrlib
import bzrlib.errors as errors
from bzrlib.errors import BzrError, UnlistableStore, TransportNotPossible
from bzrlib.trace import mutter
import bzrlib.transport as transport
from bzrlib.transport.local import LocalTransport

######################################################################
# stores

class StoreError(Exception):
    pass


class Store(object):
    """This class represents the abstract storage layout for saving information.
    
    Files can be added, but not modified once they are in.  Typically
    the hash is used as the name, or something else known to be unique,
    such as a UUID.
    """

    def __len__(self):
        raise NotImplementedError('Children should define their length')

    def get(self, fileid, suffix=None):
        """Returns a file reading from a particular entry.
        
        If suffix is present, retrieve the named suffix for fileid.
        """
        raise NotImplementedError

    def __getitem__(self, fileid):
        """DEPRECATED. Please use .get(fileid) instead."""
        raise NotImplementedError

    #def __contains__(self, fileid):
    #    """Deprecated, please use has_id"""
    #    raise NotImplementedError

    def __iter__(self):
        raise NotImplementedError

    def add(self, f, fileid):
        """Add a file object f to the store accessible from the given fileid"""
        raise NotImplementedError('Children of Store must define their method of adding entries.')

    def has_id(self, fileid, suffix=None):
        """Return True or false for the presence of fileid in the store.
        
        suffix, if present, is a per file suffix, i.e. for digital signature 
        data."""
        raise NotImplementedError

    def listable(self):
        """Return True if this store is able to be listed."""
        return hasattr(self, "__iter__")

    def copy_multi(self, other, ids, pb=None, permit_failure=False):
        """Copy texts for ids from other into self.

        If an id is present in self, it is skipped.  A count of copied
        ids is returned, which may be less than len(ids).

        :param other: Another Store object
        :param ids: A list of entry ids to be copied
        :param pb: A ProgressBar object, if none is given, the default will be created.
        :param permit_failure: Allow missing entries to be ignored
        :return: (n_copied, [failed]) The number of entries copied successfully,
            followed by a list of entries which could not be copied (because they
            were missing)
        """
        if pb is None:
            pb = bzrlib.ui.ui_factory.progress_bar()
        pb.update('preparing to copy')
        failed = set()
        count = 0
        ids = list(ids) # get the list for showing a length.
        for fileid in ids:
            count += 1
            if self.has_id(fileid):
                continue
            try:
                self._copy_one(fileid, None, other, pb)
                for suffix in self._suffixes:
                    try:
                        self._copy_one(fileid, suffix, other, pb)
                    except KeyError:
                        pass
                pb.update('copy', count, len(ids))
            except KeyError:
                if permit_failure:
                    failed.add(fileid)
                else:
                    raise
        assert count == len(ids)
        pb.clear()
        return count, failed

    def _copy_one(self, fileid, suffix, other, pb):
        """Most generic copy-one object routine.
        
        Subclasses can override this to provide an optimised
        copy between their own instances. Such overriden routines
        should call this if they have no optimised facility for a 
        specific 'other'.
        """
        mutter('Store._copy_one: %r', fileid)
        f = other.get(fileid, suffix)
        self.add(f, fileid, suffix)


class TransportStore(Store):
    """A TransportStore is a Store superclass for Stores that use Transports."""

    def add(self, f, fileid, suffix=None):
        """Add contents of a file into the store.

        f -- A file-like object, or string
        """
        mutter("add store entry %r", fileid)
        
        names = self._id_to_names(fileid, suffix)
        if self._transport.has_any(names):
            raise BzrError("store %r already contains id %r" 
                           % (self._transport.base, fileid))

        # Most of the time, just adding the file will work
        # if we find a time where it fails, (because the dir
        # doesn't exist), then create the dir, and try again
        self._add(names[0], f)


    def _add(self, relpath, f):
        """Actually add the file to the given location.
        This should be overridden by children.
        """
        raise NotImplementedError('children need to implement this function.')

    def _check_fileid(self, fileid):
        if not isinstance(fileid, basestring):
            raise TypeError('Fileids should be a string type: %s %r' % (type(fileid), fileid))
        if '\\' in fileid or '/' in fileid:
            raise ValueError("invalid store id %r" % fileid)

    def _id_to_names(self, fileid, suffix):
        """Return the names in the expected order"""
        if suffix is not None:
            fn = self._relpath(fileid, [suffix])
        else:
            fn = self._relpath(fileid)

        fn_gz = fn + '.gz'
        if self._compressed:
            return fn_gz, fn
        else:
            return fn, fn_gz

    def has_id(self, fileid, suffix=None):
        """See Store.has_id."""
        return self._transport.has_any(self._id_to_names(fileid, suffix))

    def _get_name(self, fileid, suffix=None):
        """A special check, which returns the name of an existing file.
        
        This is similar in spirit to 'has_id', but it is designed
        to return information about which file the store has.
        """
        for name in self._id_to_names(fileid, suffix=suffix):
            if self._transport.has(name):
                return name
        return None

    def _get(self, filename):
        """Return an vanilla file stream for clients to read from.

        This is the body of a template method on 'get', and should be 
        implemented by subclasses.
        """
        raise NotImplementedError

    def get(self, fileid, suffix=None):
        """See Store.get()."""
        names = self._id_to_names(fileid, suffix)
        for name in names:
            try:
                return self._get(name)
            except errors.NoSuchFile:
                pass
        raise KeyError(fileid)

    def __init__(self, a_transport, prefixed=False, compressed=False):
        assert isinstance(a_transport, transport.Transport)
        super(TransportStore, self).__init__()
        self._transport = a_transport
        self._prefixed = prefixed
        self._compressed = compressed
        self._suffixes = set()

    def _iter_files_recursive(self):
        """Iterate through the files in the transport."""
        for quoted_relpath in self._transport.iter_files_recursive():
            yield urllib.unquote(quoted_relpath)

    def __iter__(self):
        for relpath in self._iter_files_recursive():
            # worst case is one of each suffix.
            name = os.path.basename(relpath)
            if name.endswith('.gz'):
                name = name[:-3]
            skip = False
            for count in range(len(self._suffixes)):
                for suffix in self._suffixes:
                    if name.endswith('.' + suffix):
                        skip = True
            if not skip:
                yield name

    def __len__(self):
        return len(list(self.__iter__()))

    def _relpath(self, fileid, suffixes=None):
        self._check_fileid(fileid)
        if suffixes:
            for suffix in suffixes:
                if not suffix in self._suffixes:
                    raise ValueError("Unregistered suffix %r" % suffix)
                self._check_fileid(suffix)
        else:
            suffixes = []
        if self._prefixed:
            path = [hash_prefix(fileid) + fileid]
        else:
            path = [fileid]
        path.extend(suffixes)
        return transport.urlescape('.'.join(path))

    def __repr__(self):
        if self._transport is None:
            return "%s(None)" % (self.__class__.__name__)
        else:
            return "%s(%r)" % (self.__class__.__name__, self._transport.base)

    __str__ = __repr__

    def listable(self):
        """Return True if this store is able to be listed."""
        return self._transport.listable()

    def register_suffix(self, suffix):
        """Register a suffix as being expected in this store."""
        self._check_fileid(suffix)
        if suffix == 'gz':
            raise ValueError('You cannot register the "gz" suffix.')
        self._suffixes.add(suffix)

    def total_size(self):
        """Return (count, bytes)

        This is the (compressed) size stored on disk, not the size of
        the content."""
        total = 0
        count = 0
        for relpath in self._transport.iter_files_recursive():
            count += 1
            total += self._transport.stat(relpath).st_size
                
        return count, total


def ImmutableMemoryStore():
    return bzrlib.store.text.TextStore(transport.memory.MemoryTransport())
        

class CachedStore(Store):
    """A store that caches data locally, to avoid repeated downloads.
    The precacache method should be used to avoid server round-trips for
    every piece of data.
    """

    def __init__(self, store, cache_dir):
        super(CachedStore, self).__init__()
        self.source_store = store
        # This clones the source store type with a locally bound
        # transport. FIXME: it assumes a constructor is == cloning.
        # clonable store - it might be nicer to actually have a clone()
        # or something. RBC 20051003
        self.cache_store = store.__class__(LocalTransport(cache_dir))

    def get(self, id):
        mutter("Cache add %s", id)
        if id not in self.cache_store:
            self.cache_store.add(self.source_store.get(id), id)
        return self.cache_store.get(id)

    def has_id(self, fileid, suffix=None):
        """See Store.has_id."""
        if self.cache_store.has_id(fileid, suffix):
            return True
        if self.source_store.has_id(fileid, suffix):
            # We could asynchronously copy at this time
            return True
        return False


def copy_all(store_from, store_to):
    """Copy all ids from one store to another."""
    # TODO: Optional progress indicator
    if not store_from.listable():
        raise UnlistableStore(store_from)
    ids = [f for f in store_from]
    mutter('copy_all ids: %r', ids)
    store_to.copy_multi(store_from, ids)

def hash_prefix(fileid):
    return "%02x/" % (adler32(fileid) & 0xff)