~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
# 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

"""
An implementation the primary storage type CompressedTextStore.

This store keeps compressed versions of the full text. It does not
do any sort of delta compression.
"""

import os, tempfile, gzip

import bzrlib.store
from bzrlib.store import hash_prefix
from bzrlib.trace import mutter
from bzrlib.errors import BzrError, FileExists

from StringIO import StringIO
from stat import ST_SIZE

class CompressedTextStore(bzrlib.store.TransportStore):
    """Store that holds files indexed by unique names.

    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.

    Files are stored gzip compressed, with no delta compression.

    >>> st = ScratchCompressedTextStore()

    >>> st.add(StringIO('hello'), 'aa')
    >>> 'aa' in st
    True
    >>> 'foo' in st
    False

    You are not allowed to add an id that is already present.

    Entries can be retrieved as files, which may then be read.

    >>> st.add(StringIO('goodbye'), '123123')
    >>> st['123123'].read()
    'goodbye'
    """

    def __init__(self, transport, prefixed=False):
        super(CompressedTextStore, self).__init__(transport)
        self._prefixed = prefixed

    def _check_fileid(self, fileid):
        if '\\' in fileid or '/' in fileid:
            raise ValueError("invalid store id %r" % fileid)

    def _relpath(self, fileid):
        self._check_fileid(fileid)
        if self._prefixed:
            return hash_prefix(fileid) + fileid + ".gz"
        else:
            return fileid + ".gz"

    def add(self, f, fileid):
        """Add contents of a file into the store.

        f -- An open file, or file-like object."""
        # TODO: implement an add_multi which can do some of it's
        #       own piplelining, and possible take advantage of
        #       transport.put_multi(). The problem is that
        #       entries potentially need to be compressed as they
        #       are received, which implies translation, which
        #       means it isn't as straightforward as we would like.
        from cStringIO import StringIO
        from bzrlib.osutils import pumpfile
        
        mutter("add store entry %r" % (fileid))
        if isinstance(f, basestring):
            f = StringIO(f)
            
        fn = self._relpath(fileid)
        if self._transport.has(fn):
            raise BzrError("store %r already contains id %r" % (self._transport.base, fileid))

        if self._prefixed:
            try:
                self._transport.mkdir(hash_prefix(fileid))
            except FileExists:
                pass

        sio = StringIO()
        gf = gzip.GzipFile(mode='wb', fileobj=sio)
        # if pumpfile handles files that don't fit in ram,
        # so will this function
        if isinstance(f, basestring):
            gf.write(f)
        else:
            pumpfile(f, gf)
        gf.close()
        sio.seek(0)
        self._transport.put(fn, sio)

    def _do_copy(self, other, to_copy, pb, permit_failure=False):
        if isinstance(other, CompressedTextStore):
            return self._copy_multi_text(other, to_copy, pb,
                    permit_failure=permit_failure)
        return super(CompressedTextStore, self)._do_copy(other, to_copy,
                pb, permit_failure=permit_failure)

    def _copy_multi_text(self, other, to_copy, pb,
            permit_failure=False):
        # Because of _transport, we can no longer assume
        # that they are on the same filesystem, we can, however
        # assume that we only need to copy the exact bytes,
        # we don't need to process the files.

        failed = set()
        if permit_failure:
            new_to_copy = set()
            for fileid, has in zip(to_copy, other.has(to_copy)):
                if has:
                    new_to_copy.add(fileid)
                else:
                    failed.add(fileid)
            to_copy = new_to_copy
            #mutter('_copy_multi_text copying %s, failed %s' % (to_copy, failed))

        paths = [self._relpath(fileid) for fileid in to_copy]
        count = other._transport.copy_to(paths, self._transport, pb=pb)
        assert count == len(to_copy)
        return count, failed

    def __contains__(self, fileid):
        """"""
        fn = self._relpath(fileid)
        return self._transport.has(fn)

    def has(self, fileids, pb=None):
        """Return True/False for each entry in fileids.

        :param fileids: A List or generator yielding file ids.
        :return: A generator or list returning True/False for each entry.
        """
        relpaths = (self._relpath(fid) for fid in fileids)
        return self._transport.has_multi(relpaths, pb=pb)

    def get(self, fileids, permit_failure=False, pb=None):
        """Return a set of files, one for each requested entry.
        
        TODO: Write some tests to make sure that permit_failure is
              handled correctly.

        TODO: What should the exception be for a missing file?
              KeyError, or NoSuchFile?
        """

        # This next code gets a bit hairy because it can allow
        # to not request a file which doesn't seem to exist.
        # Also, the same fileid may be requested twice, so we
        # can't just build up a map.
        rel_paths = [self._relpath(fid) for fid in fileids]
        is_requested = []

        #mutter('CompressedTextStore.get(permit_failure=%s)' % permit_failure)
        if permit_failure:
            existing_paths = []
            for path, has in zip(rel_paths,
                    self._transport.has_multi(rel_paths)):
                if has:
                    existing_paths.append(path)
                    is_requested.append(True)
                else:
                    is_requested.append(False)
            #mutter('Retrieving %s out of %s' % (existing_paths, rel_paths))
        else:
            #mutter('Retrieving all %s' % (rel_paths, ))
            existing_paths = rel_paths
            is_requested = [True for x in rel_paths]

        count = 0
        for f in self._transport.get_multi(existing_paths, pb=pb):
            assert count < len(is_requested)
            while not is_requested[count]:
                yield None
                count += 1
            if hasattr(f, 'tell'):
                yield gzip.GzipFile(mode='rb', fileobj=f)
            else:
                from cStringIO import StringIO
                sio = StringIO(f.read())
                yield gzip.GzipFile(mode='rb', fileobj=sio)
            count += 1

        while count < len(is_requested):
            yield None
            count += 1

    def __iter__(self):
        for relpath, st in self._iter_relpaths():
            if relpath.endswith(".gz"):
                yield os.path.basename(relpath)[:-3]
            else:
                yield os.path.basename(relpath)

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

    def __getitem__(self, fileid):
        """Returns a file reading from a particular entry."""
        f = super(CompressedTextStore, self).__getitem__(fileid)
        # gzip.GzipFile.read() requires a tell() function
        # but some transports return objects that cannot seek
        # so buffer them in a StringIO instead
        if hasattr(f, 'tell'):
            return gzip.GzipFile(mode='rb', fileobj=f)
        else:
            from cStringIO import StringIO
            sio = StringIO(f.read())
            return gzip.GzipFile(mode='rb', fileobj=sio)

    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, st in self._iter_relpaths():
            count += 1
            total += st[ST_SIZE]
                
        return count, total


class ScratchCompressedTextStore(CompressedTextStore):
    """Self-destructing test subclass of CompressedTextStore.

    The Store only exists for the lifetime of the Python object.
    Obviously you should not put anything precious in it.
    """
    def __init__(self):
        from transport import transport
        t = transport(tempfile.mkdtemp())
        super(ScratchCompressedTextStore, self).__init__(t)

    def __del__(self):
        self._transport.delete_multi(self._transport.list_dir('.'))
        os.rmdir(self._transport.base)
        mutter("%r destroyed" % self)