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
|
# Copyright 2010 Lars Wirzenius
#
# 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 3 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, see <http://www.gnu.org/licenses/>.
import logging
import os
import StringIO
import struct
import tempfile
import tracing
import larch
def encode_refcounts(refcounts, start_id, how_many):
fmt = '!QH' + 'H' * how_many
args = [start_id, how_many] + ([0] * how_many)
keys = set(refcounts.keys())
wanted = set(range(start_id, start_id + how_many))
for key in wanted.intersection(keys):
args[2 + key - start_id] = refcounts[key]
return struct.pack(fmt, *args)
def decode_refcounts(encoded):
n = struct.calcsize('!QH')
start_id, how_many = struct.unpack('!QH', encoded[:n])
counts = struct.unpack('!' + 'H' * how_many, encoded[n:])
return zip(range(start_id, start_id + how_many), counts)
class RefcountStore(object):
'''Store node reference counts.
Each node has a reference count, which gets stored on disk.
Reference counts are grouped into blocks of ``self.per_group`` counts,
and each group is stored in its own file. This balances the
per-file overhead with the overhead of keeping a lot of unneeded
reference counts in memory.
Only those blocks that are used get loaded into memory. Blocks
that are full of zeroes are not stored in files, to save disk space.
'''
per_group = 2**15
refcountdir = 'refcounts'
def __init__(self, node_store):
self.node_store = node_store
self.refcounts = dict()
self.dirty = set()
def get_refcount(self, node_id):
'''Return reference count for a given node.'''
if node_id not in self.refcounts:
group = self._load_refcount_group(self._start_id(node_id))
if group is None:
self.refcounts[node_id] = 0
else:
for x, count in group:
if x not in self.dirty:
self.refcounts[x] = count
return self.refcounts[node_id]
def set_refcount(self, node_id, refcount):
'''Set the reference count for a given node.'''
tracing.trace('setting refcoutn for %s to %s' % (node_id, refcount))
if refcount == 0:
if node_id in self.refcounts:
del self.refcounts[node_id]
else:
self.refcounts[node_id] = refcount
self.dirty.add(node_id)
def save_refcounts(self):
'''Save all modified refcounts.'''
tracing.trace('saving refcounts (len(dirty) = %s)' %
(len(self.dirty)))
if self.dirty:
level = logging.getLogger().getEffectiveLevel()
dirname = os.path.join(self.node_store.dirname, self.refcountdir)
if not self.node_store.journal.exists(dirname):
self.node_store.journal.makedirs(dirname)
ids = sorted(self.dirty)
for start_id in range(self._start_id(ids[0]),
self._start_id(ids[-1]) + 1,
self.per_group):
encoded = encode_refcounts(self.refcounts, start_id,
self.per_group)
filename = self._group_filename(start_id)
self.node_store.journal.overwrite_file(filename, encoded)
# We re-initialize these so that they don't grow indefinitely.
self.refcounts = dict()
self.dirty = set()
def _load_refcount_group(self, start_id):
filename = self._group_filename(start_id)
if self.node_store.journal.exists(filename):
encoded = self.node_store.journal.cat(filename)
return decode_refcounts(encoded)
def _group_filename(self, start_id):
return os.path.join(self.node_store.dirname, self.refcountdir,
'refcounts-%d' % start_id)
def _start_id(self, node_id):
return (node_id / self.per_group) * self.per_group
|
