« back to all changes in this revision
Viewing changes to bzrlib/lru_cache.py
- browse files at revision 2993.1.1
- compare with another revision
- compare with revision 4017
- download diff
- download tarball
- view history from revision 2993.1.1
- Committer: Robert Collins
- Date: 2007-11-14 21:07:54 UTC
- mto: This revision was merged to the branch mainline in revision 2996.
- Revision ID: robertc@robertcollins.net-20071114210754-fbts4wyrsxk34t5b
* New module ``lru_cache`` providing a cache for use by tasks that need
semi-random access to large amounts of data. (John A Meinel)
semi-random access to large amounts of data. (John A Meinel)
- files added:
- bzrlib/lru_cache.py
- files modified:
- NEWS
added
removed
bzrlib/lru_cache.py
1
# Copyright (C) 2006 Canonical Ltd
2
#
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.
7
#
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.
12
#
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
16
17
"""A simple least-recently-used (LRU) cache."""
18
19
from collections import deque
20
import gc
21
22
23
class LRUCache(object):
24
"""A class which manages a cache of entries, removing unused ones."""
25
26
def __init__(self, max_cache=100, after_cleanup_size=None):
27
self._max_cache = max_cache
28
if after_cleanup_size is None:
29
self._after_cleanup_size = self._max_cache
30
else:
31
self._after_cleanup_size = min(after_cleanup_size, self._max_cache)
32
33
self._compact_queue_length = 4*self._max_cache
34
35
self._cache = {}
36
self._cleanup = {}
37
self._queue = deque() # Track when things are accessed
38
self._refcount = {} # number of entries in self._queue for each key
39
40
def __contains__(self, key):
41
return key in self._cache
42
43
def __getitem__(self, key):
44
val = self._cache[key]
45
self._record_access(key)
46
return val
47
48
def __len__(self):
49
return len(self._cache)
50
51
def add(self, key, value, cleanup=None):
52
"""Add a new value to the cache.
53
54
Also, if the entry is ever removed from the queue, call cleanup.
55
Passing it the key and value being removed.
56
57
:param key: The key to store it under
58
:param value: The object to store
59
:param cleanup: None or a function taking (key, value) to indicate
60
'value' sohuld be cleaned up.
61
"""
62
if key in self._cache:
63
self._remove(key)
64
self._cache[key] = value
65
self._cleanup[key] = cleanup
66
self._record_access(key)
67
68
if len(self._cache) > self._max_cache:
69
# Trigger the cleanup
70
self.cleanup()
71
72
def cleanup(self):
73
"""Clear the cache until it shrinks to the requested size.
74
75
This does not completely wipe the cache, just makes sure it is under
76
the after_cleanup_size.
77
"""
78
# Make sure the cache is shrunk to the correct size
79
while len(self._cache) > self._after_cleanup_size:
80
self._remove_lru()
81
82
def __setitem__(self, key, value):
83
"""Add a value to the cache, there will be no cleanup function."""
84
self.add(key, value, cleanup=None)
85
86
def _record_access(self, key):
87
"""Record that key was accessed."""
88
self._queue.append(key)
89
# Can't use setdefault because you can't += 1 the result
90
self._refcount[key] = self._refcount.get(key, 0) + 1
91
92
# If our access queue is too large, clean it up too
93
if len(self._queue) > self._compact_queue_length:
94
self._compact_queue()
95
96
def _compact_queue(self):
97
"""Compact the queue, leaving things in sorted last appended order."""
98
new_queue = deque()
99
for item in self._queue:
100
if self._refcount[item] == 1:
101
new_queue.append(item)
102
else:
103
self._refcount[item] -= 1
104
self._queue = new_queue
105
# All entries should be of the same size. There should be one entry in
106
# queue for each entry in cache, and all refcounts should == 1
107
assert (len(self._queue) == len(self._cache) ==
108
len(self._refcount) == sum(self._refcount.itervalues()))
109
110
def _remove(self, key):
111
"""Remove an entry, making sure to maintain the invariants."""
112
cleanup = self._cleanup.pop(key)
113
val = self._cache.pop(key)
114
if cleanup is not None:
115
cleanup(key, val)
116
return val
117
118
def _remove_lru(self):
119
"""Remove one entry from the lru, and handle consequences.
120
121
If there are no more references to the lru, then this entry should be
122
removed from the cache.
123
"""
124
key = self._queue.popleft()
125
self._refcount[key] -= 1
126
if not self._refcount[key]:
127
del self._refcount[key]
128
self._remove(key)
129
130
def clear(self):
131
"""Clear out all of the cache."""
132
# Clean up in LRU order
133
while self._cache:
134
self._remove_lru()
135
136
137
class LRUSizeCache(LRUCache):
138
"""An LRUCache that removes things based on the size of the values.
139
140
This differs in that it doesn't care how many actual items there are,
141
it just restricts the cache to be cleaned up after so much data is stored.
142
143
The values that are added must support len(value).
144
"""
145
146
def __init__(self, max_size=1024*1024, after_cleanup_size=None,
147
compute_size=None):
148
"""Create a new LRUSizeCache.
149
150
:param max_size: The max number of bytes to store before we start
151
clearing out entries.
152
:param after_cleanup_size: After cleaning up, shrink everything to this
153
size.
154
:param compute_size: A function to compute the size of the values. We
155
use a function here, so that you can pass 'len' if you are just
156
using simple strings, or a more complex function if you are using
157
something like a list of strings, or even a custom object.
158
The function should take the form "compute_size(value) => integer".
159
If not supplied, it defaults to 'len()'
160
"""
161
# This approximates that texts are > 0.5k in size. It only really
162
# effects when we clean up the queue, so we don't want it to be too
163
# large.
164
LRUCache.__init__(self, max_cache=int(max_size/512))
165
self._max_size = max_size
166
if after_cleanup_size is None:
167
self._after_cleanup_size = self._max_size
168
else:
169
self._after_cleanup_size = min(after_cleanup_size, self._max_size)
170
171
self._value_size = 0
172
self._compute_size = compute_size
173
if compute_size is None:
174
self._compute_size = len
175
176
def add(self, key, value, cleanup=None):
177
"""Add a new value to the cache.
178
179
Also, if the entry is ever removed from the queue, call cleanup.
180
Passing it the key and value being removed.
181
182
:param key: The key to store it under
183
:param value: The object to store
184
:param cleanup: None or a function taking (key, value) to indicate
185
'value' sohuld be cleaned up.
186
"""
187
if key in self._cache:
188
self._remove(key)
189
value_len = self._compute_size(value)
190
if value_len >= self._after_cleanup_size:
191
return
192
self._value_size += value_len
193
self._cache[key] = value
194
self._cleanup[key] = cleanup
195
self._record_access(key)
196
197
if self._value_size > self._max_size:
198
# Time to cleanup
199
self.cleanup()
200
201
def cleanup(self):
202
"""Clear the cache until it shrinks to the requested size.
203
204
This does not completely wipe the cache, just makes sure it is under
205
the after_cleanup_size.
206
"""
207
# Make sure the cache is shrunk to the correct size
208
while self._value_size > self._after_cleanup_size:
209
self._remove_lru()
210
211
def _remove(self, key):
212
"""Remove an entry, making sure to maintain the invariants."""
213
val = LRUCache._remove(self, key)
214
self._value_size -= self._compute_size(val)
Loggerhead is a web-based interface for Breezy
Version: 2.0.1