google_hash 0.0.0
Sign up to get free protection for your applications and to get access to all the features.
- data/README +21 -0
- data/Rakefile +11 -0
- data/VERSION +1 -0
- data/ext/extconf.rb +15 -0
- data/ext/go.cpp +109 -0
- data/ext/sparsehash-1.5.2/AUTHORS +2 -0
- data/ext/sparsehash-1.5.2/COPYING +28 -0
- data/ext/sparsehash-1.5.2/ChangeLog +167 -0
- data/ext/sparsehash-1.5.2/INSTALL +236 -0
- data/ext/sparsehash-1.5.2/Makefile.am +157 -0
- data/ext/sparsehash-1.5.2/Makefile.in +1019 -0
- data/ext/sparsehash-1.5.2/NEWS +0 -0
- data/ext/sparsehash-1.5.2/README +149 -0
- data/ext/sparsehash-1.5.2/README.windows +25 -0
- data/ext/sparsehash-1.5.2/TODO +28 -0
- data/ext/sparsehash-1.5.2/aclocal.m4 +868 -0
- data/ext/sparsehash-1.5.2/compile +99 -0
- data/ext/sparsehash-1.5.2/config.guess +1516 -0
- data/ext/sparsehash-1.5.2/config.sub +1626 -0
- data/ext/sparsehash-1.5.2/configure +8054 -0
- data/ext/sparsehash-1.5.2/configure.ac +74 -0
- data/ext/sparsehash-1.5.2/depcomp +530 -0
- data/ext/sparsehash-1.5.2/doc/dense_hash_map.html +1591 -0
- data/ext/sparsehash-1.5.2/doc/dense_hash_set.html +1445 -0
- data/ext/sparsehash-1.5.2/doc/designstyle.css +115 -0
- data/ext/sparsehash-1.5.2/doc/implementation.html +365 -0
- data/ext/sparsehash-1.5.2/doc/index.html +69 -0
- data/ext/sparsehash-1.5.2/doc/performance.html +96 -0
- data/ext/sparsehash-1.5.2/doc/sparse_hash_map.html +1527 -0
- data/ext/sparsehash-1.5.2/doc/sparse_hash_set.html +1376 -0
- data/ext/sparsehash-1.5.2/doc/sparsetable.html +1393 -0
- data/ext/sparsehash-1.5.2/experimental/Makefile +9 -0
- data/ext/sparsehash-1.5.2/experimental/README +14 -0
- data/ext/sparsehash-1.5.2/experimental/example.c +54 -0
- data/ext/sparsehash-1.5.2/experimental/libchash.c +1537 -0
- data/ext/sparsehash-1.5.2/experimental/libchash.h +252 -0
- data/ext/sparsehash-1.5.2/google-sparsehash.sln +47 -0
- data/ext/sparsehash-1.5.2/install-sh +323 -0
- data/ext/sparsehash-1.5.2/m4/acx_pthread.m4 +363 -0
- data/ext/sparsehash-1.5.2/m4/google_namespace.m4 +42 -0
- data/ext/sparsehash-1.5.2/m4/namespaces.m4 +15 -0
- data/ext/sparsehash-1.5.2/m4/stl_hash.m4 +70 -0
- data/ext/sparsehash-1.5.2/m4/stl_hash_fun.m4 +36 -0
- data/ext/sparsehash-1.5.2/m4/stl_namespace.m4 +25 -0
- data/ext/sparsehash-1.5.2/missing +360 -0
- data/ext/sparsehash-1.5.2/mkinstalldirs +158 -0
- data/ext/sparsehash-1.5.2/packages/deb.sh +74 -0
- data/ext/sparsehash-1.5.2/packages/deb/README +7 -0
- data/ext/sparsehash-1.5.2/packages/deb/changelog +107 -0
- data/ext/sparsehash-1.5.2/packages/deb/compat +1 -0
- data/ext/sparsehash-1.5.2/packages/deb/control +17 -0
- data/ext/sparsehash-1.5.2/packages/deb/copyright +35 -0
- data/ext/sparsehash-1.5.2/packages/deb/docs +16 -0
- data/ext/sparsehash-1.5.2/packages/deb/rules +117 -0
- data/ext/sparsehash-1.5.2/packages/deb/sparsehash.dirs +2 -0
- data/ext/sparsehash-1.5.2/packages/deb/sparsehash.install +2 -0
- data/ext/sparsehash-1.5.2/packages/rpm.sh +86 -0
- data/ext/sparsehash-1.5.2/packages/rpm/rpm.spec +61 -0
- data/ext/sparsehash-1.5.2/src/config.h.in +131 -0
- data/ext/sparsehash-1.5.2/src/config.h.include +23 -0
- data/ext/sparsehash-1.5.2/src/google/dense_hash_map +310 -0
- data/ext/sparsehash-1.5.2/src/google/dense_hash_set +287 -0
- data/ext/sparsehash-1.5.2/src/google/sparse_hash_map +294 -0
- data/ext/sparsehash-1.5.2/src/google/sparse_hash_set +275 -0
- data/ext/sparsehash-1.5.2/src/google/sparsehash/densehashtable.h +1062 -0
- data/ext/sparsehash-1.5.2/src/google/sparsehash/sparsehashtable.h +1015 -0
- data/ext/sparsehash-1.5.2/src/google/sparsetable +1468 -0
- data/ext/sparsehash-1.5.2/src/google/type_traits.h +250 -0
- data/ext/sparsehash-1.5.2/src/hashtable_unittest.cc +1375 -0
- data/ext/sparsehash-1.5.2/src/simple_test.cc +103 -0
- data/ext/sparsehash-1.5.2/src/sparsetable_unittest.cc +696 -0
- data/ext/sparsehash-1.5.2/src/time_hash_map.cc +488 -0
- data/ext/sparsehash-1.5.2/src/type_traits_unittest.cc +492 -0
- data/ext/sparsehash-1.5.2/src/windows/config.h +149 -0
- data/ext/sparsehash-1.5.2/src/windows/google/sparsehash/sparseconfig.h +32 -0
- data/ext/sparsehash-1.5.2/src/windows/port.cc +63 -0
- data/ext/sparsehash-1.5.2/src/windows/port.h +81 -0
- data/ext/sparsehash-1.5.2/src/words +8944 -0
- data/ext/sparsehash-1.5.2/vsprojects/hashtable_unittest/hashtable_unittest.vcproj +187 -0
- data/ext/sparsehash-1.5.2/vsprojects/sparsetable_unittest/sparsetable_unittest.vcproj +172 -0
- data/ext/sparsehash-1.5.2/vsprojects/time_hash_map/time_hash_map.vcproj +187 -0
- data/ext/sparsehash-1.5.2/vsprojects/type_traits_unittest/type_traits_unittest.vcproj +169 -0
- data/ext/test.rb +10 -0
- data/test/spec.go +70 -0
- metadata +147 -0
@@ -0,0 +1,287 @@
|
|
1
|
+
// Copyright (c) 2005, Google Inc.
|
2
|
+
// All rights reserved.
|
3
|
+
//
|
4
|
+
// Redistribution and use in source and binary forms, with or without
|
5
|
+
// modification, are permitted provided that the following conditions are
|
6
|
+
// met:
|
7
|
+
//
|
8
|
+
// * Redistributions of source code must retain the above copyright
|
9
|
+
// notice, this list of conditions and the following disclaimer.
|
10
|
+
// * Redistributions in binary form must reproduce the above
|
11
|
+
// copyright notice, this list of conditions and the following disclaimer
|
12
|
+
// in the documentation and/or other materials provided with the
|
13
|
+
// distribution.
|
14
|
+
// * Neither the name of Google Inc. nor the names of its
|
15
|
+
// contributors may be used to endorse or promote products derived from
|
16
|
+
// this software without specific prior written permission.
|
17
|
+
//
|
18
|
+
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
|
19
|
+
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
|
20
|
+
// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
|
21
|
+
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
|
22
|
+
// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
|
23
|
+
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
|
24
|
+
// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
25
|
+
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
26
|
+
// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
27
|
+
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
28
|
+
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
29
|
+
|
30
|
+
// ---
|
31
|
+
// Author: Craig Silverstein
|
32
|
+
//
|
33
|
+
// This is just a very thin wrapper over densehashtable.h, just
|
34
|
+
// like sgi stl's stl_hash_set is a very thin wrapper over
|
35
|
+
// stl_hashtable. The major thing we define is operator[], because
|
36
|
+
// we have a concept of a data_type which stl_hashtable doesn't
|
37
|
+
// (it only has a key and a value).
|
38
|
+
//
|
39
|
+
// This is more different from dense_hash_map than you might think,
|
40
|
+
// because all iterators for sets are const (you obviously can't
|
41
|
+
// change the key, and for sets there is no value).
|
42
|
+
//
|
43
|
+
// NOTE: this is exactly like sparse_hash_set.h, with the word
|
44
|
+
// "sparse" replaced by "dense", except for the addition of
|
45
|
+
// set_empty_key().
|
46
|
+
//
|
47
|
+
// YOU MUST CALL SET_EMPTY_KEY() IMMEDIATELY AFTER CONSTRUCTION.
|
48
|
+
//
|
49
|
+
// Otherwise your program will die in mysterious ways.
|
50
|
+
//
|
51
|
+
// In other respects, we adhere mostly to the STL semantics for
|
52
|
+
// hash-set. One important exception is that insert() invalidates
|
53
|
+
// iterators entirely. On the plus side, though, erase() doesn't
|
54
|
+
// invalidate iterators at all, or even change the ordering of elements.
|
55
|
+
//
|
56
|
+
// Here are a few "power user" tips:
|
57
|
+
//
|
58
|
+
// 1) set_deleted_key():
|
59
|
+
// If you want to use erase() you must call set_deleted_key(),
|
60
|
+
// in addition to set_empty_key(), after construction.
|
61
|
+
// The deleted and empty keys must differ.
|
62
|
+
//
|
63
|
+
// 2) resize(0):
|
64
|
+
// When an item is deleted, its memory isn't freed right
|
65
|
+
// away. This allows you to iterate over a hashtable,
|
66
|
+
// and call erase(), without invalidating the iterator.
|
67
|
+
// To force the memory to be freed, call resize(0).
|
68
|
+
// For tr1 compatibility, this can also be called as rehash(0).
|
69
|
+
//
|
70
|
+
// 3) min_load_factor(0.0)
|
71
|
+
// Setting the minimum load factor to 0.0 guarantees that
|
72
|
+
// the hash table will never shrink.
|
73
|
+
//
|
74
|
+
// Guide to what kind of hash_set to use:
|
75
|
+
// (1) dense_hash_set: fastest, uses the most memory
|
76
|
+
// (2) sparse_hash_set: slowest, uses the least memory
|
77
|
+
// (3) hash_set (STL): in the middle
|
78
|
+
// Typically I use sparse_hash_set when I care about space and/or when
|
79
|
+
// I need to save the hashtable on disk. I use hash_set otherwise. I
|
80
|
+
// don't personally use dense_hash_set ever; some people use it for
|
81
|
+
// small sets with lots of lookups.
|
82
|
+
//
|
83
|
+
// - dense_hash_set has, typically, a factor of 2 memory overhead (if your
|
84
|
+
// data takes up X bytes, the hash_set uses X more bytes in overhead).
|
85
|
+
// - sparse_hash_set has about 2 bits overhead per entry.
|
86
|
+
// - sparse_hash_map can be 3-7 times slower than the others for lookup and,
|
87
|
+
// especially, inserts. See time_hash_map.cc for details.
|
88
|
+
//
|
89
|
+
// See /usr/(local/)?doc/sparsehash-*/dense_hash_set.html
|
90
|
+
// for information about how to use this class.
|
91
|
+
|
92
|
+
#ifndef _DENSE_HASH_SET_H_
|
93
|
+
#define _DENSE_HASH_SET_H_
|
94
|
+
|
95
|
+
#include <google/sparsehash/sparseconfig.h>
|
96
|
+
#include <stdio.h> // for FILE * in read()/write()
|
97
|
+
#include <algorithm> // for the default template args
|
98
|
+
#include <functional> // for equal_to
|
99
|
+
#include <memory> // for alloc<>
|
100
|
+
#include <utility> // for pair<>
|
101
|
+
#include HASH_FUN_H // defined in config.h
|
102
|
+
#include <google/sparsehash/densehashtable.h>
|
103
|
+
|
104
|
+
|
105
|
+
_START_GOOGLE_NAMESPACE_
|
106
|
+
|
107
|
+
using STL_NAMESPACE::pair;
|
108
|
+
|
109
|
+
template <class Value,
|
110
|
+
class HashFcn = SPARSEHASH_HASH<Value>, // defined in sparseconfig.h
|
111
|
+
class EqualKey = STL_NAMESPACE::equal_to<Value>,
|
112
|
+
class Alloc = STL_NAMESPACE::allocator<Value> >
|
113
|
+
class dense_hash_set {
|
114
|
+
private:
|
115
|
+
// Apparently identity is not stl-standard, so we define our own
|
116
|
+
struct Identity {
|
117
|
+
Value& operator()(Value& v) const { return v; }
|
118
|
+
const Value& operator()(const Value& v) const { return v; }
|
119
|
+
};
|
120
|
+
struct SetKey {
|
121
|
+
void operator()(Value* value, const Value& new_key) const {
|
122
|
+
*value = new_key;
|
123
|
+
}
|
124
|
+
};
|
125
|
+
|
126
|
+
// The actual data
|
127
|
+
typedef dense_hashtable<Value, Value, HashFcn,
|
128
|
+
Identity, SetKey, EqualKey, Alloc> ht;
|
129
|
+
ht rep;
|
130
|
+
|
131
|
+
public:
|
132
|
+
typedef typename ht::key_type key_type;
|
133
|
+
typedef typename ht::value_type value_type;
|
134
|
+
typedef typename ht::hasher hasher;
|
135
|
+
typedef typename ht::key_equal key_equal;
|
136
|
+
typedef Alloc allocator_type;
|
137
|
+
|
138
|
+
typedef typename ht::size_type size_type;
|
139
|
+
typedef typename ht::difference_type difference_type;
|
140
|
+
typedef typename ht::const_pointer pointer;
|
141
|
+
typedef typename ht::const_pointer const_pointer;
|
142
|
+
typedef typename ht::const_reference reference;
|
143
|
+
typedef typename ht::const_reference const_reference;
|
144
|
+
|
145
|
+
typedef typename ht::const_iterator iterator;
|
146
|
+
typedef typename ht::const_iterator const_iterator;
|
147
|
+
typedef typename ht::const_local_iterator local_iterator;
|
148
|
+
typedef typename ht::const_local_iterator const_local_iterator;
|
149
|
+
|
150
|
+
|
151
|
+
// Iterator functions -- recall all iterators are const
|
152
|
+
iterator begin() const { return rep.begin(); }
|
153
|
+
iterator end() const { return rep.end(); }
|
154
|
+
|
155
|
+
// These come from tr1's unordered_set. For us, a bucket has 0 or 1 elements.
|
156
|
+
local_iterator begin(size_type i) const { return rep.begin(i); }
|
157
|
+
local_iterator end(size_type i) const { return rep.end(i); }
|
158
|
+
|
159
|
+
|
160
|
+
// Accessor functions
|
161
|
+
hasher hash_funct() const { return rep.hash_funct(); }
|
162
|
+
key_equal key_eq() const { return rep.key_eq(); }
|
163
|
+
|
164
|
+
|
165
|
+
// Constructors
|
166
|
+
explicit dense_hash_set(size_type expected_max_items_in_table = 0,
|
167
|
+
const hasher& hf = hasher(),
|
168
|
+
const key_equal& eql = key_equal())
|
169
|
+
: rep(expected_max_items_in_table, hf, eql) { }
|
170
|
+
|
171
|
+
template <class InputIterator>
|
172
|
+
dense_hash_set(InputIterator f, InputIterator l,
|
173
|
+
size_type expected_max_items_in_table = 0,
|
174
|
+
const hasher& hf = hasher(),
|
175
|
+
const key_equal& eql = key_equal())
|
176
|
+
: rep(expected_max_items_in_table, hf, eql) {
|
177
|
+
rep.insert(f, l);
|
178
|
+
}
|
179
|
+
// We use the default copy constructor
|
180
|
+
// We use the default operator=()
|
181
|
+
// We use the default destructor
|
182
|
+
|
183
|
+
void clear() { rep.clear(); }
|
184
|
+
// This clears the hash set without resizing it down to the minimum
|
185
|
+
// bucket count, but rather keeps the number of buckets constant
|
186
|
+
void clear_no_resize() { rep.clear_no_resize(); }
|
187
|
+
void swap(dense_hash_set& hs) { rep.swap(hs.rep); }
|
188
|
+
|
189
|
+
|
190
|
+
// Functions concerning size
|
191
|
+
size_type size() const { return rep.size(); }
|
192
|
+
size_type max_size() const { return rep.max_size(); }
|
193
|
+
bool empty() const { return rep.empty(); }
|
194
|
+
size_type bucket_count() const { return rep.bucket_count(); }
|
195
|
+
size_type max_bucket_count() const { return rep.max_bucket_count(); }
|
196
|
+
|
197
|
+
// These are tr1 methods. bucket() is the bucket the key is or would be in.
|
198
|
+
size_type bucket_size(size_type i) const { return rep.bucket_size(i); }
|
199
|
+
size_type bucket(const key_type& key) const { return rep.bucket(key); }
|
200
|
+
float load_factor() const {
|
201
|
+
return size() * 1.0f / bucket_count();
|
202
|
+
}
|
203
|
+
float max_load_factor() const {
|
204
|
+
float shrink, grow;
|
205
|
+
rep.get_resizing_parameters(&shrink, &grow);
|
206
|
+
return grow;
|
207
|
+
}
|
208
|
+
void max_load_factor(float new_grow) {
|
209
|
+
float shrink, grow;
|
210
|
+
rep.get_resizing_parameters(&shrink, &grow);
|
211
|
+
rep.set_resizing_parameters(shrink, new_grow);
|
212
|
+
}
|
213
|
+
// These aren't tr1 methods but perhaps ought to be.
|
214
|
+
float min_load_factor() const {
|
215
|
+
float shrink, grow;
|
216
|
+
rep.get_resizing_parameters(&shrink, &grow);
|
217
|
+
return shrink;
|
218
|
+
}
|
219
|
+
void min_load_factor(float new_shrink) {
|
220
|
+
float shrink, grow;
|
221
|
+
rep.get_resizing_parameters(&shrink, &grow);
|
222
|
+
rep.set_resizing_parameters(new_shrink, grow);
|
223
|
+
}
|
224
|
+
// Deprecated; use min_load_factor() or max_load_factor() instead.
|
225
|
+
void set_resizing_parameters(float shrink, float grow) {
|
226
|
+
return rep.set_resizing_parameters(shrink, grow);
|
227
|
+
}
|
228
|
+
|
229
|
+
void resize(size_type hint) { rep.resize(hint); }
|
230
|
+
void rehash(size_type hint) { resize(hint); } // the tr1 name
|
231
|
+
|
232
|
+
// Lookup routines
|
233
|
+
iterator find(const key_type& key) const { return rep.find(key); }
|
234
|
+
|
235
|
+
size_type count(const key_type& key) const { return rep.count(key); }
|
236
|
+
|
237
|
+
pair<iterator, iterator> equal_range(const key_type& key) const {
|
238
|
+
return rep.equal_range(key);
|
239
|
+
}
|
240
|
+
|
241
|
+
// Insertion routines
|
242
|
+
pair<iterator, bool> insert(const value_type& obj) {
|
243
|
+
pair<typename ht::iterator, bool> p = rep.insert(obj);
|
244
|
+
return pair<iterator, bool>(p.first, p.second); // const to non-const
|
245
|
+
}
|
246
|
+
template <class InputIterator>
|
247
|
+
void insert(InputIterator f, InputIterator l) { rep.insert(f, l); }
|
248
|
+
void insert(const_iterator f, const_iterator l) { rep.insert(f, l); }
|
249
|
+
// required for std::insert_iterator; the passed-in iterator is ignored
|
250
|
+
iterator insert(iterator, const value_type& obj) { return insert(obj).first; }
|
251
|
+
|
252
|
+
|
253
|
+
// Deletion and empty routines
|
254
|
+
// THESE ARE NON-STANDARD! I make you specify an "impossible" key
|
255
|
+
// value to identify deleted and empty buckets. You can change the
|
256
|
+
// deleted key as time goes on, or get rid of it entirely to be insert-only.
|
257
|
+
void set_empty_key(const key_type& key) { rep.set_empty_key(key); }
|
258
|
+
void set_deleted_key(const key_type& key) { rep.set_deleted_key(key); }
|
259
|
+
void clear_deleted_key() { rep.clear_deleted_key(); }
|
260
|
+
|
261
|
+
// These are standard
|
262
|
+
size_type erase(const key_type& key) { return rep.erase(key); }
|
263
|
+
void erase(iterator it) { rep.erase(it); }
|
264
|
+
void erase(iterator f, iterator l) { rep.erase(f, l); }
|
265
|
+
|
266
|
+
|
267
|
+
// Comparison
|
268
|
+
bool operator==(const dense_hash_set& hs) const { return rep == hs.rep; }
|
269
|
+
bool operator!=(const dense_hash_set& hs) const { return rep != hs.rep; }
|
270
|
+
|
271
|
+
|
272
|
+
// I/O -- this is an add-on for writing metainformation to disk
|
273
|
+
bool write_metadata(FILE *fp) { return rep.write_metadata(fp); }
|
274
|
+
bool read_metadata(FILE *fp) { return rep.read_metadata(fp); }
|
275
|
+
bool write_nopointer_data(FILE *fp) { return rep.write_nopointer_data(fp); }
|
276
|
+
bool read_nopointer_data(FILE *fp) { return rep.read_nopointer_data(fp); }
|
277
|
+
};
|
278
|
+
|
279
|
+
template <class Val, class HashFcn, class EqualKey, class Alloc>
|
280
|
+
inline void swap(dense_hash_set<Val, HashFcn, EqualKey, Alloc>& hs1,
|
281
|
+
dense_hash_set<Val, HashFcn, EqualKey, Alloc>& hs2) {
|
282
|
+
hs1.swap(hs2);
|
283
|
+
}
|
284
|
+
|
285
|
+
_END_GOOGLE_NAMESPACE_
|
286
|
+
|
287
|
+
#endif /* _DENSE_HASH_SET_H_ */
|
@@ -0,0 +1,294 @@
|
|
1
|
+
// Copyright (c) 2005, Google Inc.
|
2
|
+
// All rights reserved.
|
3
|
+
//
|
4
|
+
// Redistribution and use in source and binary forms, with or without
|
5
|
+
// modification, are permitted provided that the following conditions are
|
6
|
+
// met:
|
7
|
+
//
|
8
|
+
// * Redistributions of source code must retain the above copyright
|
9
|
+
// notice, this list of conditions and the following disclaimer.
|
10
|
+
// * Redistributions in binary form must reproduce the above
|
11
|
+
// copyright notice, this list of conditions and the following disclaimer
|
12
|
+
// in the documentation and/or other materials provided with the
|
13
|
+
// distribution.
|
14
|
+
// * Neither the name of Google Inc. nor the names of its
|
15
|
+
// contributors may be used to endorse or promote products derived from
|
16
|
+
// this software without specific prior written permission.
|
17
|
+
//
|
18
|
+
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
|
19
|
+
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
|
20
|
+
// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
|
21
|
+
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
|
22
|
+
// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
|
23
|
+
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
|
24
|
+
// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
25
|
+
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
26
|
+
// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
27
|
+
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
28
|
+
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
29
|
+
|
30
|
+
// ---
|
31
|
+
// Author: Craig Silverstein
|
32
|
+
//
|
33
|
+
// This is just a very thin wrapper over sparsehashtable.h, just
|
34
|
+
// like sgi stl's stl_hash_map is a very thin wrapper over
|
35
|
+
// stl_hashtable. The major thing we define is operator[], because
|
36
|
+
// we have a concept of a data_type which stl_hashtable doesn't
|
37
|
+
// (it only has a key and a value).
|
38
|
+
//
|
39
|
+
// We adhere mostly to the STL semantics for hash-map. One important
|
40
|
+
// exception is that insert() invalidates iterators entirely. On the
|
41
|
+
// plus side, though, delete() doesn't invalidate iterators at all, or
|
42
|
+
// even change the ordering of elements.
|
43
|
+
//
|
44
|
+
// Here are a few "power user" tips:
|
45
|
+
//
|
46
|
+
// 1) set_deleted_key():
|
47
|
+
// Unlike STL's hash_map, if you want to use erase() you
|
48
|
+
// *must* call set_deleted_key() after construction.
|
49
|
+
//
|
50
|
+
// 2) resize(0):
|
51
|
+
// When an item is deleted, its memory isn't freed right
|
52
|
+
// away. This is what allows you to iterate over a hashtable
|
53
|
+
// and call erase() without invalidating the iterator.
|
54
|
+
// To force the memory to be freed, call resize(0).
|
55
|
+
// For tr1 compatibility, this can also be called as rehash(0).
|
56
|
+
//
|
57
|
+
// 3) min_load_factor(0.0)
|
58
|
+
// Setting the minimum load factor to 0.0 guarantees that
|
59
|
+
// the hash table will never shrink.
|
60
|
+
//
|
61
|
+
// Guide to what kind of hash_map to use:
|
62
|
+
// (1) dense_hash_map: fastest, uses the most memory
|
63
|
+
// (2) sparse_hash_map: slowest, uses the least memory
|
64
|
+
// (3) hash_map / unordered_map (STL): in the middle
|
65
|
+
// Typically I use sparse_hash_map when I care about space and/or when
|
66
|
+
// I need to save the hashtable on disk. I use hash_map otherwise. I
|
67
|
+
// don't personally use dense_hash_map ever; some people use it for
|
68
|
+
// small maps with lots of lookups.
|
69
|
+
//
|
70
|
+
// - dense_hash_map has, typically, a factor of 2 memory overhead (if your
|
71
|
+
// data takes up X bytes, the hash_map uses X more bytes in overhead).
|
72
|
+
// - sparse_hash_map has about 2 bits overhead per entry.
|
73
|
+
// - sparse_hash_map can be 3-7 times slower than the others for lookup and,
|
74
|
+
// especially, inserts. See time_hash_map.cc for details.
|
75
|
+
//
|
76
|
+
// See /usr/(local/)?doc/sparsehash-*/sparse_hash_map.html
|
77
|
+
// for information about how to use this class.
|
78
|
+
|
79
|
+
#ifndef _SPARSE_HASH_MAP_H_
|
80
|
+
#define _SPARSE_HASH_MAP_H_
|
81
|
+
|
82
|
+
#include <google/sparsehash/sparseconfig.h>
|
83
|
+
#include <stdio.h> // for FILE * in read()/write()
|
84
|
+
#include <algorithm> // for the default template args
|
85
|
+
#include <functional> // for equal_to
|
86
|
+
#include <memory> // for alloc<>
|
87
|
+
#include <utility> // for pair<>
|
88
|
+
#include HASH_FUN_H // defined in config.h
|
89
|
+
#include <google/sparsehash/sparsehashtable.h>
|
90
|
+
|
91
|
+
|
92
|
+
_START_GOOGLE_NAMESPACE_
|
93
|
+
|
94
|
+
using STL_NAMESPACE::pair;
|
95
|
+
|
96
|
+
template <class Key, class T,
|
97
|
+
class HashFcn = SPARSEHASH_HASH<Key>, // defined in sparseconfig.h
|
98
|
+
class EqualKey = STL_NAMESPACE::equal_to<Key>,
|
99
|
+
class Alloc = STL_NAMESPACE::allocator<T> >
|
100
|
+
class sparse_hash_map {
|
101
|
+
private:
|
102
|
+
// Apparently select1st is not stl-standard, so we define our own
|
103
|
+
struct SelectKey {
|
104
|
+
const Key& operator()(const pair<const Key, T>& p) const {
|
105
|
+
return p.first;
|
106
|
+
}
|
107
|
+
};
|
108
|
+
struct SetKey {
|
109
|
+
void operator()(pair<const Key, T>* value, const Key& new_key) const {
|
110
|
+
*const_cast<Key*>(&value->first) = new_key;
|
111
|
+
// It would be nice to clear the rest of value here as well, in
|
112
|
+
// case it's taking up a lot of memory. We do this by clearing
|
113
|
+
// the value. This assumes T has a zero-arg constructor!
|
114
|
+
value->second = T();
|
115
|
+
}
|
116
|
+
};
|
117
|
+
|
118
|
+
// The actual data
|
119
|
+
typedef sparse_hashtable<pair<const Key, T>, Key, HashFcn,
|
120
|
+
SelectKey, SetKey, EqualKey, Alloc> ht;
|
121
|
+
ht rep;
|
122
|
+
|
123
|
+
public:
|
124
|
+
typedef typename ht::key_type key_type;
|
125
|
+
typedef T data_type;
|
126
|
+
typedef T mapped_type;
|
127
|
+
typedef typename ht::value_type value_type;
|
128
|
+
typedef typename ht::hasher hasher;
|
129
|
+
typedef typename ht::key_equal key_equal;
|
130
|
+
typedef Alloc allocator_type;
|
131
|
+
|
132
|
+
typedef typename ht::size_type size_type;
|
133
|
+
typedef typename ht::difference_type difference_type;
|
134
|
+
typedef typename ht::pointer pointer;
|
135
|
+
typedef typename ht::const_pointer const_pointer;
|
136
|
+
typedef typename ht::reference reference;
|
137
|
+
typedef typename ht::const_reference const_reference;
|
138
|
+
|
139
|
+
typedef typename ht::iterator iterator;
|
140
|
+
typedef typename ht::const_iterator const_iterator;
|
141
|
+
typedef typename ht::local_iterator local_iterator;
|
142
|
+
typedef typename ht::const_local_iterator const_local_iterator;
|
143
|
+
|
144
|
+
// Iterator functions
|
145
|
+
iterator begin() { return rep.begin(); }
|
146
|
+
iterator end() { return rep.end(); }
|
147
|
+
const_iterator begin() const { return rep.begin(); }
|
148
|
+
const_iterator end() const { return rep.end(); }
|
149
|
+
|
150
|
+
// These come from tr1's unordered_map. For us, a bucket has 0 or 1 elements.
|
151
|
+
local_iterator begin(size_type i) { return rep.begin(i); }
|
152
|
+
local_iterator end(size_type i) { return rep.end(i); }
|
153
|
+
const_local_iterator begin(size_type i) const { return rep.begin(i); }
|
154
|
+
const_local_iterator end(size_type i) const { return rep.end(i); }
|
155
|
+
|
156
|
+
// Accessor functions
|
157
|
+
// TODO(csilvers): implement Alloc get_allocator() const;
|
158
|
+
hasher hash_funct() const { return rep.hash_funct(); }
|
159
|
+
hasher hash_function() const { return hash_funct(); }
|
160
|
+
key_equal key_eq() const { return rep.key_eq(); }
|
161
|
+
|
162
|
+
|
163
|
+
// Constructors
|
164
|
+
explicit sparse_hash_map(size_type expected_max_items_in_table = 0,
|
165
|
+
const hasher& hf = hasher(),
|
166
|
+
const key_equal& eql = key_equal())
|
167
|
+
: rep(expected_max_items_in_table, hf, eql) { }
|
168
|
+
|
169
|
+
template <class InputIterator>
|
170
|
+
sparse_hash_map(InputIterator f, InputIterator l,
|
171
|
+
size_type expected_max_items_in_table = 0,
|
172
|
+
const hasher& hf = hasher(),
|
173
|
+
const key_equal& eql = key_equal())
|
174
|
+
: rep(expected_max_items_in_table, hf, eql) {
|
175
|
+
rep.insert(f, l);
|
176
|
+
}
|
177
|
+
// We use the default copy constructor
|
178
|
+
// We use the default operator=()
|
179
|
+
// We use the default destructor
|
180
|
+
|
181
|
+
void clear() { rep.clear(); }
|
182
|
+
void swap(sparse_hash_map& hs) { rep.swap(hs.rep); }
|
183
|
+
|
184
|
+
|
185
|
+
// Functions concerning size
|
186
|
+
size_type size() const { return rep.size(); }
|
187
|
+
size_type max_size() const { return rep.max_size(); }
|
188
|
+
bool empty() const { return rep.empty(); }
|
189
|
+
size_type bucket_count() const { return rep.bucket_count(); }
|
190
|
+
size_type max_bucket_count() const { return rep.max_bucket_count(); }
|
191
|
+
|
192
|
+
// These are tr1 methods. bucket() is the bucket the key is or would be in.
|
193
|
+
size_type bucket_size(size_type i) const { return rep.bucket_size(i); }
|
194
|
+
size_type bucket(const key_type& key) const { return rep.bucket(key); }
|
195
|
+
float load_factor() const {
|
196
|
+
return size() * 1.0f / bucket_count();
|
197
|
+
}
|
198
|
+
float max_load_factor() const {
|
199
|
+
float shrink, grow;
|
200
|
+
rep.get_resizing_parameters(&shrink, &grow);
|
201
|
+
return grow;
|
202
|
+
}
|
203
|
+
void max_load_factor(float new_grow) {
|
204
|
+
float shrink, grow;
|
205
|
+
rep.get_resizing_parameters(&shrink, &grow);
|
206
|
+
rep.set_resizing_parameters(shrink, new_grow);
|
207
|
+
}
|
208
|
+
// These aren't tr1 methods but perhaps ought to be.
|
209
|
+
float min_load_factor() const {
|
210
|
+
float shrink, grow;
|
211
|
+
rep.get_resizing_parameters(&shrink, &grow);
|
212
|
+
return shrink;
|
213
|
+
}
|
214
|
+
void min_load_factor(float new_shrink) {
|
215
|
+
float shrink, grow;
|
216
|
+
rep.get_resizing_parameters(&shrink, &grow);
|
217
|
+
rep.set_resizing_parameters(new_shrink, grow);
|
218
|
+
}
|
219
|
+
// Deprecated; use min_load_factor() or max_load_factor() instead.
|
220
|
+
void set_resizing_parameters(float shrink, float grow) {
|
221
|
+
return rep.set_resizing_parameters(shrink, grow);
|
222
|
+
}
|
223
|
+
|
224
|
+
void resize(size_type hint) { rep.resize(hint); }
|
225
|
+
void rehash(size_type hint) { resize(hint); } // the tr1 name
|
226
|
+
|
227
|
+
// Lookup routines
|
228
|
+
iterator find(const key_type& key) { return rep.find(key); }
|
229
|
+
const_iterator find(const key_type& key) const { return rep.find(key); }
|
230
|
+
|
231
|
+
data_type& operator[](const key_type& key) { // This is our value-add!
|
232
|
+
iterator it = find(key);
|
233
|
+
if (it != end()) {
|
234
|
+
return it->second;
|
235
|
+
} else {
|
236
|
+
return insert(value_type(key, data_type())).first->second;
|
237
|
+
}
|
238
|
+
}
|
239
|
+
|
240
|
+
size_type count(const key_type& key) const { return rep.count(key); }
|
241
|
+
|
242
|
+
pair<iterator, iterator> equal_range(const key_type& key) {
|
243
|
+
return rep.equal_range(key);
|
244
|
+
}
|
245
|
+
pair<const_iterator, const_iterator> equal_range(const key_type& key) const {
|
246
|
+
return rep.equal_range(key);
|
247
|
+
}
|
248
|
+
|
249
|
+
// Insertion routines
|
250
|
+
pair<iterator, bool> insert(const value_type& obj) { return rep.insert(obj); }
|
251
|
+
template <class InputIterator>
|
252
|
+
void insert(InputIterator f, InputIterator l) { rep.insert(f, l); }
|
253
|
+
void insert(const_iterator f, const_iterator l) { rep.insert(f, l); }
|
254
|
+
// required for std::insert_iterator; the passed-in iterator is ignored
|
255
|
+
iterator insert(iterator, const value_type& obj) { return insert(obj).first; }
|
256
|
+
|
257
|
+
|
258
|
+
// Deletion routines
|
259
|
+
// THESE ARE NON-STANDARD! I make you specify an "impossible" key
|
260
|
+
// value to identify deleted buckets. You can change the key as
|
261
|
+
// time goes on, or get rid of it entirely to be insert-only.
|
262
|
+
void set_deleted_key(const key_type& key) {
|
263
|
+
rep.set_deleted_key(key);
|
264
|
+
}
|
265
|
+
void clear_deleted_key() { rep.clear_deleted_key(); }
|
266
|
+
|
267
|
+
// These are standard
|
268
|
+
size_type erase(const key_type& key) { return rep.erase(key); }
|
269
|
+
void erase(iterator it) { rep.erase(it); }
|
270
|
+
void erase(iterator f, iterator l) { rep.erase(f, l); }
|
271
|
+
|
272
|
+
|
273
|
+
// Comparison
|
274
|
+
bool operator==(const sparse_hash_map& hs) const { return rep == hs.rep; }
|
275
|
+
bool operator!=(const sparse_hash_map& hs) const { return rep != hs.rep; }
|
276
|
+
|
277
|
+
|
278
|
+
// I/O -- this is an add-on for writing metainformation to disk
|
279
|
+
bool write_metadata(FILE *fp) { return rep.write_metadata(fp); }
|
280
|
+
bool read_metadata(FILE *fp) { return rep.read_metadata(fp); }
|
281
|
+
bool write_nopointer_data(FILE *fp) { return rep.write_nopointer_data(fp); }
|
282
|
+
bool read_nopointer_data(FILE *fp) { return rep.read_nopointer_data(fp); }
|
283
|
+
};
|
284
|
+
|
285
|
+
// We need a global swap as well
|
286
|
+
template <class Key, class T, class HashFcn, class EqualKey, class Alloc>
|
287
|
+
inline void swap(sparse_hash_map<Key, T, HashFcn, EqualKey, Alloc>& hm1,
|
288
|
+
sparse_hash_map<Key, T, HashFcn, EqualKey, Alloc>& hm2) {
|
289
|
+
hm1.swap(hm2);
|
290
|
+
}
|
291
|
+
|
292
|
+
_END_GOOGLE_NAMESPACE_
|
293
|
+
|
294
|
+
#endif /* _SPARSE_HASH_MAP_H_ */
|