bloom-ruby 0.2.1
Sign up to get free protection for your applications and to get access to all the features.
- data/CHANGELOG +15 -0
- data/README.md +67 -0
- data/ext/bloom-filter.c +265 -0
- data/ext/bloom-filter.h +193 -0
- data/ext/bloom_filter.c +248 -0
- data/ext/extconf.rb +4 -0
- data/ext/hash-string.c +58 -0
- data/ext/hash-string.h +56 -0
- data/ext/version.h +1 -0
- data/lib/bloom-filter.rb +1 -0
- data/test/helper.rb +4 -0
- data/test/test_basic.rb +29 -0
- data/test/test_io.rb +27 -0
- metadata +91 -0
data/CHANGELOG
ADDED
data/README.md
ADDED
@@ -0,0 +1,67 @@
|
|
1
|
+
# BloomFilter
|
2
|
+
|
3
|
+
BloomFilter is a ruby library that implements an in-memory [Bloom Filter](http://en.wikipedia.org/wiki/Bloom_filter)
|
4
|
+
|
5
|
+
## Dependencies
|
6
|
+
|
7
|
+
* ruby 1.9.1 or later
|
8
|
+
|
9
|
+
## Installation
|
10
|
+
|
11
|
+
```
|
12
|
+
gem install bloom-filter
|
13
|
+
```
|
14
|
+
|
15
|
+
## API
|
16
|
+
|
17
|
+
```
|
18
|
+
BloomFilter
|
19
|
+
.new
|
20
|
+
.load
|
21
|
+
#dump
|
22
|
+
#insert
|
23
|
+
#include?
|
24
|
+
#bits
|
25
|
+
#binary
|
26
|
+
|
27
|
+
```
|
28
|
+
|
29
|
+
## Usage
|
30
|
+
|
31
|
+
```ruby
|
32
|
+
require 'bloom-filter'
|
33
|
+
|
34
|
+
filter = BloomFilter.new
|
35
|
+
|
36
|
+
# auto-calculate optimum bitmap size based on maximum number of items stored and desired max error rate.
|
37
|
+
filter = BloomFilter.new size: 100_000, error_rate: 0.01
|
38
|
+
|
39
|
+
# specify bitmap size & number of hash functions explicitly.
|
40
|
+
filter = BloomFilter.new bits: 100_000_0, hashes: 4
|
41
|
+
|
42
|
+
filter.insert "foo"
|
43
|
+
filter.include? "foo" #=> true
|
44
|
+
filter.include? "bar" #=> false
|
45
|
+
|
46
|
+
filter.dump "/tmp/random.bloom"
|
47
|
+
filter = BloomFilter.load "/tmp/random.bloom"
|
48
|
+
|
49
|
+
bits = filter.bits #=> "10010100100111..."
|
50
|
+
binary = filter.binary #=> "\x83Ö\xAC\xEA\u00..."
|
51
|
+
|
52
|
+
filter2 = BloomFilter.new bits: 100_000_0, hashes: 4
|
53
|
+
filter2.binary = binary
|
54
|
+
filter2.include? "foo" #=> true
|
55
|
+
```
|
56
|
+
|
57
|
+
## See Also
|
58
|
+
|
59
|
+
[https://github.com/igrigorik/bloomfilter-rb](https://github.com/igrigorik/bloomfilter-rb)
|
60
|
+
|
61
|
+
## Home Page
|
62
|
+
|
63
|
+
[https://github.com/deepfryed/bloom-filter](https://github.com/deepfryed/bloom-filter)
|
64
|
+
|
65
|
+
## License
|
66
|
+
|
67
|
+
MIT
|
data/ext/bloom-filter.c
ADDED
@@ -0,0 +1,265 @@
|
|
1
|
+
/*
|
2
|
+
|
3
|
+
Copyright (c) 2005-2008, Simon Howard
|
4
|
+
|
5
|
+
Permission to use, copy, modify, and/or distribute this software
|
6
|
+
for any purpose with or without fee is hereby granted, provided
|
7
|
+
that the above copyright notice and this permission notice appear
|
8
|
+
in all copies.
|
9
|
+
|
10
|
+
THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL
|
11
|
+
WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED
|
12
|
+
WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE
|
13
|
+
AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR
|
14
|
+
CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM
|
15
|
+
LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT,
|
16
|
+
NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
|
17
|
+
CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
|
18
|
+
|
19
|
+
*/
|
20
|
+
|
21
|
+
#include <stdlib.h>
|
22
|
+
#include <string.h>
|
23
|
+
|
24
|
+
#include "bloom-filter.h"
|
25
|
+
|
26
|
+
/* malloc() / free() testing */
|
27
|
+
|
28
|
+
#ifdef ALLOC_TESTING
|
29
|
+
#include "alloc-testing.h"
|
30
|
+
#endif
|
31
|
+
|
32
|
+
/* Salt values. These salts are XORed with the output of the hash
|
33
|
+
* function to give multiple unique hashes. */
|
34
|
+
|
35
|
+
static const unsigned int salts[] = {
|
36
|
+
0x5cee4612, 0xb5587b1c, 0xa250f2b0, 0xa3bf6d2a,
|
37
|
+
0x7a81bd1a, 0x92888d7f, 0x1dc977c7, 0xedc96624,
|
38
|
+
0x920c85d9, 0xf16066b3, 0xc6f0d4b3, 0x2b76eb86,
|
39
|
+
0xcacb3893, 0x493d81c5, 0xf5a133ac, 0x039740bf,
|
40
|
+
0x162b8224, 0xf841de90, 0xc3e5090d, 0x3bce93a7,
|
41
|
+
0xf1860334, 0xe832b5f1, 0xf5b6535b, 0xe4cf4fa6,
|
42
|
+
0x8357b769, 0x1442b07a, 0x21c5863d, 0xabc0d846,
|
43
|
+
0x6dc0d77a, 0x23a3992c, 0xe12179ba, 0xd81d1e23,
|
44
|
+
0xcff4727b, 0xe957ecfb, 0xee8f391a, 0x426efa23,
|
45
|
+
0x3a34ff2c, 0x8b875d94, 0x34fd0f63, 0xf159daae,
|
46
|
+
0xaabab8b3, 0xa83a07ba, 0x4e54fb33, 0xfb82fab8,
|
47
|
+
0x2ae2888f, 0xd1a307a8, 0xbe33322d, 0x87c73f86,
|
48
|
+
0x7270fa7e, 0x68673c55, 0x2c8026d0, 0xead8e422,
|
49
|
+
0xa3ee5132, 0xecb67767, 0x1c3b1ae5, 0x47adf5b6,
|
50
|
+
0xf4518d30, 0x46e62797, 0x9889aa76, 0x1405aadf,
|
51
|
+
0xf62f9124, 0x5c435ac5, 0x35b8dfe3, 0x651c08c5,
|
52
|
+
};
|
53
|
+
|
54
|
+
BloomFilter *bloom_filter_new (unsigned int table_size, BloomFilterHashFunc hash_func, unsigned int num_functions) {
|
55
|
+
BloomFilter *filter;
|
56
|
+
|
57
|
+
/* There is a limit on the number of functions which can be
|
58
|
+
* applied, due to the table size */
|
59
|
+
|
60
|
+
if (num_functions > sizeof (salts) / sizeof (*salts)) {
|
61
|
+
return NULL;
|
62
|
+
}
|
63
|
+
|
64
|
+
/* Allocate bloom filter structure */
|
65
|
+
|
66
|
+
filter = malloc (sizeof (BloomFilter));
|
67
|
+
|
68
|
+
if (filter == NULL) {
|
69
|
+
return NULL;
|
70
|
+
}
|
71
|
+
|
72
|
+
/* Allocate table, each entry is one bit; these are packed into
|
73
|
+
* bytes. When allocating we must round the length up to the nearest
|
74
|
+
* byte. */
|
75
|
+
|
76
|
+
filter->table = calloc ((table_size + 7) / 8, 1);
|
77
|
+
|
78
|
+
if (filter->table == NULL) {
|
79
|
+
free (filter);
|
80
|
+
return NULL;
|
81
|
+
}
|
82
|
+
|
83
|
+
filter->hash_func = hash_func;
|
84
|
+
filter->num_functions = num_functions;
|
85
|
+
filter->table_size = table_size;
|
86
|
+
|
87
|
+
return filter;
|
88
|
+
}
|
89
|
+
|
90
|
+
void bloom_filter_free (BloomFilter * bloomfilter) {
|
91
|
+
free (bloomfilter->table);
|
92
|
+
free (bloomfilter);
|
93
|
+
}
|
94
|
+
|
95
|
+
void bloom_filter_insert (BloomFilter * bloomfilter, BloomFilterValue value) {
|
96
|
+
unsigned long hash;
|
97
|
+
unsigned long subhash;
|
98
|
+
unsigned int index;
|
99
|
+
unsigned int i;
|
100
|
+
|
101
|
+
/* Generate hash of the value to insert */
|
102
|
+
|
103
|
+
hash = bloomfilter->hash_func (value);
|
104
|
+
|
105
|
+
/* Generate multiple unique hashes by XORing with values in the
|
106
|
+
* salt table. */
|
107
|
+
|
108
|
+
for (i = 0; i < bloomfilter->num_functions; ++i) {
|
109
|
+
|
110
|
+
/* Generate a unique hash */
|
111
|
+
|
112
|
+
subhash = hash ^ salts[i];
|
113
|
+
|
114
|
+
/* Find the index into the table */
|
115
|
+
|
116
|
+
index = subhash % bloomfilter->table_size;
|
117
|
+
|
118
|
+
/* Insert into the table.
|
119
|
+
* index / 8 finds the byte index of the table,
|
120
|
+
* index % 8 gives the bit index within that byte to set. */
|
121
|
+
|
122
|
+
bloomfilter->table[index / 8] |= 1 << (index % 8);
|
123
|
+
}
|
124
|
+
}
|
125
|
+
|
126
|
+
int bloom_filter_query (BloomFilter * bloomfilter, BloomFilterValue value) {
|
127
|
+
unsigned long hash;
|
128
|
+
unsigned long subhash;
|
129
|
+
unsigned int index;
|
130
|
+
unsigned int i;
|
131
|
+
unsigned char b;
|
132
|
+
int bit;
|
133
|
+
|
134
|
+
/* Generate hash of the value to lookup */
|
135
|
+
|
136
|
+
hash = bloomfilter->hash_func (value);
|
137
|
+
|
138
|
+
/* Generate multiple unique hashes by XORing with values in the
|
139
|
+
* salt table. */
|
140
|
+
|
141
|
+
for (i = 0; i < bloomfilter->num_functions; ++i) {
|
142
|
+
|
143
|
+
/* Generate a unique hash */
|
144
|
+
|
145
|
+
subhash = hash ^ salts[i];
|
146
|
+
|
147
|
+
/* Find the index into the table to test */
|
148
|
+
|
149
|
+
index = subhash % bloomfilter->table_size;
|
150
|
+
|
151
|
+
/* The byte at index / 8 holds the value to test */
|
152
|
+
|
153
|
+
b = bloomfilter->table[index / 8];
|
154
|
+
bit = 1 << (index % 8);
|
155
|
+
|
156
|
+
/* Test if the particular bit is set; if it is not set,
|
157
|
+
* this value can not have been inserted. */
|
158
|
+
|
159
|
+
if ((b & bit) == 0) {
|
160
|
+
return 0;
|
161
|
+
}
|
162
|
+
}
|
163
|
+
|
164
|
+
/* All necessary bits were set. This may indicate that the value
|
165
|
+
* was inserted, or the values could have been set through other
|
166
|
+
* insertions. */
|
167
|
+
|
168
|
+
return 1;
|
169
|
+
}
|
170
|
+
|
171
|
+
void bloom_filter_read (BloomFilter * bloomfilter, unsigned char *array) {
|
172
|
+
unsigned int array_size;
|
173
|
+
|
174
|
+
/* The table is an array of bits, packed into bytes. Round up
|
175
|
+
* to the nearest byte. */
|
176
|
+
|
177
|
+
array_size = (bloomfilter->table_size + 7) / 8;
|
178
|
+
|
179
|
+
/* Copy into the buffer of the calling routine. */
|
180
|
+
|
181
|
+
memcpy (array, bloomfilter->table, array_size);
|
182
|
+
}
|
183
|
+
|
184
|
+
void bloom_filter_load (BloomFilter * bloomfilter, unsigned char *array) {
|
185
|
+
unsigned int array_size;
|
186
|
+
|
187
|
+
/* The table is an array of bits, packed into bytes. Round up
|
188
|
+
* to the nearest byte. */
|
189
|
+
|
190
|
+
array_size = (bloomfilter->table_size + 7) / 8;
|
191
|
+
|
192
|
+
/* Copy from the buffer of the calling routine. */
|
193
|
+
|
194
|
+
memcpy (bloomfilter->table, array, array_size);
|
195
|
+
}
|
196
|
+
|
197
|
+
BloomFilter *bloom_filter_union (BloomFilter * filter1, BloomFilter * filter2) {
|
198
|
+
BloomFilter *result;
|
199
|
+
unsigned int i;
|
200
|
+
unsigned int array_size;
|
201
|
+
|
202
|
+
/* To perform this operation, both filters must be created with
|
203
|
+
* the same values. */
|
204
|
+
|
205
|
+
if (filter1->table_size != filter2->table_size
|
206
|
+
|| filter1->num_functions != filter2->num_functions || filter1->hash_func != filter2->hash_func) {
|
207
|
+
return NULL;
|
208
|
+
}
|
209
|
+
|
210
|
+
/* Create a new bloom filter for the result */
|
211
|
+
|
212
|
+
result = bloom_filter_new (filter1->table_size, filter1->hash_func, filter1->num_functions);
|
213
|
+
|
214
|
+
if (result == NULL) {
|
215
|
+
return NULL;
|
216
|
+
}
|
217
|
+
|
218
|
+
/* The table is an array of bits, packed into bytes. Round up
|
219
|
+
* to the nearest byte. */
|
220
|
+
|
221
|
+
array_size = (filter1->table_size + 7) / 8;
|
222
|
+
|
223
|
+
/* Populate the table of the new filter */
|
224
|
+
|
225
|
+
for (i = 0; i < array_size; ++i) {
|
226
|
+
result->table[i] = filter1->table[i] | filter2->table[i];
|
227
|
+
}
|
228
|
+
|
229
|
+
return result;
|
230
|
+
}
|
231
|
+
|
232
|
+
BloomFilter *bloom_filter_intersection (BloomFilter * filter1, BloomFilter * filter2) {
|
233
|
+
BloomFilter *result;
|
234
|
+
unsigned int i;
|
235
|
+
unsigned int array_size;
|
236
|
+
|
237
|
+
/* To perform this operation, both filters must be created with
|
238
|
+
* the same values. */
|
239
|
+
|
240
|
+
if (filter1->table_size != filter2->table_size
|
241
|
+
|| filter1->num_functions != filter2->num_functions || filter1->hash_func != filter2->hash_func) {
|
242
|
+
return NULL;
|
243
|
+
}
|
244
|
+
|
245
|
+
/* Create a new bloom filter for the result */
|
246
|
+
|
247
|
+
result = bloom_filter_new (filter1->table_size, filter1->hash_func, filter1->num_functions);
|
248
|
+
|
249
|
+
if (result == NULL) {
|
250
|
+
return NULL;
|
251
|
+
}
|
252
|
+
|
253
|
+
/* The table is an array of bits, packed into bytes. Round up
|
254
|
+
* to the nearest byte. */
|
255
|
+
|
256
|
+
array_size = (filter1->table_size + 7) / 8;
|
257
|
+
|
258
|
+
/* Populate the table of the new filter */
|
259
|
+
|
260
|
+
for (i = 0; i < array_size; ++i) {
|
261
|
+
result->table[i] = filter1->table[i] & filter2->table[i];
|
262
|
+
}
|
263
|
+
|
264
|
+
return result;
|
265
|
+
}
|
data/ext/bloom-filter.h
ADDED
@@ -0,0 +1,193 @@
|
|
1
|
+
/*
|
2
|
+
|
3
|
+
Copyright (c) 2005-2008, Simon Howard
|
4
|
+
|
5
|
+
Permission to use, copy, modify, and/or distribute this software
|
6
|
+
for any purpose with or without fee is hereby granted, provided
|
7
|
+
that the above copyright notice and this permission notice appear
|
8
|
+
in all copies.
|
9
|
+
|
10
|
+
THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL
|
11
|
+
WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED
|
12
|
+
WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE
|
13
|
+
AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR
|
14
|
+
CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM
|
15
|
+
LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT,
|
16
|
+
NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
|
17
|
+
CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
|
18
|
+
|
19
|
+
*/
|
20
|
+
|
21
|
+
/**
|
22
|
+
* @file bloom-filter.h
|
23
|
+
*
|
24
|
+
* @brief Bloom filter
|
25
|
+
*
|
26
|
+
* A bloom filter is a space efficient data structure that can be
|
27
|
+
* used to test whether a given element is part of a set. Lookups
|
28
|
+
* will occasionally generate false positives, but never false
|
29
|
+
* negatives.
|
30
|
+
*
|
31
|
+
* To create a bloom filter, use @ref bloom_filter_new. To destroy a
|
32
|
+
* bloom filter, use @ref bloom_filter_free.
|
33
|
+
*
|
34
|
+
* To insert a value into a bloom filter, use @ref bloom_filter_insert.
|
35
|
+
*
|
36
|
+
* To query whether a value is part of the set, use
|
37
|
+
* @ref bloom_filter_query.
|
38
|
+
*/
|
39
|
+
|
40
|
+
#ifndef ALGORITHM_BLOOM_FILTER_H
|
41
|
+
#define ALGORITHM_BLOOM_FILTER_H
|
42
|
+
|
43
|
+
#ifdef __cplusplus
|
44
|
+
extern "C" {
|
45
|
+
#endif
|
46
|
+
|
47
|
+
/**
|
48
|
+
* A value stored in a @ref BloomFilter.
|
49
|
+
*/
|
50
|
+
|
51
|
+
typedef void *BloomFilterValue;
|
52
|
+
|
53
|
+
/**
|
54
|
+
* Hash function used to generate hash values for values inserted into a
|
55
|
+
* bloom filter.
|
56
|
+
*
|
57
|
+
* @param data The value to generate a hash value for.
|
58
|
+
* @return The hash value.
|
59
|
+
*/
|
60
|
+
|
61
|
+
typedef unsigned long (*BloomFilterHashFunc) (BloomFilterValue data);
|
62
|
+
|
63
|
+
/**
|
64
|
+
* A bloom filter structure.
|
65
|
+
*/
|
66
|
+
|
67
|
+
typedef struct BloomFilter {
|
68
|
+
BloomFilterHashFunc hash_func;
|
69
|
+
unsigned char *table;
|
70
|
+
unsigned int table_size;
|
71
|
+
unsigned int num_functions;
|
72
|
+
} BloomFilter;
|
73
|
+
|
74
|
+
|
75
|
+
/**
|
76
|
+
* Create a new bloom filter.
|
77
|
+
*
|
78
|
+
* @param table_size The size of the bloom filter. The greater
|
79
|
+
* the table size, the more elements can be
|
80
|
+
* stored, and the lesser the chance of false
|
81
|
+
* positives.
|
82
|
+
* @param hash_func Hash function to use on values stored in the
|
83
|
+
* filter.
|
84
|
+
* @param num_functions Number of hash functions to apply to each
|
85
|
+
* element on insertion. This running time for
|
86
|
+
* insertion and queries is proportional to this
|
87
|
+
* value. The more functions applied, the lesser
|
88
|
+
* the chance of false positives. The maximum
|
89
|
+
* number of functions is 64.
|
90
|
+
* @return A new hash table structure, or NULL if it
|
91
|
+
* was not possible to allocate the new bloom
|
92
|
+
* filter.
|
93
|
+
*/
|
94
|
+
|
95
|
+
BloomFilter *bloom_filter_new (unsigned int table_size, BloomFilterHashFunc hash_func, unsigned int num_functions);
|
96
|
+
|
97
|
+
/**
|
98
|
+
* Destroy a bloom filter.
|
99
|
+
*
|
100
|
+
* @param bloomfilter The bloom filter to destroy.
|
101
|
+
*/
|
102
|
+
|
103
|
+
void bloom_filter_free (BloomFilter * bloomfilter);
|
104
|
+
|
105
|
+
/**
|
106
|
+
* Insert a value into a bloom filter.
|
107
|
+
*
|
108
|
+
* @param bloomfilter The bloom filter.
|
109
|
+
* @param value The value to insert.
|
110
|
+
*/
|
111
|
+
|
112
|
+
void bloom_filter_insert (BloomFilter * bloomfilter, BloomFilterValue value);
|
113
|
+
|
114
|
+
/**
|
115
|
+
* Query a bloom filter for a particular value.
|
116
|
+
*
|
117
|
+
* @param bloomfilter The bloom filter.
|
118
|
+
* @param value The value to look up.
|
119
|
+
* @return Zero if the value was definitely not
|
120
|
+
* inserted into the filter. Non-zero
|
121
|
+
* indicates that it either may or may not
|
122
|
+
* have been inserted.
|
123
|
+
*/
|
124
|
+
|
125
|
+
int bloom_filter_query (BloomFilter * bloomfilter, BloomFilterValue value);
|
126
|
+
|
127
|
+
/**
|
128
|
+
* Read the contents of a bloom filter into an array.
|
129
|
+
*
|
130
|
+
* @param bloomfilter The bloom filter.
|
131
|
+
* @param array Pointer to the array to read into. This
|
132
|
+
* should be (table_size + 7) / 8 bytes in
|
133
|
+
* length.
|
134
|
+
*/
|
135
|
+
|
136
|
+
void bloom_filter_read (BloomFilter * bloomfilter, unsigned char *array);
|
137
|
+
|
138
|
+
/**
|
139
|
+
* Load the contents of a bloom filter from an array.
|
140
|
+
* The data loaded should be the output read from @ref bloom_filter_read,
|
141
|
+
* from a bloom filter created using the same arguments used to create
|
142
|
+
* the original filter.
|
143
|
+
*
|
144
|
+
* @param bloomfilter The bloom filter.
|
145
|
+
* @param array Pointer to the array to load from. This
|
146
|
+
* should be (table_size + 7) / 8 bytes in
|
147
|
+
* length.
|
148
|
+
*/
|
149
|
+
|
150
|
+
void bloom_filter_load (BloomFilter * bloomfilter, unsigned char *array);
|
151
|
+
|
152
|
+
/**
|
153
|
+
* Find the union of two bloom filters. Values are present in the
|
154
|
+
* resulting filter if they are present in either of the original
|
155
|
+
* filters.
|
156
|
+
*
|
157
|
+
* Both of the original filters must have been created using the
|
158
|
+
* same parameters to @ref bloom_filter_new.
|
159
|
+
*
|
160
|
+
* @param filter1 The first filter.
|
161
|
+
* @param filter2 The second filter.
|
162
|
+
* @return A new filter which is an intersection of the
|
163
|
+
* two filters, or NULL if it was not possible
|
164
|
+
* to allocate memory for the new filter, or
|
165
|
+
* if the two filters specified were created
|
166
|
+
* with different parameters.
|
167
|
+
*/
|
168
|
+
|
169
|
+
BloomFilter *bloom_filter_union (BloomFilter * filter1, BloomFilter * filter2);
|
170
|
+
|
171
|
+
/**
|
172
|
+
* Find the intersection of two bloom filters. Values are only ever
|
173
|
+
* present in the resulting filter if they are present in both of the
|
174
|
+
* original filters.
|
175
|
+
*
|
176
|
+
* Both of the original filters must have been created using the
|
177
|
+
* same parameters to @ref bloom_filter_new.
|
178
|
+
*
|
179
|
+
* @param filter1 The first filter.
|
180
|
+
* @param filter2 The second filter.
|
181
|
+
* @return A new filter which is an intersection of the
|
182
|
+
* two filters, or NULL if it was not possible
|
183
|
+
* to allocate memory for the new filter, or
|
184
|
+
* if the two filters specified were created
|
185
|
+
* with different parameters.
|
186
|
+
*/
|
187
|
+
|
188
|
+
BloomFilter *bloom_filter_intersection (BloomFilter * filter1, BloomFilter * filter2);
|
189
|
+
|
190
|
+
#ifdef __cplusplus
|
191
|
+
}
|
192
|
+
#endif
|
193
|
+
#endif /* #ifndef ALGORITHM_BLOOM_FILTER_H */
|
data/ext/bloom_filter.c
ADDED
@@ -0,0 +1,248 @@
|
|
1
|
+
#include <sys/types.h>
|
2
|
+
#include <sys/stat.h>
|
3
|
+
#include <fcntl.h>
|
4
|
+
#include <unistd.h>
|
5
|
+
#include <errno.h>
|
6
|
+
#include <string.h>
|
7
|
+
#include <stdlib.h>
|
8
|
+
#include <math.h>
|
9
|
+
|
10
|
+
#include "ruby/ruby.h"
|
11
|
+
#include "bloom-filter.h"
|
12
|
+
#include "hash-string.h"
|
13
|
+
#include "version.h"
|
14
|
+
|
15
|
+
#define TO_S(v) rb_funcall(v, rb_intern("to_s"), 0)
|
16
|
+
#define CSTRING(v) RSTRING_PTR(TO_S(v))
|
17
|
+
|
18
|
+
typedef struct FileHeader {
|
19
|
+
uint64_t table_size;
|
20
|
+
uint64_t num_functions;
|
21
|
+
} FileHeader;
|
22
|
+
|
23
|
+
static void bloom_free(BloomFilter *filter) {
|
24
|
+
if (filter)
|
25
|
+
bloom_filter_free(filter);
|
26
|
+
}
|
27
|
+
|
28
|
+
VALUE bloom_allocate(VALUE klass) {
|
29
|
+
BloomFilter *filter = 0;
|
30
|
+
return Data_Wrap_Struct(klass, 0, bloom_free, filter);
|
31
|
+
}
|
32
|
+
|
33
|
+
BloomFilter* bloom_handle(VALUE self) {
|
34
|
+
BloomFilter *filter = 0;
|
35
|
+
Data_Get_Struct(self, BloomFilter, filter);
|
36
|
+
if (!filter)
|
37
|
+
rb_raise(rb_eArgError, "invalid BloomFilter instance");
|
38
|
+
return filter;
|
39
|
+
}
|
40
|
+
|
41
|
+
VALUE bloom_initialize(int argc, VALUE *argv, VALUE self) {
|
42
|
+
double error;
|
43
|
+
size_t nbits, nhash, nmax;
|
44
|
+
|
45
|
+
VALUE max_size, error_rate, bitmap_size, hash_count, options;
|
46
|
+
BloomFilter *filter = 0;
|
47
|
+
|
48
|
+
rb_scan_args(argc, argv, "01", &options);
|
49
|
+
if (!NIL_P(options) && TYPE(options) != T_HASH)
|
50
|
+
rb_raise(rb_eArgError, "invalid options, expect hash");
|
51
|
+
|
52
|
+
if (NIL_P(options)) {
|
53
|
+
nbits = 1000000;
|
54
|
+
nhash = 4;
|
55
|
+
}
|
56
|
+
else {
|
57
|
+
max_size = rb_hash_aref(options, ID2SYM(rb_intern("size")));
|
58
|
+
error_rate = rb_hash_aref(options, ID2SYM(rb_intern("error_rate")));
|
59
|
+
bitmap_size = rb_hash_aref(options, ID2SYM(rb_intern("bits")));
|
60
|
+
hash_count = rb_hash_aref(options, ID2SYM(rb_intern("hashes")));
|
61
|
+
|
62
|
+
nhash = NIL_P(hash_count) ? 4 : NUM2ULONG(hash_count);
|
63
|
+
|
64
|
+
if (!NIL_P(bitmap_size))
|
65
|
+
nbits = NUM2ULONG(bitmap_size);
|
66
|
+
else if (!NIL_P(max_size)) {
|
67
|
+
nmax = NUM2ULONG(max_size);
|
68
|
+
error = NIL_P(error_rate) ? 0.01 : NUM2DBL(error_rate);
|
69
|
+
nbits = ceil(fabs(log(error) * (double)nmax / pow(log(2), 2)));
|
70
|
+
nhash = ceil(0.7 * (double)nbits / (double)nmax);
|
71
|
+
}
|
72
|
+
else
|
73
|
+
rb_raise(rb_eArgError, "requires either size & error_rate or bits & hashes");
|
74
|
+
}
|
75
|
+
|
76
|
+
filter = bloom_filter_new(nbits, string_nocase_hash, nhash);
|
77
|
+
|
78
|
+
if (!filter)
|
79
|
+
rb_raise(rb_eNoMemError, "unable to allocate memory for BloomFilter");
|
80
|
+
|
81
|
+
DATA_PTR(self) = filter;
|
82
|
+
return self;
|
83
|
+
}
|
84
|
+
|
85
|
+
|
86
|
+
VALUE bloom_insert(VALUE klass, VALUE string) {
|
87
|
+
BloomFilter *filter = bloom_handle(klass);
|
88
|
+
bloom_filter_insert(filter, (BloomFilterValue)CSTRING(string));
|
89
|
+
return Qtrue;
|
90
|
+
}
|
91
|
+
|
92
|
+
VALUE bloom_include(VALUE klass, VALUE string) {
|
93
|
+
BloomFilter *filter = bloom_handle(klass);
|
94
|
+
return bloom_filter_query(filter, (BloomFilterValue)CSTRING(string)) ? Qtrue : Qfalse;
|
95
|
+
}
|
96
|
+
|
97
|
+
VALUE bloom_dump(VALUE klass, VALUE file) {
|
98
|
+
int fd;
|
99
|
+
void *buffer;
|
100
|
+
uint64_t nbits;
|
101
|
+
FileHeader header;
|
102
|
+
BloomFilter *filter = bloom_handle(klass);
|
103
|
+
|
104
|
+
nbits = (filter->table_size + 7) / 8;
|
105
|
+
buffer = malloc(nbits);
|
106
|
+
|
107
|
+
if (!buffer)
|
108
|
+
rb_raise(rb_eNoMemError, "out of memory dumping BloomFilter");
|
109
|
+
|
110
|
+
bloom_filter_read(filter, buffer);
|
111
|
+
|
112
|
+
fd = open(CSTRING(file), O_WRONLY|O_CREAT|O_TRUNC, S_IRUSR|S_IWUSR);
|
113
|
+
if (fd == -1) {
|
114
|
+
free(buffer);
|
115
|
+
rb_raise(rb_eIOError, "unable to open file. %s", strerror(errno));
|
116
|
+
}
|
117
|
+
|
118
|
+
header.table_size = filter->table_size;
|
119
|
+
header.num_functions = filter->num_functions;
|
120
|
+
|
121
|
+
if (write(fd, &header, sizeof(header)) == -1) {
|
122
|
+
free(buffer);
|
123
|
+
close(fd);
|
124
|
+
rb_raise(rb_eIOError, "error dumping BloomFilter: %s\n", strerror(errno));
|
125
|
+
}
|
126
|
+
|
127
|
+
if (write(fd, buffer, nbits) != -1) {
|
128
|
+
free(buffer);
|
129
|
+
close(fd);
|
130
|
+
return Qtrue;
|
131
|
+
}
|
132
|
+
else {
|
133
|
+
free(buffer);
|
134
|
+
close(fd);
|
135
|
+
rb_raise(rb_eIOError, "error dumping BloomFilter: %s\n", strerror(errno));
|
136
|
+
}
|
137
|
+
|
138
|
+
return Qfalse; // not reachable
|
139
|
+
}
|
140
|
+
|
141
|
+
VALUE bloom_bits(VALUE klass) {
|
142
|
+
BloomFilter *filter = bloom_handle(klass);
|
143
|
+
|
144
|
+
VALUE bitmap;
|
145
|
+
char *buffer;
|
146
|
+
unsigned char b;
|
147
|
+
int i = 0, bit, nbits = filter->table_size;
|
148
|
+
|
149
|
+
buffer = (char *)malloc(nbits);
|
150
|
+
if (!buffer)
|
151
|
+
rb_raise(rb_eNoMemError, "out of memory dumping BloomFilter");
|
152
|
+
|
153
|
+
for (i = 0; i < nbits; i++) {
|
154
|
+
b = filter->table[i / 8];
|
155
|
+
bit = 1 << (i % 8);
|
156
|
+
|
157
|
+
if ((b & bit) == 0)
|
158
|
+
buffer[i] = '0';
|
159
|
+
else
|
160
|
+
buffer[i] = '1';
|
161
|
+
}
|
162
|
+
|
163
|
+
bitmap = rb_str_new(buffer, nbits);
|
164
|
+
free(buffer);
|
165
|
+
return bitmap;
|
166
|
+
}
|
167
|
+
|
168
|
+
VALUE bloom_binary(VALUE klass) {
|
169
|
+
BloomFilter *filter = bloom_handle(klass);
|
170
|
+
|
171
|
+
VALUE bitmap;
|
172
|
+
char *buffer;
|
173
|
+
|
174
|
+
int nbytes = (filter->table_size + 7) / 8;
|
175
|
+
buffer = (char *)malloc(nbytes);
|
176
|
+
|
177
|
+
if (!buffer)
|
178
|
+
rb_raise(rb_eNoMemError, "out of memory dumping BloomFilter");
|
179
|
+
|
180
|
+
bloom_filter_read(filter, buffer);
|
181
|
+
|
182
|
+
bitmap = rb_str_new(buffer, nbytes);
|
183
|
+
free(buffer);
|
184
|
+
return bitmap;
|
185
|
+
}
|
186
|
+
|
187
|
+
VALUE bloom_binary_set(VALUE klass, VALUE buffer) {
|
188
|
+
BloomFilter *filter = bloom_handle(klass);
|
189
|
+
char* ptr = (char *) RSTRING_PTR(buffer);
|
190
|
+
bloom_filter_load(filter, ptr);
|
191
|
+
return Qtrue;
|
192
|
+
}
|
193
|
+
|
194
|
+
VALUE bloom_load(VALUE klass, VALUE file) {
|
195
|
+
int fd;
|
196
|
+
void *buffer;
|
197
|
+
size_t nbits, bytes;
|
198
|
+
FileHeader header;
|
199
|
+
BloomFilter *filter;
|
200
|
+
VALUE instance;
|
201
|
+
|
202
|
+
fd = open(CSTRING(file), O_RDONLY);
|
203
|
+
if (fd == -1)
|
204
|
+
rb_raise(rb_eIOError, "unable to open file: %s", strerror(errno));
|
205
|
+
|
206
|
+
if (read(fd, &header, sizeof(header)) != sizeof(header)) {
|
207
|
+
close(fd);
|
208
|
+
rb_raise(rb_eIOError, "unable to read file, header corrupted\n");
|
209
|
+
}
|
210
|
+
|
211
|
+
nbits = (header.table_size + 7) / 8;
|
212
|
+
buffer = malloc(nbits);
|
213
|
+
if (!buffer) {
|
214
|
+
close(fd);
|
215
|
+
rb_raise(rb_eNoMemError, "out of memory dumping BloomFilter");
|
216
|
+
}
|
217
|
+
|
218
|
+
bytes = read(fd, buffer, nbits);
|
219
|
+
if (bytes != nbits) {
|
220
|
+
free(buffer);
|
221
|
+
close(fd);
|
222
|
+
rb_raise(rb_eStandardError, "unable to load BloomFilter, expected %ld but got %ld bytes", nbits, bytes);
|
223
|
+
}
|
224
|
+
|
225
|
+
filter = bloom_filter_new(header.table_size, string_nocase_hash, header.num_functions);
|
226
|
+
bloom_filter_load(filter, buffer);
|
227
|
+
instance = Data_Wrap_Struct(klass, 0, bloom_free, filter);
|
228
|
+
|
229
|
+
free(buffer);
|
230
|
+
close(fd);
|
231
|
+
return instance;
|
232
|
+
}
|
233
|
+
|
234
|
+
void Init_bloom_filter() {
|
235
|
+
VALUE cBloom = rb_define_class("BloomFilter", rb_cObject);
|
236
|
+
|
237
|
+
rb_define_method(cBloom, "initialize", RUBY_METHOD_FUNC(bloom_initialize), -1);
|
238
|
+
rb_define_method(cBloom, "dump", RUBY_METHOD_FUNC(bloom_dump), 1);
|
239
|
+
rb_define_method(cBloom, "bits", RUBY_METHOD_FUNC(bloom_bits), 0);
|
240
|
+
rb_define_method(cBloom, "binary", RUBY_METHOD_FUNC(bloom_binary), 0);
|
241
|
+
rb_define_method(cBloom, "binary=", RUBY_METHOD_FUNC(bloom_binary_set), 1);
|
242
|
+
rb_define_method(cBloom, "insert", RUBY_METHOD_FUNC(bloom_insert), 1);
|
243
|
+
rb_define_method(cBloom, "include?", RUBY_METHOD_FUNC(bloom_include), 1);
|
244
|
+
|
245
|
+
rb_define_alloc_func(cBloom, bloom_allocate);
|
246
|
+
rb_define_singleton_method(cBloom, "load", RUBY_METHOD_FUNC(bloom_load), 1);
|
247
|
+
rb_define_const(cBloom, "VERSION", rb_str_new2(RUBY_BLOOM_FILTER_VERSION));
|
248
|
+
}
|
data/ext/extconf.rb
ADDED
data/ext/hash-string.c
ADDED
@@ -0,0 +1,58 @@
|
|
1
|
+
/*
|
2
|
+
|
3
|
+
Copyright (c) 2005-2008, Simon Howard
|
4
|
+
|
5
|
+
Permission to use, copy, modify, and/or distribute this software
|
6
|
+
for any purpose with or without fee is hereby granted, provided
|
7
|
+
that the above copyright notice and this permission notice appear
|
8
|
+
in all copies.
|
9
|
+
|
10
|
+
THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL
|
11
|
+
WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED
|
12
|
+
WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE
|
13
|
+
AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR
|
14
|
+
CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM
|
15
|
+
LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT,
|
16
|
+
NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
|
17
|
+
CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
|
18
|
+
|
19
|
+
*/
|
20
|
+
|
21
|
+
#include <ctype.h>
|
22
|
+
|
23
|
+
#include "hash-string.h"
|
24
|
+
|
25
|
+
/* String hash function */
|
26
|
+
|
27
|
+
unsigned long string_hash (void *string) {
|
28
|
+
/* This is the djb2 string hash function */
|
29
|
+
|
30
|
+
unsigned long result = 5381;
|
31
|
+
unsigned char *p;
|
32
|
+
|
33
|
+
p = (unsigned char *) string;
|
34
|
+
|
35
|
+
while (*p != '\0') {
|
36
|
+
result = ((result << 5) ^ result) ^ (*p);
|
37
|
+
++p;
|
38
|
+
}
|
39
|
+
|
40
|
+
return result;
|
41
|
+
}
|
42
|
+
|
43
|
+
/* The same function, with a tolower on every character so that
|
44
|
+
* case is ignored. This code is duplicated for performance. */
|
45
|
+
|
46
|
+
unsigned long string_nocase_hash (void *string) {
|
47
|
+
unsigned long result = 5381;
|
48
|
+
unsigned char *p;
|
49
|
+
|
50
|
+
p = (unsigned char *) string;
|
51
|
+
|
52
|
+
while (*p != '\0') {
|
53
|
+
result = ((result << 5) ^ result) ^ tolower (*p);
|
54
|
+
++p;
|
55
|
+
}
|
56
|
+
|
57
|
+
return result;
|
58
|
+
}
|
data/ext/hash-string.h
ADDED
@@ -0,0 +1,56 @@
|
|
1
|
+
/*
|
2
|
+
|
3
|
+
Copyright (c) 2005-2008, Simon Howard
|
4
|
+
|
5
|
+
Permission to use, copy, modify, and/or distribute this software
|
6
|
+
for any purpose with or without fee is hereby granted, provided
|
7
|
+
that the above copyright notice and this permission notice appear
|
8
|
+
in all copies.
|
9
|
+
|
10
|
+
THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL
|
11
|
+
WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED
|
12
|
+
WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE
|
13
|
+
AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR
|
14
|
+
CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM
|
15
|
+
LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT,
|
16
|
+
NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
|
17
|
+
CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
|
18
|
+
|
19
|
+
*/
|
20
|
+
|
21
|
+
/**
|
22
|
+
* @file hash-string.h
|
23
|
+
*
|
24
|
+
* Hash functions for text strings. For more information
|
25
|
+
* see @ref string_hash or @ref string_nocase_hash.
|
26
|
+
*/
|
27
|
+
|
28
|
+
#ifndef ALGORITHM_HASH_STRING_H
|
29
|
+
#define ALGORITHM_HASH_STRING_H
|
30
|
+
|
31
|
+
#ifdef __cplusplus
|
32
|
+
extern "C" {
|
33
|
+
#endif
|
34
|
+
|
35
|
+
/**
|
36
|
+
* Generate a hash key from a string.
|
37
|
+
*
|
38
|
+
* @param string The string.
|
39
|
+
* @return A hash key for the string.
|
40
|
+
*/
|
41
|
+
|
42
|
+
unsigned long string_hash (void *string);
|
43
|
+
|
44
|
+
/**
|
45
|
+
* Generate a hash key from a string, ignoring the case of letters.
|
46
|
+
*
|
47
|
+
* @param string The string.
|
48
|
+
* @return A hash key for the string.
|
49
|
+
*/
|
50
|
+
|
51
|
+
unsigned long string_nocase_hash (void *string);
|
52
|
+
|
53
|
+
#ifdef __cplusplus
|
54
|
+
}
|
55
|
+
#endif
|
56
|
+
#endif /* #ifndef ALGORITHM_HASH_STRING_H */
|
data/ext/version.h
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
#define RUBY_BLOOM_FILTER_VERSION "0.2.1"
|
data/lib/bloom-filter.rb
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
require 'bloom_filter'
|
data/test/helper.rb
ADDED
data/test/test_basic.rb
ADDED
@@ -0,0 +1,29 @@
|
|
1
|
+
require 'helper'
|
2
|
+
|
3
|
+
describe 'BloomFilter primitives' do
|
4
|
+
it 'should create one with default size' do
|
5
|
+
assert BloomFilter.new
|
6
|
+
end
|
7
|
+
|
8
|
+
it 'should create one with given size' do
|
9
|
+
assert BloomFilter.new(size: 100)
|
10
|
+
end
|
11
|
+
|
12
|
+
it 'should insert' do
|
13
|
+
assert filter = BloomFilter.new(size: 100)
|
14
|
+
assert filter.insert("foo")
|
15
|
+
end
|
16
|
+
|
17
|
+
it 'should allow membership checks' do
|
18
|
+
assert filter = BloomFilter.new(size: 100)
|
19
|
+
assert filter.insert("foo")
|
20
|
+
assert filter.include?("foo")
|
21
|
+
assert !filter.include?("bar")
|
22
|
+
end
|
23
|
+
|
24
|
+
it 'should return bits as a string' do
|
25
|
+
assert filter = BloomFilter.new(bits: 5, hashes: 2)
|
26
|
+
assert filter.insert("a")
|
27
|
+
assert_equal "10010", filter.bits
|
28
|
+
end
|
29
|
+
end
|
data/test/test_io.rb
ADDED
@@ -0,0 +1,27 @@
|
|
1
|
+
require 'helper'
|
2
|
+
require 'tempfile'
|
3
|
+
|
4
|
+
describe 'BloomFilter load & dump' do
|
5
|
+
it 'should dump and load a filter' do
|
6
|
+
file = Tempfile.new("bloom-filter-test")
|
7
|
+
|
8
|
+
assert filter = BloomFilter.new(size: 100)
|
9
|
+
assert filter.insert("foo")
|
10
|
+
assert filter.dump(file.path), "dump filter"
|
11
|
+
assert filter = BloomFilter.load(file.path)
|
12
|
+
assert filter.include?("foo")
|
13
|
+
assert !filter.include?("bar")
|
14
|
+
end
|
15
|
+
|
16
|
+
it 'should accept assigning the bits directly' do
|
17
|
+
assert filter = BloomFilter.new(bits: 80)
|
18
|
+
assert filter.insert("foo")
|
19
|
+
puts filter.binary.inspect
|
20
|
+
assert filter2 = BloomFilter.new(bits: 80)
|
21
|
+
assert filter2.binary = filter.binary
|
22
|
+
puts
|
23
|
+
puts filter2.binary.inspect
|
24
|
+
assert filter2.include?("foo")
|
25
|
+
assert !filter2.include?("bar")
|
26
|
+
end
|
27
|
+
end
|
metadata
ADDED
@@ -0,0 +1,91 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: bloom-ruby
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.2.1
|
5
|
+
prerelease:
|
6
|
+
platform: ruby
|
7
|
+
authors:
|
8
|
+
- Bharanee Rathna
|
9
|
+
autorequire:
|
10
|
+
bindir: bin
|
11
|
+
cert_chain: []
|
12
|
+
date: 2013-01-21 00:00:00.000000000 Z
|
13
|
+
dependencies:
|
14
|
+
- !ruby/object:Gem::Dependency
|
15
|
+
name: rake
|
16
|
+
requirement: !ruby/object:Gem::Requirement
|
17
|
+
none: false
|
18
|
+
requirements:
|
19
|
+
- - ! '>='
|
20
|
+
- !ruby/object:Gem::Version
|
21
|
+
version: '0'
|
22
|
+
type: :development
|
23
|
+
prerelease: false
|
24
|
+
version_requirements: !ruby/object:Gem::Requirement
|
25
|
+
none: false
|
26
|
+
requirements:
|
27
|
+
- - ! '>='
|
28
|
+
- !ruby/object:Gem::Version
|
29
|
+
version: '0'
|
30
|
+
- !ruby/object:Gem::Dependency
|
31
|
+
name: rake-compiler
|
32
|
+
requirement: !ruby/object:Gem::Requirement
|
33
|
+
none: false
|
34
|
+
requirements:
|
35
|
+
- - ! '>='
|
36
|
+
- !ruby/object:Gem::Version
|
37
|
+
version: '0'
|
38
|
+
type: :development
|
39
|
+
prerelease: false
|
40
|
+
version_requirements: !ruby/object:Gem::Requirement
|
41
|
+
none: false
|
42
|
+
requirements:
|
43
|
+
- - ! '>='
|
44
|
+
- !ruby/object:Gem::Version
|
45
|
+
version: '0'
|
46
|
+
description: A fast Bloom Filter library for Ruby for unices.
|
47
|
+
email:
|
48
|
+
- deepfryed@gmail.com
|
49
|
+
executables: []
|
50
|
+
extensions:
|
51
|
+
- ext/extconf.rb
|
52
|
+
extra_rdoc_files: []
|
53
|
+
files:
|
54
|
+
- ext/bloom-filter.c
|
55
|
+
- ext/hash-string.c
|
56
|
+
- ext/bloom_filter.c
|
57
|
+
- ext/bloom-filter.h
|
58
|
+
- ext/hash-string.h
|
59
|
+
- ext/version.h
|
60
|
+
- ext/extconf.rb
|
61
|
+
- test/helper.rb
|
62
|
+
- test/test_io.rb
|
63
|
+
- test/test_basic.rb
|
64
|
+
- README.md
|
65
|
+
- CHANGELOG
|
66
|
+
- lib/bloom-filter.rb
|
67
|
+
homepage: http://github.com/deepfryed/bloom-filter
|
68
|
+
licenses: []
|
69
|
+
post_install_message:
|
70
|
+
rdoc_options: []
|
71
|
+
require_paths:
|
72
|
+
- lib
|
73
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
74
|
+
none: false
|
75
|
+
requirements:
|
76
|
+
- - ! '>='
|
77
|
+
- !ruby/object:Gem::Version
|
78
|
+
version: '0'
|
79
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
80
|
+
none: false
|
81
|
+
requirements:
|
82
|
+
- - ! '>='
|
83
|
+
- !ruby/object:Gem::Version
|
84
|
+
version: '0'
|
85
|
+
requirements: []
|
86
|
+
rubyforge_project:
|
87
|
+
rubygems_version: 1.8.25
|
88
|
+
signing_key:
|
89
|
+
specification_version: 3
|
90
|
+
summary: A fast Bloom Filter library for Ruby.
|
91
|
+
test_files: []
|