digest-murmurhash 0.2.3 → 0.3.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/README.md +16 -0
- data/Rakefile +11 -1
- data/digest-murmurhash.gemspec +3 -3
- data/ext/digest/murmurhash/extconf.rb +1 -2
- data/ext/digest/murmurhash/murmurhash.c +253 -0
- data/ext/digest/murmurhash/murmurhash.h +52 -35
- data/ext/digest/murmurhash/murmurhash1.c +18 -54
- data/ext/digest/murmurhash/murmurhash2.c +17 -56
- data/ext/digest/murmurhash/murmurhash2a.c +18 -42
- data/ext/digest/murmurhash/murmurhash64a.c +87 -0
- data/ext/digest/murmurhash/murmurhash64a.h +13 -0
- data/ext/digest/murmurhash/murmurhash64b.c +93 -0
- data/ext/digest/murmurhash/murmurhash64b.h +13 -0
- data/ext/digest/murmurhash/murmurhash_aligned2.c +87 -0
- data/ext/digest/murmurhash/murmurhash_aligned2.h +13 -0
- data/ext/digest/murmurhash/murmurhash_neutral2.c +87 -0
- data/ext/digest/murmurhash/murmurhash_neutral2.h +13 -0
- data/lib/digest/murmurhash/version.rb +1 -1
- data/spec/bench.rb +21 -31
- data/spec/digest_spec.rb +54 -34
- data/spec/exception_spec.rb +19 -4
- data/spec/mem_spec.rb +3 -6
- data/spec/spec_helper.rb +21 -0
- metadata +15 -7
- data/ext/digest/murmurhash/init.c +0 -39
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 3ae0e93d61ad1118ff16e6a8ee49e2624bfe93c6
|
4
|
+
data.tar.gz: 221199ad7ef8c94e0eba38d57ec6f0387652e051
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 1b4b91306a2c297fb8809037693c9ec30ca87352bc047dcad25b099513183e4534b0b260104133ea94054c129eccf5192604331b493e5076361b8c3e104ca610
|
7
|
+
data.tar.gz: 58bad31ff82316ca28e66a37dd0709f08bbb59e7df2839f78b80ca1dee00799ee925a3e8f25dab272d1c5d893708a017ae2d8d882db5029aca73b09f475e82fa
|
data/README.md
CHANGED
@@ -10,6 +10,14 @@
|
|
10
10
|
|
11
11
|
**Digest::MurmurHash2A** use algorithm MurmurHash2A(32-bit).
|
12
12
|
|
13
|
+
**Digest::MurmurHash64A** use algorithm MurmurHash64A(64-bit).
|
14
|
+
|
15
|
+
**Digest::MurmurHash64B** use algorithm MurmurHash64B(64-bit).
|
16
|
+
|
17
|
+
**Digest::MurmurHashNeutral2** use algorithm MurmurHashNeutral2(32-bit).
|
18
|
+
|
19
|
+
**Digest::MurmurHashAligned2** use algorithm MurmurHashAligned2(32-bit).
|
20
|
+
|
13
21
|
(**Digest::MurmurHash** exist to write only version number)
|
14
22
|
|
15
23
|
All classes compliance Digest API of Ruby.
|
@@ -40,6 +48,14 @@ p Digest::MurmurHash2.file("./LICENSE.txt").hexdigest #=> '78678326'
|
|
40
48
|
|
41
49
|
**Digest::MurmurHash2A** < Digest::StringBuffer
|
42
50
|
|
51
|
+
**Digest::MurmurHash64A** < Digest::StringBuffer
|
52
|
+
|
53
|
+
**Digest::MurmurHash64B** < Digest::StringBuffer
|
54
|
+
|
55
|
+
**Digest::MurmurHashNeutral2** < Digest::StringBuffer
|
56
|
+
|
57
|
+
**Digest::MurmurHashAligned2** < Digest::StringBuffer
|
58
|
+
|
43
59
|
## Installation
|
44
60
|
|
45
61
|
Add this line to your application's Gemfile:
|
data/Rakefile
CHANGED
@@ -17,5 +17,15 @@ Rake::ExtensionTask.new('murmurhash', spec) do |ext|
|
|
17
17
|
ext.lib_dir = 'lib/digest/murmurhash'
|
18
18
|
end
|
19
19
|
|
20
|
+
require 'digest/murmurhash/version'
|
21
|
+
desc "gem reinstall v#{Digest::MurmurHash::VERSION}"
|
22
|
+
task :reinstall do |t|
|
23
|
+
system "gem uninstall murmurhash"
|
24
|
+
system "rake clean"
|
25
|
+
system "bundle exec rake"
|
26
|
+
system "rake install"
|
27
|
+
system "irb -rdigest/murmurhash"
|
28
|
+
end
|
29
|
+
|
20
30
|
|
21
|
-
task :default => [:spec]
|
31
|
+
task :default => [:spec, :build]
|
data/digest-murmurhash.gemspec
CHANGED
@@ -8,8 +8,8 @@ Gem::Specification.new do |spec|
|
|
8
8
|
spec.version = Digest::MurmurHash::VERSION
|
9
9
|
spec.author = "ksss"
|
10
10
|
spec.email = "co000ri@gmail.com"
|
11
|
-
spec.description = %q{Digest::
|
12
|
-
spec.summary = %q{Digest::
|
11
|
+
spec.description = %q{Digest::MurmurHash is class collections of use algorithm MurmurHash desiged by Austin Appleby.}
|
12
|
+
spec.summary = %q{Digest::MurmurHash is class collections of use algorithm MurmurHash desiged by Austin Appleby.}
|
13
13
|
spec.homepage = ""
|
14
14
|
spec.license = "MIT"
|
15
15
|
|
@@ -23,5 +23,5 @@ Gem::Specification.new do |spec|
|
|
23
23
|
spec.add_development_dependency "bundler", "~> 1.3"
|
24
24
|
spec.add_development_dependency "rake"
|
25
25
|
spec.add_development_dependency "rspec", ['~> 2.11']
|
26
|
-
spec.add_development_dependency "rake-compiler", ["~> 0.
|
26
|
+
spec.add_development_dependency "rake-compiler", ["~> 0.8.3"]
|
27
27
|
end
|
@@ -0,0 +1,253 @@
|
|
1
|
+
#include "murmurhash.h"
|
2
|
+
#include "murmurhash1.h"
|
3
|
+
#include "murmurhash2.h"
|
4
|
+
#include "murmurhash2a.h"
|
5
|
+
#include "murmurhash64a.h"
|
6
|
+
#include "murmurhash64b.h"
|
7
|
+
#include "murmurhash_neutral2.h"
|
8
|
+
#include "murmurhash_aligned2.h"
|
9
|
+
|
10
|
+
VALUE cDigest_MurmurHash1,
|
11
|
+
cDigest_MurmurHash2,
|
12
|
+
cDigest_MurmurHash2A,
|
13
|
+
cDigest_MurmurHash64A,
|
14
|
+
cDigest_MurmurHash64B,
|
15
|
+
cDigest_MurmurHashNeutral2,
|
16
|
+
cDigest_MurmurHashAligned2;
|
17
|
+
ID id_seed,
|
18
|
+
id_DEFAULT_SEED;
|
19
|
+
|
20
|
+
VALUE
|
21
|
+
murmur_seed_get32(VALUE self)
|
22
|
+
{
|
23
|
+
if (!rb_ivar_defined(self, id_seed)) {
|
24
|
+
rb_ivar_set(self, id_seed, rb_usascii_str_new(DEFAULT_SEED, 4));
|
25
|
+
}
|
26
|
+
return rb_ivar_get(self, id_seed);
|
27
|
+
}
|
28
|
+
|
29
|
+
VALUE
|
30
|
+
murmur_seed_set32(VALUE self, VALUE obj)
|
31
|
+
{
|
32
|
+
StringValue(obj);
|
33
|
+
if (RSTRING_LEN(obj) != 4) {
|
34
|
+
rb_raise(rb_eArgError, "seed string should 64 bit chars");
|
35
|
+
}
|
36
|
+
return rb_ivar_set(self, id_seed, obj);
|
37
|
+
}
|
38
|
+
|
39
|
+
VALUE
|
40
|
+
murmur_seed_get64(VALUE self)
|
41
|
+
{
|
42
|
+
if (!rb_ivar_defined(self, id_seed)) {
|
43
|
+
rb_ivar_set(self, id_seed, rb_usascii_str_new(DEFAULT_SEED, 8));
|
44
|
+
}
|
45
|
+
return rb_ivar_get(self, id_seed);
|
46
|
+
}
|
47
|
+
|
48
|
+
VALUE
|
49
|
+
murmur_seed_set64(VALUE self, VALUE obj)
|
50
|
+
{
|
51
|
+
StringValue(obj);
|
52
|
+
if (RSTRING_LEN(obj) != 8) {
|
53
|
+
rb_raise(rb_eArgError, "seed string should 64 bit chars");
|
54
|
+
}
|
55
|
+
return rb_ivar_set(self, id_seed, obj);
|
56
|
+
}
|
57
|
+
|
58
|
+
uint32_t
|
59
|
+
_murmur_finish32(VALUE self, uint32_t (*process)(const char *, uint32_t, uint32_t))
|
60
|
+
{
|
61
|
+
const char *seed = RSTRING_PTR(murmur_seed_get32(self));
|
62
|
+
MURMURHASH(self, ptr);
|
63
|
+
return process(ptr->buffer, ptr->p - ptr->buffer, *(uint32_t*)seed);
|
64
|
+
}
|
65
|
+
|
66
|
+
uint64_t
|
67
|
+
_murmur_finish64(VALUE self, uint64_t (*process)(const char *, uint32_t, uint64_t))
|
68
|
+
{
|
69
|
+
const char *seed = RSTRING_PTR(murmur_seed_get64(self));
|
70
|
+
MURMURHASH(self, ptr);
|
71
|
+
return process(ptr->buffer, ptr->p - ptr->buffer, *(uint64_t*)seed);
|
72
|
+
}
|
73
|
+
|
74
|
+
|
75
|
+
uint32_t
|
76
|
+
_murmur_s_digest32(int argc, VALUE *argv, VALUE klass, uint32_t (*process)(const char *, uint32_t, uint32_t))
|
77
|
+
{
|
78
|
+
VALUE str;
|
79
|
+
const char *seed;
|
80
|
+
|
81
|
+
if (argc < 1)
|
82
|
+
rb_raise(rb_eArgError, "no data given");
|
83
|
+
|
84
|
+
str = *argv;
|
85
|
+
|
86
|
+
StringValue(str);
|
87
|
+
|
88
|
+
if (1 < argc) {
|
89
|
+
StringValue(argv[1]);
|
90
|
+
if (RSTRING_LEN(argv[1]) != 4) {
|
91
|
+
rb_raise(rb_eArgError, "seed string should 32 bit chars");
|
92
|
+
}
|
93
|
+
seed = RSTRING_PTR(argv[1]);
|
94
|
+
} else {
|
95
|
+
seed = RSTRING_PTR(rb_const_get(klass, id_DEFAULT_SEED));
|
96
|
+
}
|
97
|
+
|
98
|
+
return process(RSTRING_PTR(str), RSTRING_LEN(str), *(uint32_t*)seed);
|
99
|
+
}
|
100
|
+
|
101
|
+
uint64_t
|
102
|
+
_murmur_s_digest64(int argc, VALUE *argv, VALUE klass, uint64_t (*process)(const char *, uint32_t, uint64_t))
|
103
|
+
{
|
104
|
+
VALUE str;
|
105
|
+
const char *seed;
|
106
|
+
|
107
|
+
if (argc < 1)
|
108
|
+
rb_raise(rb_eArgError, "no data given");
|
109
|
+
|
110
|
+
str = *argv;
|
111
|
+
|
112
|
+
StringValue(str);
|
113
|
+
|
114
|
+
if (1 < argc) {
|
115
|
+
StringValue(argv[1]);
|
116
|
+
if (RSTRING_LEN(argv[1]) != 8) {
|
117
|
+
rb_raise(rb_eArgError, "seed string should 64 bit chars");
|
118
|
+
}
|
119
|
+
seed = RSTRING_PTR(argv[1]);
|
120
|
+
} else {
|
121
|
+
seed = RSTRING_PTR(rb_const_get(klass, id_DEFAULT_SEED));
|
122
|
+
}
|
123
|
+
|
124
|
+
return process(RSTRING_PTR(str), RSTRING_LEN(str), *(uint64_t*)seed);
|
125
|
+
}
|
126
|
+
|
127
|
+
/*
|
128
|
+
* from https://github.com/ruby/ruby/blob/trunk/ext/digest/digest.c
|
129
|
+
* Copyright (C) 1995-2001 Yukihiro Matsumoto
|
130
|
+
* Copyright (C) 2001-2006 Akinori MUSHA
|
131
|
+
*/
|
132
|
+
VALUE
|
133
|
+
hexencode_str_new(VALUE str_digest)
|
134
|
+
{
|
135
|
+
char *digest;
|
136
|
+
size_t digest_len;
|
137
|
+
size_t i;
|
138
|
+
VALUE str;
|
139
|
+
char *p;
|
140
|
+
static const char hex[] = {
|
141
|
+
'0', '1', '2', '3', '4', '5', '6', '7', '8', '9',
|
142
|
+
'a', 'b', 'c', 'd', 'e', 'f'
|
143
|
+
};
|
144
|
+
|
145
|
+
StringValue(str_digest);
|
146
|
+
digest = RSTRING_PTR(str_digest);
|
147
|
+
digest_len = RSTRING_LEN(str_digest);
|
148
|
+
|
149
|
+
if (LONG_MAX / 2 < digest_len) {
|
150
|
+
rb_raise(rb_eRuntimeError, "digest string too long");
|
151
|
+
}
|
152
|
+
|
153
|
+
str = rb_usascii_str_new(0, digest_len * 2);
|
154
|
+
|
155
|
+
for (i = 0, p = RSTRING_PTR(str); i < digest_len; i++) {
|
156
|
+
unsigned char byte = digest[i];
|
157
|
+
|
158
|
+
p[i + i] = hex[byte >> 4];
|
159
|
+
p[i + i + 1] = hex[byte & 0x0f];
|
160
|
+
}
|
161
|
+
|
162
|
+
return str;
|
163
|
+
}
|
164
|
+
|
165
|
+
void
|
166
|
+
Init_murmurhash(void)
|
167
|
+
{
|
168
|
+
VALUE mDigest, cDigest_StringBuffer;
|
169
|
+
|
170
|
+
id_seed = rb_intern("seed");
|
171
|
+
id_DEFAULT_SEED = rb_intern("DEFAULT_SEED");
|
172
|
+
|
173
|
+
/* Digest::MurmurHash is require that Digest::StringBuffer */
|
174
|
+
mDigest = rb_path2class("Digest");
|
175
|
+
cDigest_StringBuffer = rb_path2class("Digest::StringBuffer");
|
176
|
+
|
177
|
+
/* class Digest::MurmurHash1 < Digest::StringBuffer */
|
178
|
+
cDigest_MurmurHash1 = rb_define_class_under(mDigest, "MurmurHash1", cDigest_StringBuffer);
|
179
|
+
rb_define_const(cDigest_MurmurHash1, "DEFAULT_SEED", rb_usascii_str_new(DEFAULT_SEED, 4));
|
180
|
+
rb_define_singleton_method(cDigest_MurmurHash1, "digest", murmur1_s_digest, -1);
|
181
|
+
rb_define_singleton_method(cDigest_MurmurHash1, "hexdigest", murmur1_s_hexdigest, -1);
|
182
|
+
rb_define_singleton_method(cDigest_MurmurHash1, "rawdigest", murmur1_s_rawdigest, -1);
|
183
|
+
rb_define_private_method(cDigest_MurmurHash1, "finish", murmur1_finish, 0);
|
184
|
+
rb_define_method(cDigest_MurmurHash1, "to_i", murmur1_to_i, 0);
|
185
|
+
rb_define_method(cDigest_MurmurHash1, "seed", murmur_seed_get32, 0);
|
186
|
+
rb_define_method(cDigest_MurmurHash1, "seed=", murmur_seed_set32, 1);
|
187
|
+
|
188
|
+
/* class Digest::MurmurHash2 < Digest::StringBuffer */
|
189
|
+
cDigest_MurmurHash2 = rb_define_class_under(mDigest, "MurmurHash2", cDigest_StringBuffer);
|
190
|
+
rb_define_const(cDigest_MurmurHash2, "DEFAULT_SEED", rb_usascii_str_new(DEFAULT_SEED, 4));
|
191
|
+
rb_define_singleton_method(cDigest_MurmurHash2, "digest", murmur2_s_digest, -1);
|
192
|
+
rb_define_singleton_method(cDigest_MurmurHash2, "hexdigest", murmur2_s_hexdigest, -1);
|
193
|
+
rb_define_singleton_method(cDigest_MurmurHash2, "rawdigest", murmur2_s_rawdigest, -1);
|
194
|
+
rb_define_private_method(cDigest_MurmurHash2, "finish", murmur2_finish, 0);
|
195
|
+
rb_define_method(cDigest_MurmurHash2, "to_i", murmur2_to_i, 0);
|
196
|
+
rb_define_method(cDigest_MurmurHash2, "seed", murmur_seed_get32, 0);
|
197
|
+
rb_define_method(cDigest_MurmurHash2, "seed=", murmur_seed_set32, 1);
|
198
|
+
|
199
|
+
/* class Digest::MurmurHash2A < Digest::StringBuffer */
|
200
|
+
cDigest_MurmurHash2A = rb_define_class_under(mDigest, "MurmurHash2A", cDigest_StringBuffer);
|
201
|
+
rb_define_const(cDigest_MurmurHash2A, "DEFAULT_SEED", rb_usascii_str_new(DEFAULT_SEED, 4));
|
202
|
+
rb_define_singleton_method(cDigest_MurmurHash2A, "digest", murmur2a_s_digest, -1);
|
203
|
+
rb_define_singleton_method(cDigest_MurmurHash2A, "hexdigest", murmur2a_s_hexdigest, -1);
|
204
|
+
rb_define_singleton_method(cDigest_MurmurHash2A, "rawdigest", murmur2a_s_rawdigest, -1);
|
205
|
+
rb_define_private_method(cDigest_MurmurHash2A, "finish", murmur2a_finish, 0);
|
206
|
+
rb_define_method(cDigest_MurmurHash2A, "to_i", murmur2a_to_i, 0);
|
207
|
+
rb_define_method(cDigest_MurmurHash2A, "seed", murmur_seed_get32, 0);
|
208
|
+
rb_define_method(cDigest_MurmurHash2A, "seed=", murmur_seed_set32, 1);
|
209
|
+
|
210
|
+
/* class Digest::MurmurHash64A < Digest::StringBuffer */
|
211
|
+
cDigest_MurmurHash64A = rb_define_class_under(mDigest, "MurmurHash64A", cDigest_StringBuffer);
|
212
|
+
rb_define_const(cDigest_MurmurHash64A, "DEFAULT_SEED", rb_usascii_str_new(DEFAULT_SEED, 8));
|
213
|
+
rb_define_singleton_method(cDigest_MurmurHash64A, "digest", murmur64a_s_digest, -1);
|
214
|
+
rb_define_singleton_method(cDigest_MurmurHash64A, "hexdigest", murmur64a_s_hexdigest, -1);
|
215
|
+
rb_define_singleton_method(cDigest_MurmurHash64A, "rawdigest", murmur64a_s_rawdigest, -1);
|
216
|
+
rb_define_private_method(cDigest_MurmurHash64A, "finish", murmur64a_finish, 0);
|
217
|
+
rb_define_method(cDigest_MurmurHash64A, "to_i", murmur64a_to_i, 0);
|
218
|
+
rb_define_method(cDigest_MurmurHash64A, "seed", murmur_seed_get64, 0);
|
219
|
+
rb_define_method(cDigest_MurmurHash64A, "seed=", murmur_seed_set64, 1);
|
220
|
+
|
221
|
+
/* class Digest::MurmurHash64B < Digest::StringBuffer */
|
222
|
+
cDigest_MurmurHash64B = rb_define_class_under(mDigest, "MurmurHash64B", cDigest_StringBuffer);
|
223
|
+
rb_define_const(cDigest_MurmurHash64B, "DEFAULT_SEED", rb_usascii_str_new(DEFAULT_SEED, 8));
|
224
|
+
rb_define_singleton_method(cDigest_MurmurHash64B, "digest", murmur64b_s_digest, -1);
|
225
|
+
rb_define_singleton_method(cDigest_MurmurHash64B, "hexdigest", murmur64b_s_hexdigest, -1);
|
226
|
+
rb_define_singleton_method(cDigest_MurmurHash64B, "rawdigest", murmur64b_s_rawdigest, -1);
|
227
|
+
rb_define_private_method(cDigest_MurmurHash64B, "finish", murmur64b_finish, 0);
|
228
|
+
rb_define_method(cDigest_MurmurHash64B, "to_i", murmur64b_to_i, 0);
|
229
|
+
rb_define_method(cDigest_MurmurHash64B, "seed", murmur_seed_get64, 0);
|
230
|
+
rb_define_method(cDigest_MurmurHash64B, "seed=", murmur_seed_set64, 1);
|
231
|
+
|
232
|
+
/* class Digest::MurmurHashNeutral2 < Digest::StringBuffer */
|
233
|
+
cDigest_MurmurHashNeutral2 = rb_define_class_under(mDigest, "MurmurHashNeutral2", cDigest_StringBuffer);
|
234
|
+
rb_define_const(cDigest_MurmurHashNeutral2, "DEFAULT_SEED", rb_usascii_str_new(DEFAULT_SEED, 4));
|
235
|
+
rb_define_singleton_method(cDigest_MurmurHashNeutral2, "digest", murmur_neutral2_s_digest, -1);
|
236
|
+
rb_define_singleton_method(cDigest_MurmurHashNeutral2, "hexdigest", murmur_neutral2_s_hexdigest, -1);
|
237
|
+
rb_define_singleton_method(cDigest_MurmurHashNeutral2, "rawdigest", murmur_neutral2_s_rawdigest, -1);
|
238
|
+
rb_define_private_method(cDigest_MurmurHashNeutral2, "finish", murmur_neutral2_finish, 0);
|
239
|
+
rb_define_method(cDigest_MurmurHashNeutral2, "to_i", murmur_neutral2_to_i, 0);
|
240
|
+
rb_define_method(cDigest_MurmurHashNeutral2, "seed", murmur_seed_get32, 0);
|
241
|
+
rb_define_method(cDigest_MurmurHashNeutral2, "seed=", murmur_seed_set32, 1);
|
242
|
+
|
243
|
+
/* class Digest::MurmurHashAligned2 < Digest::StringBuffer */
|
244
|
+
cDigest_MurmurHashAligned2 = rb_define_class_under(mDigest, "MurmurHashAligned2", cDigest_StringBuffer);
|
245
|
+
rb_define_const(cDigest_MurmurHashAligned2, "DEFAULT_SEED", rb_usascii_str_new(DEFAULT_SEED, 4));
|
246
|
+
rb_define_singleton_method(cDigest_MurmurHashAligned2, "digest", murmur_aligned2_s_digest, -1);
|
247
|
+
rb_define_singleton_method(cDigest_MurmurHashAligned2, "hexdigest", murmur_aligned2_s_hexdigest, -1);
|
248
|
+
rb_define_singleton_method(cDigest_MurmurHashAligned2, "rawdigest", murmur_aligned2_s_rawdigest, -1);
|
249
|
+
rb_define_private_method(cDigest_MurmurHashAligned2, "finish", murmur_aligned2_finish, 0);
|
250
|
+
rb_define_method(cDigest_MurmurHashAligned2, "to_i", murmur_aligned2_to_i, 0);
|
251
|
+
rb_define_method(cDigest_MurmurHashAligned2, "seed", murmur_seed_get32, 0);
|
252
|
+
rb_define_method(cDigest_MurmurHashAligned2, "seed=", murmur_seed_set32, 1);
|
253
|
+
}
|
@@ -3,7 +3,51 @@
|
|
3
3
|
|
4
4
|
#include "ruby.h"
|
5
5
|
|
6
|
+
#if defined(_MSC_VER)
|
7
|
+
# define BIG_CONSTANT(x) (x)
|
8
|
+
// Other compilers
|
9
|
+
#else // defined(_MSC_VER)
|
10
|
+
# define BIG_CONSTANT(x) (x##LLU)
|
11
|
+
#endif // !defined(_MSC_VER)
|
12
|
+
|
13
|
+
#define DEFAULT_SEED "\x00\x00\x00\x00\x00\x00\x00\x00"
|
6
14
|
#define MURMURHASH_MAGIC 0x5bd1e995
|
15
|
+
#define MURMURHASH_MAGIC64A BIG_CONSTANT(0xc6a4a7935bd1e995)
|
16
|
+
#if INTEGER_PACK_LITTLE_ENDIAN
|
17
|
+
# define ASSINE_BY_ENDIAN_32(digest, h) do { \
|
18
|
+
(digest)[3] = (h) >> 24; \
|
19
|
+
(digest)[2] = (h) >> 16; \
|
20
|
+
(digest)[1] = (h) >> 8; \
|
21
|
+
(digest)[0] = (h); \
|
22
|
+
} while (0)
|
23
|
+
# define ASSINE_BY_ENDIAN_64(digest, h) do { \
|
24
|
+
(digest)[7] = (h) >> 56; \
|
25
|
+
(digest)[6] = (h) >> 48; \
|
26
|
+
(digest)[5] = (h) >> 40; \
|
27
|
+
(digest)[4] = (h) >> 32; \
|
28
|
+
(digest)[3] = (h) >> 24; \
|
29
|
+
(digest)[2] = (h) >> 16; \
|
30
|
+
(digest)[1] = (h) >> 8; \
|
31
|
+
(digest)[0] = (h); \
|
32
|
+
} while (0)
|
33
|
+
#else
|
34
|
+
# define ASSINE_BY_ENDIAN_32(digest, h) do { \
|
35
|
+
(digest)[0] = (h) >> 24; \
|
36
|
+
(digest)[1] = (h) >> 16; \
|
37
|
+
(digest)[2] = (h) >> 8; \
|
38
|
+
(digest)[3] = (h); \
|
39
|
+
} while (0)
|
40
|
+
# define ASSINE_BY_ENDIAN_64(digest, h) do { \
|
41
|
+
(digest)[0] = (h) >> 56; \
|
42
|
+
(digest)[1] = (h) >> 48; \
|
43
|
+
(digest)[2] = (h) >> 40; \
|
44
|
+
(digest)[3] = (h) >> 32; \
|
45
|
+
(digest)[4] = (h) >> 24; \
|
46
|
+
(digest)[5] = (h) >> 16; \
|
47
|
+
(digest)[6] = (h) >> 8; \
|
48
|
+
(digest)[7] = (h); \
|
49
|
+
} while (0)
|
50
|
+
#endif
|
7
51
|
|
8
52
|
/* should be same type structure to digest/stringbuffer */
|
9
53
|
typedef struct {
|
@@ -19,43 +63,16 @@ typedef struct {
|
|
19
63
|
rb_raise(rb_eArgError, "NULL found for " # name " when shouldn't be.'"); \
|
20
64
|
}
|
21
65
|
|
22
|
-
|
23
|
-
|
24
|
-
* Copyright (C) 1995-2001 Yukihiro Matsumoto
|
25
|
-
* Copyright (C) 2001-2006 Akinori MUSHA
|
26
|
-
*/
|
27
|
-
static VALUE
|
28
|
-
hexencode_str_new(VALUE str_digest)
|
29
|
-
{
|
30
|
-
char *digest;
|
31
|
-
size_t digest_len;
|
32
|
-
size_t i;
|
33
|
-
VALUE str;
|
34
|
-
char *p;
|
35
|
-
static const char hex[] = {
|
36
|
-
'0', '1', '2', '3', '4', '5', '6', '7', '8', '9',
|
37
|
-
'a', 'b', 'c', 'd', 'e', 'f'
|
38
|
-
};
|
39
|
-
|
40
|
-
StringValue(str_digest);
|
41
|
-
digest = RSTRING_PTR(str_digest);
|
42
|
-
digest_len = RSTRING_LEN(str_digest);
|
43
|
-
|
44
|
-
if (LONG_MAX / 2 < digest_len) {
|
45
|
-
rb_raise(rb_eRuntimeError, "digest string too long");
|
46
|
-
}
|
47
|
-
|
48
|
-
str = rb_usascii_str_new(0, digest_len * 2);
|
66
|
+
VALUE murmur_seed_get32(VALUE self);
|
67
|
+
VALUE murmur_seed_get64(VALUE self);
|
49
68
|
|
50
|
-
|
51
|
-
unsigned char byte = digest[i];
|
52
|
-
|
53
|
-
p[i + i] = hex[byte >> 4];
|
54
|
-
p[i + i + 1] = hex[byte & 0x0f];
|
55
|
-
}
|
69
|
+
extern ID id_seed, id_DEFAULT_SEED;
|
56
70
|
|
57
|
-
|
58
|
-
|
71
|
+
uint32_t _murmur_finish32(VALUE self, uint32_t (*process)(const char *, uint32_t, uint32_t));
|
72
|
+
uint64_t _murmur_finish64(VALUE self, uint64_t (*process)(const char *, uint32_t, uint64_t));
|
73
|
+
uint32_t _murmur_s_digest32(int argc, VALUE *argv, VALUE klass, uint32_t (*process)(const char *, uint32_t, uint32_t));
|
74
|
+
uint64_t _murmur_s_digest64(int argc, VALUE *argv, VALUE klass, uint64_t (*process)(const char *, uint32_t, uint64_t));
|
75
|
+
VALUE hexencode_str_new(VALUE str_digest);
|
59
76
|
|
60
77
|
#endif /* ifndef MURMURHASH_INCLUDED */
|
61
78
|
|
@@ -4,27 +4,23 @@
|
|
4
4
|
|
5
5
|
#include "murmurhash1.h"
|
6
6
|
|
7
|
-
|
8
|
-
|
9
|
-
|
10
|
-
|
11
|
-
h *= m;
|
12
|
-
h ^= h >> r;
|
13
|
-
return h;
|
14
|
-
}
|
7
|
+
#define murmur1(r) do { \
|
8
|
+
h *= m; \
|
9
|
+
h ^= h >> r; \
|
10
|
+
} while(0)
|
15
11
|
|
16
12
|
static uint32_t
|
17
|
-
murmur_hash_process1(const char *data, uint32_t length)
|
13
|
+
murmur_hash_process1(const char *data, uint32_t length, uint32_t seed)
|
18
14
|
{
|
19
15
|
const uint32_t m = MURMURHASH_MAGIC;
|
20
16
|
const uint8_t r = 16;
|
21
17
|
uint32_t h;
|
22
18
|
|
23
|
-
h = length * m;
|
19
|
+
h = seed ^ (length * m);
|
24
20
|
|
25
21
|
while (4 <= length) {
|
26
22
|
h += *(uint32_t*)data;
|
27
|
-
|
23
|
+
murmur1(r);
|
28
24
|
data += 4;
|
29
25
|
length -= 4;
|
30
26
|
}
|
@@ -36,11 +32,11 @@ murmur_hash_process1(const char *data, uint32_t length)
|
|
36
32
|
h += data[1] << 8;
|
37
33
|
case 1:
|
38
34
|
h += data[0];
|
39
|
-
|
35
|
+
murmur1(r);
|
40
36
|
}
|
41
37
|
|
42
|
-
|
43
|
-
|
38
|
+
murmur1(10);
|
39
|
+
murmur1(17);
|
44
40
|
|
45
41
|
return h;
|
46
42
|
}
|
@@ -48,49 +44,27 @@ murmur_hash_process1(const char *data, uint32_t length)
|
|
48
44
|
VALUE
|
49
45
|
murmur1_finish(VALUE self)
|
50
46
|
{
|
51
|
-
uint32_t h;
|
52
47
|
uint8_t digest[4];
|
53
|
-
|
54
|
-
|
55
|
-
h = murmur_hash_process1(ptr->buffer, ptr->p - ptr->buffer);
|
56
|
-
|
57
|
-
digest[0] = h >> 24;
|
58
|
-
digest[1] = h >> 16;
|
59
|
-
digest[2] = h >> 8;
|
60
|
-
digest[3] = h;
|
48
|
+
uint64_t h;
|
61
49
|
|
50
|
+
h = _murmur_finish32(self, murmur_hash_process1);
|
51
|
+
ASSINE_BY_ENDIAN_32(digest, h);
|
62
52
|
return rb_str_new((const char*) digest, 4);
|
63
53
|
}
|
64
54
|
|
65
55
|
VALUE
|
66
56
|
murmur1_to_i(VALUE self)
|
67
57
|
{
|
68
|
-
|
69
|
-
return UINT2NUM(murmur_hash_process1(ptr->buffer, ptr->p - ptr->buffer));
|
58
|
+
return UINT2NUM(_murmur_finish32(self, murmur_hash_process1));
|
70
59
|
}
|
71
60
|
|
72
61
|
VALUE
|
73
62
|
murmur1_s_digest(int argc, VALUE *argv, VALUE klass)
|
74
63
|
{
|
75
|
-
VALUE str;
|
76
|
-
uint32_t h;
|
77
64
|
uint8_t digest[4];
|
78
|
-
|
79
|
-
|
80
|
-
|
81
|
-
|
82
|
-
str = *argv++;
|
83
|
-
argc--;
|
84
|
-
|
85
|
-
StringValue(str);
|
86
|
-
|
87
|
-
h = murmur_hash_process1(RSTRING_PTR(str), RSTRING_LEN(str));
|
88
|
-
|
89
|
-
digest[0] = h >> 24;
|
90
|
-
digest[1] = h >> 16;
|
91
|
-
digest[2] = h >> 8;
|
92
|
-
digest[3] = h;
|
93
|
-
|
65
|
+
uint32_t h;
|
66
|
+
h = _murmur_s_digest32(argc, argv, klass, murmur_hash_process1);
|
67
|
+
ASSINE_BY_ENDIAN_32(digest, h);
|
94
68
|
return rb_str_new((const char*) digest, 4);
|
95
69
|
}
|
96
70
|
|
@@ -103,15 +77,5 @@ murmur1_s_hexdigest(int argc, VALUE *argv, VALUE klass)
|
|
103
77
|
VALUE
|
104
78
|
murmur1_s_rawdigest(int argc, VALUE *argv, VALUE klass)
|
105
79
|
{
|
106
|
-
|
107
|
-
|
108
|
-
if (argc < 1)
|
109
|
-
rb_raise(rb_eArgError, "no data given");
|
110
|
-
|
111
|
-
str = *argv++;
|
112
|
-
argc--;
|
113
|
-
|
114
|
-
StringValue(str);
|
115
|
-
|
116
|
-
return UINT2NUM(murmur_hash_process1(RSTRING_PTR(str), RSTRING_LEN(str)));
|
80
|
+
return UINT2NUM(_murmur_s_digest32(argc, argv, klass, murmur_hash_process1));
|
117
81
|
}
|