ngramdistance-ffi 1.0.0
Sign up to get free protection for your applications and to get access to all the features.
- data/.gitignore +4 -0
- data/CHANGELOG.markdown +3 -0
- data/Gemfile +8 -0
- data/README.markdown +21 -0
- data/Rakefile +19 -0
- data/VERSION +1 -0
- data/ext/ngramdistance/.gitignore +3 -0
- data/ext/ngramdistance/extconf.rb +2 -0
- data/ext/ngramdistance/ngramdistance.c +140 -0
- data/ext/ngramdistance/ngramdistance.h +1 -0
- data/lib/ngramdistance.rb +12 -0
- data/ngramdistance-ffi.gemspec +64 -0
- data/spec/ngramdistance_spec.rb +19 -0
- data/spec/spec_helper.rb +3 -0
- metadata +112 -0
data/.gitignore
ADDED
data/CHANGELOG.markdown
ADDED
data/Gemfile
ADDED
data/README.markdown
ADDED
@@ -0,0 +1,21 @@
|
|
1
|
+
ngramdistance-ffi
|
2
|
+
===============
|
3
|
+
|
4
|
+
Converted to FFI by Bali for Ruby portability.
|
5
|
+
|
6
|
+
This gem originally based on ngram distance.
|
7
|
+
|
8
|
+
Tested on:
|
9
|
+
|
10
|
+
* MRI 1.9.2
|
11
|
+
|
12
|
+
Known Issues
|
13
|
+
============
|
14
|
+
* The C extension uses `char*` strings, and so Unicode strings will give incorrect distances.
|
15
|
+
|
16
|
+
Including in Gemfile
|
17
|
+
====================
|
18
|
+
|
19
|
+
gem ' ngramdistance-ffi', :require => ' ngramdistance'
|
20
|
+
|
21
|
+
- end -
|
data/Rakefile
ADDED
@@ -0,0 +1,19 @@
|
|
1
|
+
begin
|
2
|
+
require 'jeweler'
|
3
|
+
Jeweler::Tasks.new do |gemspec|
|
4
|
+
gemspec.name = " ngramdistance-ffi"
|
5
|
+
gemspec.summary = "An FFI version of the ngramdistance gem."
|
6
|
+
gemspec.description = "Provides a fast, cross-Ruby implementation of the ngramdistance distance algorithm."
|
7
|
+
gemspec.email = "dbalatero@gmail.com"
|
8
|
+
gemspec.homepage = "http://github.com/dbalatero/ ngramdistance-ffi"
|
9
|
+
gemspec.authors = ["Bali"]
|
10
|
+
gemspec.add_dependency "ffi"
|
11
|
+
gemspec.add_development_dependency "rspec"
|
12
|
+
gemspec.add_development_dependency "jeweler"
|
13
|
+
end
|
14
|
+
|
15
|
+
Jeweler::GemcutterTasks.new
|
16
|
+
rescue LoadError
|
17
|
+
puts "Jeweler not available. Install it with: gem install jeweler"
|
18
|
+
end
|
19
|
+
|
data/VERSION
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
1.0.0
|
@@ -0,0 +1,140 @@
|
|
1
|
+
#include <string.h>
|
2
|
+
#include <stdlib.h>
|
3
|
+
#include <stdio.h>
|
4
|
+
#ifdef NGRAM_CASE_INSENSITIVE
|
5
|
+
#include <ctype.h>
|
6
|
+
#define eq(x, y) (tolower(x) == tolower(y))
|
7
|
+
#else
|
8
|
+
#define eq(x, y) ((x) == (y))
|
9
|
+
#endif
|
10
|
+
|
11
|
+
#ifndef max
|
12
|
+
#define max( a, b ) ( ((a) > (b)) ? (a) : (b) )
|
13
|
+
#endif
|
14
|
+
|
15
|
+
#ifndef min
|
16
|
+
#define min( a, b ) ( ((a) < (b)) ? (a) : (b) )
|
17
|
+
#endif
|
18
|
+
|
19
|
+
|
20
|
+
char* substring(const char* str, size_t begin, size_t len)
|
21
|
+
{
|
22
|
+
if (str == 0 || strlen(str) == 0 || strlen(str) < begin || strlen(str) < (begin+len))
|
23
|
+
return 0;
|
24
|
+
|
25
|
+
return strndup(str + begin, len);
|
26
|
+
}
|
27
|
+
|
28
|
+
|
29
|
+
float ngramdistance (const char *source, const char *target,int n) {
|
30
|
+
int sl = strlen(source);
|
31
|
+
int tl = strlen(target);
|
32
|
+
|
33
|
+
if (sl == 0 || tl == 0) {
|
34
|
+
if (sl == tl) {
|
35
|
+
return 1;
|
36
|
+
}
|
37
|
+
else {
|
38
|
+
return 0;
|
39
|
+
}
|
40
|
+
}
|
41
|
+
int cost = 0;
|
42
|
+
if (sl < n || tl < n) {
|
43
|
+
int ni = min(sl,tl);
|
44
|
+
int i=0;
|
45
|
+
for (i=0;i<ni;i++) {
|
46
|
+
if (eq(source[i],target[i])) {
|
47
|
+
cost++;
|
48
|
+
}
|
49
|
+
}
|
50
|
+
return (float) cost/(float)max(sl, tl);
|
51
|
+
}
|
52
|
+
int char_len = sl+n-1;
|
53
|
+
char* sa = calloc(char_len+1,sizeof(char));
|
54
|
+
float* p; //'previous' cost array, horizontally
|
55
|
+
float* d; // cost array, horizontally
|
56
|
+
float* _d; //placeholder to assist in swapping p and d
|
57
|
+
|
58
|
+
//construct sa with prefix
|
59
|
+
int i=0;
|
60
|
+
for (i=0;i<char_len;i++) {
|
61
|
+
if (i < n-1) {
|
62
|
+
sa[i]=0; //add prefix
|
63
|
+
}
|
64
|
+
else {
|
65
|
+
sa[i] = source[i-n+1];
|
66
|
+
}
|
67
|
+
}
|
68
|
+
int float_arr_len = sl+1;
|
69
|
+
p = calloc( float_arr_len+1 , sizeof( float));
|
70
|
+
d = calloc( float_arr_len+1 , sizeof( float));
|
71
|
+
|
72
|
+
// indexes into strings s and t
|
73
|
+
i=0; // iterates through source
|
74
|
+
int j=0; // iterates through target
|
75
|
+
|
76
|
+
char* t_j = calloc(n+1,sizeof(char)); // jth n-gram of t
|
77
|
+
for (i = 0; i<=sl; i++) {
|
78
|
+
p[i] = i;
|
79
|
+
}
|
80
|
+
|
81
|
+
for (j = 1; j<=tl; j++) {
|
82
|
+
//construct t_j n-gram
|
83
|
+
if (j < n) {
|
84
|
+
int ti=0;
|
85
|
+
for (ti=0;ti<n-j;ti++) {
|
86
|
+
t_j[ti]=0; //add prefix
|
87
|
+
}
|
88
|
+
for (ti=n-j;ti<n;ti++) {
|
89
|
+
t_j[ti]=target[ti-(n-j)];
|
90
|
+
}
|
91
|
+
}
|
92
|
+
else {
|
93
|
+
free (t_j);
|
94
|
+
t_j = substring(target,j-n, n);
|
95
|
+
}
|
96
|
+
d[0] = j;
|
97
|
+
for (i=1; i<=sl; i++) {
|
98
|
+
cost = 0;
|
99
|
+
int tn=n;
|
100
|
+
//compare sa to t_j
|
101
|
+
int ni=0;
|
102
|
+
for (ni=0;ni<n;ni++) {
|
103
|
+
if (!eq(sa[i-1+ni] , t_j[ni])) {
|
104
|
+
cost++;
|
105
|
+
}
|
106
|
+
else if (eq(sa[i-1+ni], 0)) { //discount matches on prefix
|
107
|
+
tn--;
|
108
|
+
}
|
109
|
+
}
|
110
|
+
float ec = (float) cost/(float)tn;
|
111
|
+
// minimum of cell to the left+1, to the top+1, diagonally left and up +cost
|
112
|
+
d[i] = min(min(d[i-1]+1, p[i]+1), p[i-1]+ec);
|
113
|
+
}
|
114
|
+
// copy current distance counts to 'previous row' distance counts
|
115
|
+
_d = p;
|
116
|
+
p = d;
|
117
|
+
d = _d;
|
118
|
+
}
|
119
|
+
float val = p[sl];
|
120
|
+
free(p);
|
121
|
+
free(d);
|
122
|
+
free(t_j);
|
123
|
+
free(sa);
|
124
|
+
// our last action in the above loop was to switch d and p, so p now
|
125
|
+
// actually has the most recent cost counts
|
126
|
+
return 1.0f - ((float) val / (float)max(tl, sl));
|
127
|
+
}
|
128
|
+
|
129
|
+
|
130
|
+
#ifdef TEST
|
131
|
+
#include <stdio.h>
|
132
|
+
#include "ngramdistance.h"
|
133
|
+
|
134
|
+
int main (int argc, char **argv) {
|
135
|
+
float distance;
|
136
|
+
if (argc < 3) return -1;
|
137
|
+
distance = ngramdistance(argv[1], argv[2],3);
|
138
|
+
printf("%s vs %s: %f\n", argv[1], argv[2],distance);
|
139
|
+
}
|
140
|
+
#endif
|
@@ -0,0 +1 @@
|
|
1
|
+
unsigned ngramdistance(const char *, const char *, int n);
|
@@ -0,0 +1,12 @@
|
|
1
|
+
require 'ffi'
|
2
|
+
|
3
|
+
module NGramDistance
|
4
|
+
extend FFI::Library
|
5
|
+
|
6
|
+
# Try loading in order.
|
7
|
+
library = File.dirname(__FILE__) + "/../ext/ngramdistance/ngramdistance"
|
8
|
+
candidates = ['.bundle', '.so', '.dylib', ''].map { |ext| library + ext }
|
9
|
+
ffi_lib(candidates)
|
10
|
+
|
11
|
+
attach_function :distance, :ngramdistance, [:string, :string, :int], :float
|
12
|
+
end
|
@@ -0,0 +1,64 @@
|
|
1
|
+
# Generated by jeweler
|
2
|
+
# DO NOT EDIT THIS FILE DIRECTLY
|
3
|
+
# Instead, edit Jeweler::Tasks in Rakefile, and run the gemspec command
|
4
|
+
# -*- encoding: utf-8 -*-
|
5
|
+
|
6
|
+
Gem::Specification.new do |s|
|
7
|
+
s.name = %q{ngramdistance-ffi}
|
8
|
+
s.version = "1.0.0"
|
9
|
+
|
10
|
+
s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
|
11
|
+
s.authors = ["Bali"]
|
12
|
+
s.date = %q{2012-06-18}
|
13
|
+
s.description = %q{Provides a fast, cross-Ruby implementation of the ngramdistance distance algorithm.}
|
14
|
+
s.email = %q{mailbali@gmail.com}
|
15
|
+
s.extensions = ["ext/ngramdistance/extconf.rb"]
|
16
|
+
s.extra_rdoc_files = [
|
17
|
+
"README.markdown"
|
18
|
+
]
|
19
|
+
s.files = [
|
20
|
+
".gitignore",
|
21
|
+
"CHANGELOG.markdown",
|
22
|
+
"Gemfile",
|
23
|
+
"README.markdown",
|
24
|
+
"Rakefile",
|
25
|
+
"VERSION",
|
26
|
+
"ext/ngramdistance/.gitignore",
|
27
|
+
"ext/ngramdistance/extconf.rb",
|
28
|
+
"ext/ngramdistance/ngramdistance.c",
|
29
|
+
"ext/ngramdistance/ngramdistance.h",
|
30
|
+
"ngramdistance-ffi.gemspec",
|
31
|
+
"lib/ngramdistance.rb",
|
32
|
+
"spec/ngramdistance_spec.rb",
|
33
|
+
"spec/spec_helper.rb"
|
34
|
+
]
|
35
|
+
s.homepage = %q{https://github.com/pecbali/ngramdistance-ffi}
|
36
|
+
s.rdoc_options = ["--charset=UTF-8"]
|
37
|
+
s.require_paths = ["lib"]
|
38
|
+
s.rubygems_version = %q{1.3.7}
|
39
|
+
s.summary = %q{An FFI version of the ngramdistance gem.}
|
40
|
+
s.test_files = [
|
41
|
+
"spec/ngramdistance_spec.rb",
|
42
|
+
"spec/spec_helper.rb"
|
43
|
+
]
|
44
|
+
|
45
|
+
if s.respond_to? :specification_version then
|
46
|
+
current_version = Gem::Specification::CURRENT_SPECIFICATION_VERSION
|
47
|
+
s.specification_version = 3
|
48
|
+
|
49
|
+
if Gem::Version.new(Gem::VERSION) >= Gem::Version.new('1.2.0') then
|
50
|
+
s.add_runtime_dependency(%q<ffi>, [">= 0"])
|
51
|
+
s.add_development_dependency(%q<rspec>, [">= 0"])
|
52
|
+
s.add_development_dependency(%q<jeweler>, [">= 0"])
|
53
|
+
else
|
54
|
+
s.add_dependency(%q<ffi>, [">= 0"])
|
55
|
+
s.add_dependency(%q<rspec>, [">= 0"])
|
56
|
+
s.add_dependency(%q<jeweler>, [">= 0"])
|
57
|
+
end
|
58
|
+
else
|
59
|
+
s.add_dependency(%q<ffi>, [">= 0"])
|
60
|
+
s.add_dependency(%q<rspec>, [">= 0"])
|
61
|
+
s.add_dependency(%q<jeweler>, [">= 0"])
|
62
|
+
end
|
63
|
+
end
|
64
|
+
|
@@ -0,0 +1,19 @@
|
|
1
|
+
require 'spec_helper'
|
2
|
+
|
3
|
+
describe NGramDistance do
|
4
|
+
fixtures = [
|
5
|
+
["university", "univearsitty", 0.750000,
|
6
|
+
["university", "university", 1.0],
|
7
|
+
["hello", "jello",0.633333],
|
8
|
+
["hello", "heloll", 0.666667],
|
9
|
+
["hello", "saint", 0.0000],
|
10
|
+
["hello", "", 0.0000]
|
11
|
+
]
|
12
|
+
|
13
|
+
fixtures.each do |w1, w2, d|
|
14
|
+
it "should calculate a distance of #{d} between #{w1} and #{w2}" do
|
15
|
+
NGramDistance.distance(w1, w2,3).should be_close( d,0.05)
|
16
|
+
NGramDistance.distance(w2, w1,3).should be_close d,0.05)
|
17
|
+
end
|
18
|
+
end
|
19
|
+
end
|
data/spec/spec_helper.rb
ADDED
metadata
ADDED
@@ -0,0 +1,112 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: ngramdistance-ffi
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 1.0.0
|
5
|
+
prerelease:
|
6
|
+
platform: ruby
|
7
|
+
authors:
|
8
|
+
- Bali
|
9
|
+
autorequire:
|
10
|
+
bindir: bin
|
11
|
+
cert_chain: []
|
12
|
+
date: 2012-06-18 00:00:00.000000000 Z
|
13
|
+
dependencies:
|
14
|
+
- !ruby/object:Gem::Dependency
|
15
|
+
name: ffi
|
16
|
+
requirement: !ruby/object:Gem::Requirement
|
17
|
+
none: false
|
18
|
+
requirements:
|
19
|
+
- - ! '>='
|
20
|
+
- !ruby/object:Gem::Version
|
21
|
+
version: '0'
|
22
|
+
type: :runtime
|
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: rspec
|
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
|
+
- !ruby/object:Gem::Dependency
|
47
|
+
name: jeweler
|
48
|
+
requirement: !ruby/object:Gem::Requirement
|
49
|
+
none: false
|
50
|
+
requirements:
|
51
|
+
- - ! '>='
|
52
|
+
- !ruby/object:Gem::Version
|
53
|
+
version: '0'
|
54
|
+
type: :development
|
55
|
+
prerelease: false
|
56
|
+
version_requirements: !ruby/object:Gem::Requirement
|
57
|
+
none: false
|
58
|
+
requirements:
|
59
|
+
- - ! '>='
|
60
|
+
- !ruby/object:Gem::Version
|
61
|
+
version: '0'
|
62
|
+
description: Provides a fast, cross-Ruby implementation of the ngramdistance distance
|
63
|
+
algorithm.
|
64
|
+
email: mailbali@gmail.com
|
65
|
+
executables: []
|
66
|
+
extensions:
|
67
|
+
- ext/ngramdistance/extconf.rb
|
68
|
+
extra_rdoc_files:
|
69
|
+
- README.markdown
|
70
|
+
files:
|
71
|
+
- .gitignore
|
72
|
+
- CHANGELOG.markdown
|
73
|
+
- Gemfile
|
74
|
+
- README.markdown
|
75
|
+
- Rakefile
|
76
|
+
- VERSION
|
77
|
+
- ext/ngramdistance/.gitignore
|
78
|
+
- ext/ngramdistance/extconf.rb
|
79
|
+
- ext/ngramdistance/ngramdistance.c
|
80
|
+
- ext/ngramdistance/ngramdistance.h
|
81
|
+
- ngramdistance-ffi.gemspec
|
82
|
+
- lib/ngramdistance.rb
|
83
|
+
- spec/ngramdistance_spec.rb
|
84
|
+
- spec/spec_helper.rb
|
85
|
+
homepage: https://github.com/pecbali/ngramdistance-ffi
|
86
|
+
licenses: []
|
87
|
+
post_install_message:
|
88
|
+
rdoc_options:
|
89
|
+
- --charset=UTF-8
|
90
|
+
require_paths:
|
91
|
+
- lib
|
92
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
93
|
+
none: false
|
94
|
+
requirements:
|
95
|
+
- - ! '>='
|
96
|
+
- !ruby/object:Gem::Version
|
97
|
+
version: '0'
|
98
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
99
|
+
none: false
|
100
|
+
requirements:
|
101
|
+
- - ! '>='
|
102
|
+
- !ruby/object:Gem::Version
|
103
|
+
version: '0'
|
104
|
+
requirements: []
|
105
|
+
rubyforge_project:
|
106
|
+
rubygems_version: 1.8.24
|
107
|
+
signing_key:
|
108
|
+
specification_version: 3
|
109
|
+
summary: An FFI version of the ngramdistance gem.
|
110
|
+
test_files:
|
111
|
+
- spec/ngramdistance_spec.rb
|
112
|
+
- spec/spec_helper.rb
|