similar_text 0.0.1
Sign up to get free protection for your applications and to get access to all the features.
- data/.gitignore +6 -0
- data/Gemfile +4 -0
- data/README.markdown +24 -0
- data/Rakefile +27 -0
- data/ext/similar_text/extconf.rb +5 -0
- data/ext/similar_text/similar_text.c +87 -0
- data/lib/similar_text.rb +8 -0
- data/lib/similar_text/version.rb +3 -0
- data/similar_text.gemspec +25 -0
- data/test/similar_text_test.rb +14 -0
- data/test/test_init.rb +3 -0
- metadata +58 -0
data/.gitignore
ADDED
data/Gemfile
ADDED
data/README.markdown
ADDED
@@ -0,0 +1,24 @@
|
|
1
|
+
Ruby similar_text
|
2
|
+
=================
|
3
|
+
|
4
|
+
Port of PHP similar_text function to Ruby, built as a native extension.
|
5
|
+
|
6
|
+
INSTALL
|
7
|
+
=======
|
8
|
+
|
9
|
+
You can install this extension via gems:
|
10
|
+
|
11
|
+
$ sudo gem install similar_text
|
12
|
+
|
13
|
+
USAGE
|
14
|
+
=====
|
15
|
+
|
16
|
+
Load extension:
|
17
|
+
|
18
|
+
require 'similar_text'
|
19
|
+
|
20
|
+
And use it by calling one of two methods (similar or similar?):
|
21
|
+
|
22
|
+
"Hello, World!".similar("Hello World!") # 96.0
|
23
|
+
"Hello, World!".similar?(Hello World!") # false
|
24
|
+
|
data/Rakefile
ADDED
@@ -0,0 +1,27 @@
|
|
1
|
+
require "bundler/gem_tasks"
|
2
|
+
require 'rubygems'
|
3
|
+
require 'rake'
|
4
|
+
require 'rake/extensiontask'
|
5
|
+
require 'rake/testtask'
|
6
|
+
|
7
|
+
task :default => [:test]
|
8
|
+
|
9
|
+
desc "Run tests"
|
10
|
+
Rake::TestTask.new("test") do |t|
|
11
|
+
t.pattern = 'test/*_test.rb'
|
12
|
+
t.libs << "test"
|
13
|
+
t.verbose = true
|
14
|
+
t.warning = true
|
15
|
+
end
|
16
|
+
|
17
|
+
Rake::ExtensionTask.new("similar_text") do |extension|
|
18
|
+
extension.lib_dir = "lib/similar_text"
|
19
|
+
end
|
20
|
+
|
21
|
+
task :chmod do
|
22
|
+
File.chmod(0775, 'lib/similar_text/similar_text.so')
|
23
|
+
end
|
24
|
+
task :build => [:clean, :compile, :chmod]
|
25
|
+
|
26
|
+
Bundler::GemHelper.install_tasks
|
27
|
+
|
@@ -0,0 +1,87 @@
|
|
1
|
+
#include <ruby.h>
|
2
|
+
#include <string.h>
|
3
|
+
|
4
|
+
static void similar_text_similar_str(const char *txt1, int len1, const char *txt2, int len2, int *pos1, int *pos2, int *max)
|
5
|
+
{
|
6
|
+
char *p, *q;
|
7
|
+
char *end1 = (char *) txt1 + len1;
|
8
|
+
char *end2 = (char *) txt2 + len2;
|
9
|
+
int l;
|
10
|
+
|
11
|
+
*max = 0;
|
12
|
+
for (p = (char *) txt1; p < end1; p++) {
|
13
|
+
for (q = (char *) txt2; q < end2; q++) {
|
14
|
+
for (l = 0; (p + l < end1) && (q + l < end2) && (p[l] == q[l]); l++);
|
15
|
+
if (l > *max) {
|
16
|
+
*max = l;
|
17
|
+
*pos1 = p - txt1;
|
18
|
+
*pos2 = q - txt2;
|
19
|
+
}
|
20
|
+
}
|
21
|
+
}
|
22
|
+
}
|
23
|
+
|
24
|
+
static int similar_text_similar_char(const char *txt1, int len1, const char *txt2, int len2)
|
25
|
+
{
|
26
|
+
int sum;
|
27
|
+
int pos1, pos2, max;
|
28
|
+
|
29
|
+
similar_text_similar_str(txt1, len1, txt2, len2, &pos1, &pos2, &max);
|
30
|
+
if ((sum = max)) {
|
31
|
+
if (pos1 && pos2) {
|
32
|
+
sum += similar_text_similar_char(txt1, pos1,
|
33
|
+
txt2, pos2);
|
34
|
+
}
|
35
|
+
if ((pos1 + max < len1) && (pos2 + max < len2)) {
|
36
|
+
sum += similar_text_similar_char(txt1 + pos1 + max, len1 - pos1 - max,
|
37
|
+
txt2 + pos2 + max, len2 - pos2 - max);
|
38
|
+
}
|
39
|
+
}
|
40
|
+
|
41
|
+
return sum;
|
42
|
+
}
|
43
|
+
|
44
|
+
int similar_text(const char *txt1, const char *txt2, double *percent)
|
45
|
+
{
|
46
|
+
size_t t1_len, t2_len;
|
47
|
+
int sim;
|
48
|
+
|
49
|
+
t1_len = strlen(txt1);
|
50
|
+
t2_len = strlen(txt2);
|
51
|
+
|
52
|
+
sim = similar_text_similar_char(txt1, t1_len, txt2, t2_len);
|
53
|
+
*percent = sim * 200.0 / (t1_len + t2_len);
|
54
|
+
|
55
|
+
return sim;
|
56
|
+
}
|
57
|
+
|
58
|
+
static VALUE t_similar(VALUE str1, VALUE str2)
|
59
|
+
{
|
60
|
+
double percent;
|
61
|
+
|
62
|
+
similar_text(StringValueCStr(str1), StringValueCStr(str2), &percent);
|
63
|
+
|
64
|
+
return rb_float_new(percent);
|
65
|
+
}
|
66
|
+
|
67
|
+
static VALUE t_similar_bool(VALUE str1, VALUE str2)
|
68
|
+
{
|
69
|
+
double percent;
|
70
|
+
|
71
|
+
similar_text(StringValueCStr(str1), StringValueCStr(str2), &percent);
|
72
|
+
|
73
|
+
if (percent == 100.0) {
|
74
|
+
return Qtrue;
|
75
|
+
}
|
76
|
+
else {
|
77
|
+
return Qfalse;
|
78
|
+
}
|
79
|
+
}
|
80
|
+
|
81
|
+
void Init_similar_text()
|
82
|
+
{
|
83
|
+
rb_cString = rb_define_class("String", rb_cObject);
|
84
|
+
rb_define_method(rb_cString, "similar", t_similar, 1);
|
85
|
+
rb_define_method(rb_cString, "similar?", t_similar_bool, 1);
|
86
|
+
}
|
87
|
+
|
data/lib/similar_text.rb
ADDED
@@ -0,0 +1,25 @@
|
|
1
|
+
# -*- encoding: utf-8 -*-
|
2
|
+
$:.push File.expand_path("../lib", __FILE__)
|
3
|
+
require "similar_text/version"
|
4
|
+
|
5
|
+
Gem::Specification.new do |s|
|
6
|
+
s.name = "similar_text"
|
7
|
+
s.version = SimilarText::VERSION
|
8
|
+
s.authors = ["Arthur Murauskas"]
|
9
|
+
s.email = ["arthur.murauskas@gmail.com"]
|
10
|
+
s.homepage = ""
|
11
|
+
s.summary = %q{Analogue of the similar_text function in PHP}
|
12
|
+
s.description = %q{Analogue of the similar_text function in PHP}
|
13
|
+
|
14
|
+
s.rubyforge_project = "similar_text"
|
15
|
+
|
16
|
+
s.files = `git ls-files`.split("\n")
|
17
|
+
s.test_files = `git ls-files -- {test,spec,features}/*`.split("\n")
|
18
|
+
s.executables = `git ls-files -- bin/*`.split("\n").map{ |f| File.basename(f) }
|
19
|
+
s.require_paths << "lib"
|
20
|
+
s.extensions << "ext/similar_text/extconf.rb"
|
21
|
+
|
22
|
+
# specify any dependencies here; for example:
|
23
|
+
# s.add_development_dependency "rspec"
|
24
|
+
# s.add_runtime_dependency "rest-client"
|
25
|
+
end
|
@@ -0,0 +1,14 @@
|
|
1
|
+
require 'test_init.rb'
|
2
|
+
|
3
|
+
class SimilarTextTest < Test::Unit::TestCase
|
4
|
+
def test_similar
|
5
|
+
assert_equal(100.0, "Hello, World!".similar("Hello, World!"))
|
6
|
+
assert_equal(96.0, "Hello, World!".similar("Hello World!"))
|
7
|
+
end
|
8
|
+
|
9
|
+
def test_similar?
|
10
|
+
assert_equal(true, "Hello, World!".similar?("Hello, World!"))
|
11
|
+
assert_equal(false, "Hello WORLD!".similar?("Hello, World!"))
|
12
|
+
end
|
13
|
+
end
|
14
|
+
|
data/test/test_init.rb
ADDED
metadata
ADDED
@@ -0,0 +1,58 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: similar_text
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.0.1
|
5
|
+
prerelease:
|
6
|
+
platform: ruby
|
7
|
+
authors:
|
8
|
+
- Arthur Murauskas
|
9
|
+
autorequire:
|
10
|
+
bindir: bin
|
11
|
+
cert_chain: []
|
12
|
+
date: 2011-09-12 00:00:00.000000000Z
|
13
|
+
dependencies: []
|
14
|
+
description: Analogue of the similar_text function in PHP
|
15
|
+
email:
|
16
|
+
- arthur.murauskas@gmail.com
|
17
|
+
executables: []
|
18
|
+
extensions:
|
19
|
+
- ext/similar_text/extconf.rb
|
20
|
+
extra_rdoc_files: []
|
21
|
+
files:
|
22
|
+
- .gitignore
|
23
|
+
- Gemfile
|
24
|
+
- README.markdown
|
25
|
+
- Rakefile
|
26
|
+
- ext/similar_text/extconf.rb
|
27
|
+
- ext/similar_text/similar_text.c
|
28
|
+
- lib/similar_text.rb
|
29
|
+
- lib/similar_text/version.rb
|
30
|
+
- similar_text.gemspec
|
31
|
+
- test/similar_text_test.rb
|
32
|
+
- test/test_init.rb
|
33
|
+
homepage: ''
|
34
|
+
licenses: []
|
35
|
+
post_install_message:
|
36
|
+
rdoc_options: []
|
37
|
+
require_paths:
|
38
|
+
- lib
|
39
|
+
- lib
|
40
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
41
|
+
none: false
|
42
|
+
requirements:
|
43
|
+
- - ! '>='
|
44
|
+
- !ruby/object:Gem::Version
|
45
|
+
version: '0'
|
46
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
47
|
+
none: false
|
48
|
+
requirements:
|
49
|
+
- - ! '>='
|
50
|
+
- !ruby/object:Gem::Version
|
51
|
+
version: '0'
|
52
|
+
requirements: []
|
53
|
+
rubyforge_project: similar_text
|
54
|
+
rubygems_version: 1.8.6
|
55
|
+
signing_key:
|
56
|
+
specification_version: 3
|
57
|
+
summary: Analogue of the similar_text function in PHP
|
58
|
+
test_files: []
|