cmigemo 1.0.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- data/History.txt +6 -0
- data/Manifest.txt +7 -0
- data/README.rdoc +68 -0
- data/Rakefile +74 -0
- data/ext/cmigemo.c +117 -0
- data/ext/extconf.rb +6 -0
- data/test/test_cmigemo.rb +41 -0
- metadata +77 -0
data/History.txt
ADDED
data/Manifest.txt
ADDED
data/README.rdoc
ADDED
@@ -0,0 +1,68 @@
|
|
1
|
+
= Ruby/CMigemo
|
2
|
+
|
3
|
+
* http://github.com/Constellation/ruby-cmigemo
|
4
|
+
|
5
|
+
== DESCRIPTION:
|
6
|
+
|
7
|
+
Ruby Binding for C/Migemo
|
8
|
+
|
9
|
+
== FEATURES/PROBLEMS:
|
10
|
+
|
11
|
+
* get regexp text which match to Japanese phrase
|
12
|
+
* get regexp text faster
|
13
|
+
* load dict file which have any encoding
|
14
|
+
|
15
|
+
== SYNOPSIS:
|
16
|
+
|
17
|
+
require 'cmigemo'
|
18
|
+
$KCODE = 'u'
|
19
|
+
dict_path = File.expand_path('~/dict')
|
20
|
+
cmigemo = CMigemo.new
|
21
|
+
{
|
22
|
+
"migemo-dict" => CMigemo::MIGEMO,
|
23
|
+
"han2zen.dat" => CMigemo::HAN2ZEN,
|
24
|
+
"hira2kata.dat" => CMigemo::HIRA2KATA,
|
25
|
+
"roma2hira.dat" => CMigemo::ROMA2HIRA,
|
26
|
+
}.each do |dict_file, type|
|
27
|
+
cmigemo.load(type, File.join(dict_path, dict_file));
|
28
|
+
end
|
29
|
+
|
30
|
+
# get regexp text
|
31
|
+
reg = Regexp.new(cmigemo.query('seiza'))
|
32
|
+
|
33
|
+
puts "match!" if "星座" =~ reg
|
34
|
+
puts "match!" if "正座" =~ reg
|
35
|
+
|
36
|
+
== REQUIREMENTS:
|
37
|
+
|
38
|
+
* C/Migemo
|
39
|
+
|
40
|
+
== INSTALL:
|
41
|
+
|
42
|
+
gem source -a http://gemcutter.org
|
43
|
+
sudo gem install cmigemo
|
44
|
+
|
45
|
+
== LICENSE:
|
46
|
+
|
47
|
+
(The MIT License)
|
48
|
+
|
49
|
+
Copyright (c) 2009 Constellation
|
50
|
+
|
51
|
+
Permission is hereby granted, free of charge, to any person obtaining
|
52
|
+
a copy of this software and associated documentation files (the
|
53
|
+
'Software'), to deal in the Software without restriction, including
|
54
|
+
without limitation the rights to use, copy, modify, merge, publish,
|
55
|
+
distribute, sublicense, and/or sell copies of the Software, and to
|
56
|
+
permit persons to whom the Software is furnished to do so, subject to
|
57
|
+
the following conditions:
|
58
|
+
|
59
|
+
The above copyright notice and this permission notice shall be
|
60
|
+
included in all copies or substantial portions of the Software.
|
61
|
+
|
62
|
+
THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND,
|
63
|
+
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
64
|
+
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
|
65
|
+
IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
|
66
|
+
CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
|
67
|
+
TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
|
68
|
+
SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
data/Rakefile
ADDED
@@ -0,0 +1,74 @@
|
|
1
|
+
# -*- ruby -*-
|
2
|
+
require 'rubygems'
|
3
|
+
require 'hoe'
|
4
|
+
require 'pp'
|
5
|
+
|
6
|
+
$version = '1.0.0'
|
7
|
+
$readme = 'README.rdoc'
|
8
|
+
$rdoc_opts = %W(--main #{$readme} --charset utf-8 --line-numbers)
|
9
|
+
|
10
|
+
Hoe.spec'cmigemo' do |p|
|
11
|
+
developer('Constellation', 'utatane.tea@gmail.com')
|
12
|
+
p.description = <<-EOF
|
13
|
+
Ruby binding for C/Migemo
|
14
|
+
use any encoding dictionary
|
15
|
+
faster response
|
16
|
+
EOF
|
17
|
+
p.name = 'cmigemo'
|
18
|
+
p.readme_file = $readme
|
19
|
+
p.extra_rdoc_files = FileList[$readme]
|
20
|
+
p.need_tar = false
|
21
|
+
p.need_zip = false
|
22
|
+
p.summary = 'Ruby binding for C/Migemo'
|
23
|
+
p.test_globs = 'spec/**/*_spec.rb'
|
24
|
+
p.url = 'http://github.com/Constellation/ruby-cmigemo'
|
25
|
+
p.version = $version
|
26
|
+
p.spec_extras[:extensions] = FileList["ext/extconf.rb"].to_a
|
27
|
+
spec_extras[:rdoc_options] = $rdoc_opts
|
28
|
+
# p.spec_extras[:rdoc_options] = $rdoc_opts
|
29
|
+
#p.clean_globs =
|
30
|
+
# self.rubyforge_name = 'cmigemox' # if different than 'cmigemo'
|
31
|
+
end
|
32
|
+
|
33
|
+
ext = "ext"
|
34
|
+
ext_so = "ext/cmigemo.so"
|
35
|
+
ext_files = FileList[
|
36
|
+
"ext/cmigemo.c",
|
37
|
+
"ext/extconf.rb",
|
38
|
+
"ext/Makefile",
|
39
|
+
"lib"
|
40
|
+
]
|
41
|
+
|
42
|
+
# ext tasks
|
43
|
+
desc "build ext"
|
44
|
+
task :ext => ["ext/Makefile", ext_so]
|
45
|
+
file "ext/Makefile" => ["ext/extconf.rb"] do
|
46
|
+
Dir.chdir(ext) do
|
47
|
+
ruby "extconf.rb"
|
48
|
+
end
|
49
|
+
end
|
50
|
+
file ext_so => ext_files do
|
51
|
+
Dir.chdir(ext) do
|
52
|
+
sh 'make'
|
53
|
+
end
|
54
|
+
cp ext_so, "lib"
|
55
|
+
end
|
56
|
+
directory "lib"
|
57
|
+
|
58
|
+
#namespace :dict do
|
59
|
+
# desc "get cp932 dict"
|
60
|
+
# task :cp932 do
|
61
|
+
# puts "get cp932 dict"
|
62
|
+
# end
|
63
|
+
# desc "get utf-8 dict"
|
64
|
+
# task :utf8 do
|
65
|
+
# puts "get utf8 dict"
|
66
|
+
# end
|
67
|
+
# desc "get euc-jp dict"
|
68
|
+
# task :eucjp do
|
69
|
+
# puts "get eucjp dict"
|
70
|
+
# end
|
71
|
+
#end
|
72
|
+
|
73
|
+
|
74
|
+
# vim: syntax=ruby
|
data/ext/cmigemo.c
ADDED
@@ -0,0 +1,117 @@
|
|
1
|
+
#include "ruby.h"
|
2
|
+
#include "migemo.h"
|
3
|
+
|
4
|
+
struct RMigemo {
|
5
|
+
int open;
|
6
|
+
migemo *migemo;
|
7
|
+
};
|
8
|
+
|
9
|
+
static void rb_cmigemo_free(struct RMigemo*);
|
10
|
+
static VALUE rb_cmigemo_close(VALUE);
|
11
|
+
static VALUE rb_cmigemo_s_allocate(VALUE);
|
12
|
+
static VALUE rb_cmigemo_initialize(VALUE);
|
13
|
+
static VALUE rb_cmigemo_load(VALUE, VALUE, VALUE);
|
14
|
+
static VALUE rb_cmigemo_query(VALUE, VALUE);
|
15
|
+
|
16
|
+
static VALUE cCMigemo;
|
17
|
+
|
18
|
+
static VALUE
|
19
|
+
rb_cmigemo_s_allocate(VALUE klass)
|
20
|
+
{
|
21
|
+
VALUE obj;
|
22
|
+
struct RMigemo *rm;
|
23
|
+
obj = Data_Make_Struct(klass, struct RMigemo, NULL, rb_cmigemo_free, rm);
|
24
|
+
rm->open = 0;
|
25
|
+
rm->migemo = NULL;
|
26
|
+
return obj;
|
27
|
+
}
|
28
|
+
|
29
|
+
static void rb_cmigemo_free(struct RMigemo *rm)
|
30
|
+
{
|
31
|
+
if(rm->open) migemo_close(rm->migemo);
|
32
|
+
free(rm);
|
33
|
+
}
|
34
|
+
|
35
|
+
static VALUE
|
36
|
+
rb_cmigemo_initialize(VALUE self)
|
37
|
+
{
|
38
|
+
char *dict;
|
39
|
+
struct RMigemo *rm;
|
40
|
+
int result;
|
41
|
+
|
42
|
+
Data_Get_Struct(self, struct RMigemo, rm);
|
43
|
+
rm->migemo = migemo_open(NULL);
|
44
|
+
rm->open = 1;
|
45
|
+
|
46
|
+
return self;
|
47
|
+
}
|
48
|
+
|
49
|
+
static VALUE
|
50
|
+
rb_cmigemo_close(VALUE self)
|
51
|
+
{
|
52
|
+
struct RMigemo *rm;
|
53
|
+
Data_Get_Struct(self, struct RMigemo, rm);
|
54
|
+
if(rm->open){
|
55
|
+
migemo_close(rm->migemo);
|
56
|
+
rm->open = 0;
|
57
|
+
}
|
58
|
+
return Qnil;
|
59
|
+
}
|
60
|
+
|
61
|
+
static VALUE
|
62
|
+
rb_cmigemo_query(VALUE self, VALUE rstr)
|
63
|
+
{
|
64
|
+
struct RMigemo *rm;
|
65
|
+
unsigned char* str;
|
66
|
+
migemo *m;
|
67
|
+
VALUE result;
|
68
|
+
|
69
|
+
Data_Get_Struct(self, struct RMigemo, rm);
|
70
|
+
if(rm->open){
|
71
|
+
str = migemo_query(rm->migemo, (unsigned char*)StringValuePtr(rstr));
|
72
|
+
result = rb_tainted_str_new2(str);
|
73
|
+
migemo_release(rm->migemo, str);
|
74
|
+
return result;
|
75
|
+
} else {
|
76
|
+
rb_raise(rb_eException, "migemo closed");
|
77
|
+
}
|
78
|
+
}
|
79
|
+
|
80
|
+
static VALUE
|
81
|
+
rb_cmigemo_load(VALUE self, VALUE id, VALUE rstr)
|
82
|
+
{
|
83
|
+
struct RMigemo *rm;
|
84
|
+
int result;
|
85
|
+
int dict_id = NUM2INT(id);
|
86
|
+
char *dict = StringValuePtr(rstr);
|
87
|
+
Data_Get_Struct(self, struct RMigemo, rm);
|
88
|
+
|
89
|
+
if(rm->open){
|
90
|
+
result = migemo_load(rm->migemo, dict_id, dict);
|
91
|
+
if(result == MIGEMO_DICTID_INVALID){
|
92
|
+
rb_raise(rb_eArgError, "invalid dict file");
|
93
|
+
}
|
94
|
+
|
95
|
+
return INT2NUM(result);
|
96
|
+
} else {
|
97
|
+
rb_raise(rb_eException, "migemo closed");
|
98
|
+
}
|
99
|
+
}
|
100
|
+
|
101
|
+
void Init_cmigemo()
|
102
|
+
{
|
103
|
+
cCMigemo = rb_define_class("CMigemo", rb_cData);
|
104
|
+
|
105
|
+
rb_define_alloc_func(cCMigemo, rb_cmigemo_s_allocate);
|
106
|
+
rb_define_private_method(cCMigemo, "initialize", rb_cmigemo_initialize, 0);
|
107
|
+
rb_define_method(cCMigemo, "query", rb_cmigemo_query, 1);
|
108
|
+
rb_define_method(cCMigemo, "load", rb_cmigemo_load, 2);
|
109
|
+
rb_define_method(cCMigemo, "close", rb_cmigemo_close, 0);
|
110
|
+
|
111
|
+
rb_define_const(cCMigemo, "MIGEMO", INT2NUM(MIGEMO_DICTID_MIGEMO));
|
112
|
+
rb_define_const(cCMigemo, "ROMA2HIRA", INT2NUM(MIGEMO_DICTID_ROMA2HIRA));
|
113
|
+
rb_define_const(cCMigemo, "HIRA2KATA", INT2NUM(MIGEMO_DICTID_HIRA2KATA));
|
114
|
+
rb_define_const(cCMigemo, "HAN2ZEN", INT2NUM(MIGEMO_DICTID_HAN2ZEN));
|
115
|
+
rb_define_const(cCMigemo, "ZEN2HAN", INT2NUM(MIGEMO_DICTID_ZEN2HAN));
|
116
|
+
}
|
117
|
+
|
data/ext/extconf.rb
ADDED
@@ -0,0 +1,41 @@
|
|
1
|
+
require File.dirname(__FILE__) + '/../lib/cmigemo'
|
2
|
+
$KCODE = 'u'
|
3
|
+
class TestCmigemo < Test::Unit::TestCase
|
4
|
+
|
5
|
+
def setup
|
6
|
+
@cmigemo = CMigemo::new
|
7
|
+
d = File.expand_path(File.join(File.dirname(__FILE__), '../dict/utf-8'))
|
8
|
+
@dicts = {
|
9
|
+
File.join(d, "migemo-dict") => CMigemo::MIGEMO,
|
10
|
+
File.join(d, "han2zen.dat") => CMigemo::HAN2ZEN,
|
11
|
+
File.join(d, "hira2kata.dat") => CMigemo::HIRA2KATA,
|
12
|
+
File.join(d, "roma2hira.dat") => CMigemo::ROMA2HIRA,
|
13
|
+
}
|
14
|
+
end
|
15
|
+
|
16
|
+
def test_new
|
17
|
+
assert_instance_of CMigemo, @cmigemo
|
18
|
+
end
|
19
|
+
|
20
|
+
def load_dicts c=@cmigemo
|
21
|
+
@dicts.each do |dict, type|
|
22
|
+
c.load(type, dict)
|
23
|
+
end
|
24
|
+
end
|
25
|
+
|
26
|
+
def test_query
|
27
|
+
load_dicts
|
28
|
+
reg = Regexp.new(@cmigemo.query 'seiza')
|
29
|
+
assert reg =~ "正座"
|
30
|
+
assert reg =~ "星座"
|
31
|
+
end
|
32
|
+
|
33
|
+
def test_close
|
34
|
+
cmigemo = CMigemo::new
|
35
|
+
cmigemo.close
|
36
|
+
assert_raise Exception do
|
37
|
+
cmigemo.query 'seiza'
|
38
|
+
end
|
39
|
+
end
|
40
|
+
|
41
|
+
end
|
metadata
ADDED
@@ -0,0 +1,77 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: cmigemo
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 1.0.0
|
5
|
+
platform: ruby
|
6
|
+
authors:
|
7
|
+
- Constellation
|
8
|
+
autorequire:
|
9
|
+
bindir: bin
|
10
|
+
cert_chain: []
|
11
|
+
|
12
|
+
date: 2009-10-09 00:00:00 +09:00
|
13
|
+
default_executable:
|
14
|
+
dependencies:
|
15
|
+
- !ruby/object:Gem::Dependency
|
16
|
+
name: hoe
|
17
|
+
type: :development
|
18
|
+
version_requirement:
|
19
|
+
version_requirements: !ruby/object:Gem::Requirement
|
20
|
+
requirements:
|
21
|
+
- - ">="
|
22
|
+
- !ruby/object:Gem::Version
|
23
|
+
version: 2.3.3
|
24
|
+
version:
|
25
|
+
description: " Ruby binding for C/Migemo\n use any encoding dictionary\n faster response\n"
|
26
|
+
email:
|
27
|
+
- utatane.tea@gmail.com
|
28
|
+
executables: []
|
29
|
+
|
30
|
+
extensions:
|
31
|
+
- ext/extconf.rb
|
32
|
+
extra_rdoc_files:
|
33
|
+
- History.txt
|
34
|
+
- Manifest.txt
|
35
|
+
- README.rdoc
|
36
|
+
files:
|
37
|
+
- History.txt
|
38
|
+
- Manifest.txt
|
39
|
+
- README.rdoc
|
40
|
+
- Rakefile
|
41
|
+
- ext/cmigemo.c
|
42
|
+
- ext/extconf.rb
|
43
|
+
- test/test_cmigemo.rb
|
44
|
+
has_rdoc: true
|
45
|
+
homepage: http://github.com/Constellation/ruby-cmigemo
|
46
|
+
licenses: []
|
47
|
+
|
48
|
+
post_install_message:
|
49
|
+
rdoc_options:
|
50
|
+
- --main
|
51
|
+
- README.rdoc
|
52
|
+
- --charset
|
53
|
+
- utf-8
|
54
|
+
- --line-numbers
|
55
|
+
require_paths:
|
56
|
+
- ext
|
57
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
58
|
+
requirements:
|
59
|
+
- - ">="
|
60
|
+
- !ruby/object:Gem::Version
|
61
|
+
version: "0"
|
62
|
+
version:
|
63
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
64
|
+
requirements:
|
65
|
+
- - ">="
|
66
|
+
- !ruby/object:Gem::Version
|
67
|
+
version: "0"
|
68
|
+
version:
|
69
|
+
requirements: []
|
70
|
+
|
71
|
+
rubyforge_project: cmigemo
|
72
|
+
rubygems_version: 1.3.5
|
73
|
+
signing_key:
|
74
|
+
specification_version: 3
|
75
|
+
summary: Ruby binding for C/Migemo
|
76
|
+
test_files: []
|
77
|
+
|