rubyrubi 0.2.0 → 0.3.0
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.
- checksums.yaml +4 -4
- data/lib/rubyrubi/client.rb +4 -0
- data/lib/rubyrubi/parser/base.rb +29 -22
- data/lib/rubyrubi/version.rb +1 -1
- metadata +2 -2
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: b005848f15f70889efb8ea6ab3fa98105e880721
|
4
|
+
data.tar.gz: 86f9fc910744efc2d59b0a9849b472add0277ee2
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 033a61ebedf61ae3f2f2fe5dd762d9a8929f7bca7e25395a9ff045bb253ac20e9abdd3c9c99b1fad04a761454306ba352149970948b6a3708fc20a95f427cc45
|
7
|
+
data.tar.gz: ee1ce99f822b8de75a6d008a58ef5d6991e02139ab3c1615331957c18d6809c0c06432946b0127c53aede60826906cb02a07f765ff08d34fb8107143377d74ef
|
data/lib/rubyrubi/client.rb
CHANGED
data/lib/rubyrubi/parser/base.rb
CHANGED
@@ -1,15 +1,18 @@
|
|
1
1
|
module Rubyrubi
|
2
2
|
module Parser
|
3
|
-
KANJI =
|
4
|
-
OKURI = /[\p{hiragana}\p{katakana}
|
3
|
+
KANJI = /[一-龠]+/
|
4
|
+
OKURI = /[\p{hiragana}\p{katakana}ー=〜・]+/
|
5
5
|
KOMEI = /固名商品/
|
6
6
|
|
7
|
+
def self.ruby_tag(text, ruby)
|
8
|
+
"<ruby>#{text}<rp>(</rp><rt>#{ruby}</rt><rp>)</rp></ruby>"
|
9
|
+
end
|
10
|
+
|
7
11
|
class Base
|
8
12
|
def initialize(result)
|
9
13
|
result = result.with_indifferent_access
|
10
14
|
@result = result
|
11
15
|
@original = result[:ResultSet][:ma_result][:word_list][:word]
|
12
|
-
#p @original
|
13
16
|
end
|
14
17
|
|
15
18
|
def parse()
|
@@ -20,34 +23,38 @@ module Rubyrubi
|
|
20
23
|
end
|
21
24
|
|
22
25
|
def add_rubi_and_okuri(word)
|
23
|
-
#p word
|
24
26
|
if word['feature'] && word['feature'].scan(KOMEI).size > 0
|
25
27
|
return word.clone
|
26
28
|
end
|
27
29
|
|
28
|
-
kanji = word['surface'].scan(KANJI)
|
29
|
-
|
30
|
-
|
31
|
-
return word.clone
|
32
|
-
end
|
30
|
+
kanji = word['surface'].scan(KANJI)
|
31
|
+
okuri = word['surface'].scan(OKURI)
|
32
|
+
yomi = '' + word['reading']
|
33
33
|
|
34
|
-
|
35
|
-
rubi = kanji == nil && okuri == nil ? nil
|
36
|
-
: word['reading'].gsub(Regexp.new("#{okuri}$"), '')
|
34
|
+
return word.clone if !kanji || kanji.size == 0
|
37
35
|
|
38
|
-
|
39
|
-
'kanji' =>
|
40
|
-
|
41
|
-
|
42
|
-
|
43
|
-
|
36
|
+
kanji.map.with_index do |k, i|
|
37
|
+
ret = { 'kanji' => k }
|
38
|
+
oku = okuri[i]
|
39
|
+
rubi = ''
|
40
|
+
if oku
|
41
|
+
ret.merge!({ 'okuri' => oku })
|
42
|
+
rubi = yomi.slice(0, yomi.index(oku))
|
43
|
+
yomi = yomi.slice(yomi.index(oku) + oku.length, yomi.length)
|
44
|
+
else
|
45
|
+
rubi = yomi
|
46
|
+
end
|
47
|
+
ret.merge!({ 'rubi' => rubi })
|
48
|
+
end
|
44
49
|
end
|
45
50
|
|
46
51
|
def create_markup(data)
|
47
|
-
data.map do |e, i|
|
48
|
-
if e
|
49
|
-
|
50
|
-
|
52
|
+
data.map.with_index do |e, i|
|
53
|
+
if e.class == Array
|
54
|
+
e.map do |r|
|
55
|
+
Parser.ruby_tag(r['kanji'], r['rubi']) << "#{r['okuri'] ? r['okuri'] : ''}"
|
56
|
+
end.join('')
|
57
|
+
elsif e['surface']
|
51
58
|
"#{e['surface']}"
|
52
59
|
end
|
53
60
|
end
|
data/lib/rubyrubi/version.rb
CHANGED
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: rubyrubi
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.
|
4
|
+
version: 0.3.0
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- 110chang
|
8
8
|
autorequire:
|
9
9
|
bindir: exe
|
10
10
|
cert_chain: []
|
11
|
-
date: 2016-02-
|
11
|
+
date: 2016-02-29 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: bundler
|