furigana 0.0.2 → 0.0.3
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/lib/furigana/reader.rb +8 -21
- data/lib/furigana/version.rb +1 -1
- metadata +1 -1
data/lib/furigana/reader.rb
CHANGED
@@ -3,55 +3,42 @@ module Furigana
|
|
3
3
|
class Reader
|
4
4
|
def reading(text)
|
5
5
|
Mecab.tokenize(text).inject([]) do |list, token|
|
6
|
-
with_furigana = add_furigana(
|
7
|
-
list
|
6
|
+
with_furigana = add_furigana(token)
|
7
|
+
list += with_furigana if with_furigana
|
8
8
|
list
|
9
9
|
end
|
10
10
|
end
|
11
11
|
|
12
12
|
private
|
13
13
|
|
14
|
-
def yomi_to_hiragana(token)
|
15
|
-
token[:reading] = choose_reading(token[:surface_form], token[:reading])
|
16
|
-
token
|
17
|
-
end
|
18
|
-
|
19
14
|
def k2h(k)
|
20
15
|
return nil if k.nil?
|
21
16
|
NKF.nkf("-h1 -w", k)
|
22
17
|
end
|
23
18
|
|
24
|
-
def kana?(str)
|
25
|
-
/^[ぁ-んァ-ンー]+$/.match(str)
|
26
|
-
end
|
27
|
-
|
28
|
-
def choose_reading(surface_form, reading)
|
29
|
-
!kana?(surface_form) ? k2h(reading) : surface_form
|
30
|
-
end
|
31
|
-
|
32
19
|
def sdiff(first, second)
|
33
20
|
Diff::LCS.sdiff(first, second)
|
34
21
|
end
|
35
22
|
|
36
23
|
def diff_token_surface_form_and_reading(token)
|
37
|
-
sdiff(token[:surface_form], token[:reading])
|
24
|
+
sdiff(k2h(token[:surface_form]), k2h(token[:reading]))
|
38
25
|
end
|
39
26
|
|
40
27
|
def add_furigana(token)
|
41
28
|
states = { kanji_and_yomi: '!', yomi: '+', kana: '=' }
|
42
29
|
kanji, yomi = 0, 1
|
43
30
|
|
44
|
-
list =
|
31
|
+
list = []
|
45
32
|
on_kanji = false
|
46
33
|
diff_token_surface_form_and_reading(token).each do |part|
|
47
34
|
case part.action
|
48
35
|
when states[:kanji_and_yomi]
|
49
|
-
list
|
50
|
-
list[kanji] += part.old_element
|
51
|
-
list[yomi] += part.new_element
|
36
|
+
list.push ['',''] unless on_kanji
|
37
|
+
list.last[kanji] += part.old_element
|
38
|
+
list.last[yomi] += part.new_element
|
52
39
|
on_kanji = true
|
53
40
|
when states[:yomi]
|
54
|
-
list[yomi] += part.new_element
|
41
|
+
list.last[yomi] += part.new_element
|
55
42
|
when states[:kana]
|
56
43
|
on_kanji = false
|
57
44
|
end
|
data/lib/furigana/version.rb
CHANGED