ruby 雅虎JAPANが提供するテキスト解析WebAPIを利用して，テキストに振り仮名をる

Posted 2021-05-16

<?xml version="1.0" encoding="UTF-8"?>
<ResultSet xmlns="urn:yahoo:jp:jlp:FuriganaService" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="urn:yahoo:jp:jlp:FuriganaService http://jlp.yahooapis.jp/FuriganaService/V1/furigana.xsd">
  <Result>
    <WordList>
      <Word>
        <Surface>ルビ</Surface>
        <Furigana>るび</Furigana>
        <Roman>rubi</Roman>
      </Word>
      <Word>
        <Surface>振り</Surface>
        <Furigana>ぶり</Furigana>
        <Roman>buri</Roman>
        <SubWordList>
          <SubWord>
            <Surface>振</Surface>
            <Furigana>ぶ</Furigana>
            <Roman>bu</Roman>
          </SubWord>
          <SubWord>
            <Surface>り</Surface>
            <Furigana>り</Furigana>
            <Roman>ri</Roman>
          </SubWord>
        </SubWordList>
      </Word>
      <Word>
        <Surface>:</Surface>
      </Word>
      <Word>
        <Surface>漢字</Surface>
        <Furigana>かんじ</Furigana>
        <Roman>kanzi</Roman>
      </Word>
      <Word>
        <Surface>かな交じり</Surface>
        <Furigana>かなまじり</Furigana>
        <Roman>kanamaziri</Roman>
        <SubWordList>
          <SubWord>
            <Surface>かな</Surface>
            <Furigana>かな</Furigana>
            <Roman>kana</Roman>
          </SubWord>
          <SubWord>
            <Surface>交</Surface>
            <Furigana>ま</Furigana>
            <Roman>ma</Roman>
          </SubWord>
          <SubWord>
            <Surface>じり</Surface>
            <Furigana>じり</Furigana>
            <Roman>ziri</Roman>
          </SubWord>
        </SubWordList>
      </Word>
      <Word>
        <Surface>文</Surface>
        <Furigana>ぶん</Furigana>
        <Roman>bun</Roman>
      </Word>
      <Word>
        <Surface>に</Surface>
        <Furigana>に</Furigana>
        <Roman>ni</Roman>
      </Word>
      <Word>
        <Surface>、</Surface>
      </Word>
      <Word>
        <Surface>ひらがな</Surface>
        <Furigana>ひらがな</Furigana>
        <Roman>hiragana</Roman>
      </Word>
      <Word>
        <Surface>と</Surface>
        <Furigana>と</Furigana>
        <Roman>to</Roman>
      </Word>
      <Word>
        <Surface>ローマ字</Surface>
        <Furigana>ろーまじ</Furigana>
        <Roman>ro-mazi</Roman>
        <SubWordList>
          <SubWord>
            <Surface>ローマ</Surface>
            <Furigana>ろーま</Furigana>
            <Roman>ro-ma</Roman>
          </SubWord>
          <SubWord>
            <Surface>字</Surface>
            <Furigana>じ</Furigana>
            <Roman>zi</Roman>
          </SubWord>
        </SubWordList>
      </Word>
      <Word>
        <Surface>の</Surface>
        <Furigana>の</Furigana>
        <Roman>no</Roman>
      </Word>
      <Word>
        <Surface>ふりがな</Surface>
        <Furigana>ふりがな</Furigana>
        <Roman>hurigana</Roman>
      </Word>
      <Word>
        <Surface>（</Surface>
      </Word>
      <Word>
        <Surface>ルビ</Surface>
        <Furigana>るび</Furigana>
        <Roman>rubi</Roman>
      </Word>
      <Word>
        <Surface>）</Surface>
      </Word>
      <Word>
        <Surface>を</Surface>
        <Furigana>を</Furigana>
        <Roman>wo</Roman>
      </Word>
      <Word>
        <Surface>付け</Surface>
        <Furigana>つけ</Furigana>
        <Roman>tuke</Roman>
        <SubWordList>
          <SubWord>
            <Surface>付</Surface>
            <Furigana>つ</Furigana>
            <Roman>tu</Roman>
          </SubWord>
          <SubWord>
            <Surface>け</Surface>
            <Furigana>け</Furigana>
            <Roman>ke</Roman>
          </SubWord>
        </SubWordList>
      </Word>
      <Word>
        <Surface>ます</Surface>
        <Furigana>ます</Furigana>
        <Roman>masu</Roman>
      </Word>
      <Word>
        <Surface>。</Surface>
      </Word>
    </WordList>
  </Result>
</ResultSet>

require 'pp'
require 'mechanize'

# Yahoo! 振り仮名API
# Ref: http://developer.yahoo.co.jp/webapi/jlp/furigana/v1/furigana.html

class Furigana
  attr_accessor :sentence
  def initialize appid=''
    @appid = appid
    @a = Mechanize.new
  end
  def send_request
    # Request XML Data with GET method
    @xml = @a.get(request_url).xml

    # Construct [[Surface, Furigana, Roman], ...] array
    word.inject([]){|res, elements|
      res << %w(Surface Furigana Roman).inject([]){|r, target_node|
        r << elements.search(target_node).text
      }
    }
  end
  def word
    # Extract <Word/> nodes
    xml = @xml.search('Word')

    # remove <SubWord/> nodes
    xml.search('SubWord').map{|element|
      element.children.remove
    }
    xml
  end
  def surface
    @xml.search('Surface')
      .map{|element|
        element.text
      }
  end
  def furigana
    @xml.search('Furigana')
      .map{|element|
        element.text
      }
  end
  def sentence_is str
    @sentence = str
    self
  end

  private
  def request_url
    "#{base_url}#{service_name}?#{appid}&#{sentence}"
  end
  def service_name
    service = 'furigana'
  end
  def appid
    "appid=#{@appid}"
  end
  def base_url
    base_url = 'http://jlp.yahooapis.jp/FuriganaService/V1/'
  end
  def sentence
    "sentence=#{@sentence}"
  end

  alias_method :and_request_send, :send_request
end


# Usage:

appid = 'LongLongYOUREappIDfooabcdBRAbrabrafoobarBrabRabrabrabra-'

f = Furigana.new(appid)
#f.sentence = 'ルビ振り:漢字かな交じり文に、ひらがなとローマ字のふりがな（ルビ）を付けます。'
#pp res = f.send_request
pp f.sentence_is('ルビ振り:漢字かな交じり文に、ひらがなとローマ字のふりがな（ルビ）を付けます。').and_request_send

=begin
[["ルビ", "るび", "rubi"],
 ["振り", "ぶり", "buri"],
 [":", "", ""],
 ["漢字", "かんじ", "kanzi"],
 ["かな交じり", "かなまじり", "kanamaziri"],
 ["文", "ぶん", "bun"],
 ["に", "に", "ni"],
 ["、", "", ""],
 ["ひらがな", "ひらがな", "hiragana"],
 ["と", "と", "to"],
 ["ローマ字", "ろーまじ", "ro-mazi"],
 ["の", "の", "no"],
 ["ふりがな", "ふりがな", "hurigana"],
 ["（", "", ""],
 ["ルビ", "るび", "rubi"],
 ["）", "", ""],
 ["を", "を", "wo"],
 ["付け", "つけ", "tuke"],
 ["ます", "ます", "masu"],
 ["。", "", ""]]
=end