class TaiwaneseNewsParser::Parser::Cna

Public Class Methods

domain() click to toggle source
# File lib/taiwanese_news_parser/parser/cna.rb, line 2
def self.domain
  'cna.com.tw'
end
names() click to toggle source
# File lib/taiwanese_news_parser/parser/cna.rb, line 6
def self.names
  %{中央社}
end
parse_url_id(url) click to toggle source
# File lib/taiwanese_news_parser/parser/cna.rb, line 56
def self.parse_url_id(url)
  url[%r{/(\d+)(?:\-\d)?\.},1]
end

Public Instance Methods

doc() click to toggle source
# File lib/taiwanese_news_parser/parser/cna.rb, line 10
def doc
  @raw = open(url).read
  @doc = Nokogiri::HTML(@raw)
end
parse() click to toggle source

url = 'www.cna.com.tw/News/aSaM/201304120296-1.aspx'

# File lib/taiwanese_news_parser/parser/cna.rb, line 16
def parse
  @article[:title] = doc.at_css('.news_content h1, .news_content h2').text

  @article[:company_name] = '中央社'

  @article[:content] = doc.css('.news_content .box_2').text

  @article[:reporter_name] = parse_reporter_name()

  match = doc.css('.news_content .box_2').text.strip.match( /(\d{3})(\d{2})(\d{2})/ )
  date = []
  date[0] = match[1].to_i + 1911
  date[1] = match[2]
  date[2] = match[3]
  date_string = date.join('/') + ' ' + doc.css('.date').text
  @article[:published_at] = Time.parse(date_string)

  clean_up

  @article
end
parse_reporter_name() click to toggle source
# File lib/taiwanese_news_parser/parser/cna.rb, line 38
def parse_reporter_name
  text = doc.css('.news_content .box_2').text
  text = text[/(中央社(.*?)\d{1,2}日/,1]
  cities = %w{台北 新北 台中 台南 高雄 基隆 新竹 嘉義 桃園 新竹 苗栗 彰化 南投 雲林 嘉義 屏東 宜蘭 花蓮 台東 澎湖 金門 連江}
  cities.find do |city|
    text.gsub!(/#{city}(?:縣市)?$/,'')
  end
  # TODO proper location name removal
  if match = text.match(%r{記者(.+)})
    reporter_name = match[1]
  end
  reporter_name
end
reproduced?() click to toggle source
# File lib/taiwanese_news_parser/parser/cna.rb, line 52
def reproduced?
  false
end