Class: TaiwaneseNewsParser::Parser::LibertyTimes

Inherits:
TaiwaneseNewsParser::Parser show all
Defined in:
lib/taiwanese_news_parser/parser/liberty_times.rb

Instance Attribute Summary

Attributes inherited from TaiwaneseNewsParser::Parser

#article, #url

Class Method Summary collapse

Instance Method Summary collapse

Methods inherited from TaiwaneseNewsParser::Parser

applicable_parser, #clean_up, #initialize, #reproduced?, subclasses

Constructor Details

This class inherits a constructor from TaiwaneseNewsParser::Parser

Class Method Details

.applicable?(url) ⇒ Boolean

Returns:

  • (Boolean) 


10
11
12
 
# File 'lib/taiwanese_news_parser/parser/liberty_times.rb', line 10

def self.applicable?(url)
  url.include?('iservice.ltn.com.tw')
end

.domain ⇒ Object 



2
3
4
 
# File 'lib/taiwanese_news_parser/parser/liberty_times.rb', line 2

def self.domain
  'libertytimes.com.tw'
end

.names ⇒ Object 



6
7
8
 
# File 'lib/taiwanese_news_parser/parser/liberty_times.rb', line 6

def self.names
  %{自由時報}
end

.parse_url_id(url) ⇒ Object 



62
63
64
 
# File 'lib/taiwanese_news_parser/parser/liberty_times.rb', line 62

def self.parse_url_id(url)
  url[%r{news\.php\?no=(\d+)},1]
end

Instance Method Details

#clean_url ⇒ Object 



57
58
59
60
 
# File 'lib/taiwanese_news_parser/parser/liberty_times.rb', line 57

def clean_url
  cleaner = TaiwaneseNewsParser::UrlCleaner.new('no')
  @article[:url] = cleaner.clean(@article[:url])
end

#doc ⇒ Object 



14
15
16
17
 
# File 'lib/taiwanese_news_parser/parser/liberty_times.rb', line 14

def doc
  @raw = open(url).read
  @doc = Nokogiri::HTML(@raw)
end

#parse ⇒ Object 



19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
 
# File 'lib/taiwanese_news_parser/parser/liberty_times.rb', line 19

def parse
  # new layout uses utf-8
  @article[:title] = doc.at_css('.content h1 text()').text
  @article[:company_name] = parse_company_name
  @article[:content] = doc.css('.content p').text

  time = doc.at_css('.content .date').text[%r{\d{4}-\d{1,2}-\d{1,2} \d{2}:\d{2}}]
  if time.nil?
    match = doc.at_css('.content .date').text.match(%r{(\d{2}):(\d{2})})
    now = Time.now
    today = Date.today
    @article[:published_at] = Time.new(today.year, today.month, today.day, match[1].to_i, match[2].to_i)
  else
    @article[:published_at] = Time.parse("#{time}:00")
  end

  @article[:reporter_name] = parse_reporter_name()

  clean_up

  @article
end

#parse_company_name ⇒ Object 



53
54
55
 
# File 'lib/taiwanese_news_parser/parser/liberty_times.rb', line 53

def parse_company_name
  '自由時報'
end

#parse_reporter_name ⇒ Object 



42
43
44
45
46
47
48
49
50
51
 
# File 'lib/taiwanese_news_parser/parser/liberty_times.rb', line 42

def parse_reporter_name
  if match = @article[:content].match(%r{〔(.*?)[/／╱](.*?)〕})
    reporter_name = match[1][%r{記者(.+)},1]
  elsif match = @article[:content].match(%r{記者(.+?)[/／╱]})
    reporter_name = match[1]
  elsif match = @article[:content].match(%r{（文／(.*?)）})
    reporter_name = match[1]
  end
  reporter_name
end