Class: WebMiner

Inherits:
Object
  • Object
show all
Defined in:
lib/webminer.rb

Overview

Main class, just a place holder

Defined Under Namespace

Classes: Util

Class Method Summary collapse

Class Method Details

.create_story(item, topic) ⇒ Object 



15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
 
# File 'lib/webminer.rb', line 15

def self.create_story(item, topic)
  params=CGI::parse(item.link)

  link = params["url"][0]
  puts Story.class
  if Story.where(:link => link).exists?
    return nil
  end

  story = Story.new
  story.title=item.title
  story.link = link

  if item.source != nil
    story.source = item.source
  else if story.link != nil
         
         matchdata = story.link.match'https?://(?:w{3}\.)?(.*?)/.*' #Since <source> tag is rarely set in Google News, we use domain name to identify source, e.g. www.nytimes.com, but still, only http(s) is going to work      
         story.source=matchdata[1]
       else 
         return nil
       end


  end #if

  story.topic=topic
  story.date_time=item.date.rfc2822

  puts "Title: "+story.title
  puts "Link: "+story.link
  puts "Source: "+story.source
  puts "Date/time:"+story.date_time.to_s
  
  #    raw_text = self.read_story(story.link)
  if !(self.get_parser_dictionary.keys.include? story.source)
    puts "Skipping text extraction from non-trusted source "+story.source+":"+story.title
    #      story.raw_content = open(story.link) do |s| Iconv.conv('UTF-8','ISO_8859-1',s.read) end
  else
    story.content = Iconv.conv('UTF-8','ISO_8859-1',self.extract_text(story.link, self.get_parser_dictionary[story.source]))
    if story.content == nil
      puts "Text extraction failed"
    end
  end

  begin
    story.save
    puts "Saved #{story.title}"
  rescue ex
    puts "Exception saving #{story.title}"
    pp ex
  end
  
  #doc.css 'div[class="storyText"]' cbsnews.com
  #'span[id="articleText"]' reuters
  return story

end

.extract_text(link, css_path) ⇒ Object

def



88
89
90
91
92
93
94
95
96
97
98
99
100
101
 
# File 'lib/webminer.rb', line 88

def self.extract_text(link, css_path)
  begin
    f=open(link)
    doc = Nokogiri::HTML(f)
    text_nodes = doc.css(css_path)
    raw_text = text_nodes.to_s
    f.close
    return Util.strip_story(raw_text)
  rescue => ex
    puts ex.message
    puts "Error fetching link:"+link
    return nil
  end
end

.get_parser_dictionary ⇒ Object 



103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
 
# File 'lib/webminer.rb', line 103

def self.get_parser_dictionary
  return {
  "google.com" => 'div[id="hostednews-article"]',
  "cbsnews.com" => 'div[id="contentBody"]',
  "reuters.com" => 'span[id="articleText"]',
  "latimes.com" => 'div[id="story-body-text"]',
  "csmonitor.com" => 'div[id="mainColumn"]',
  "npr.org" => 'div[id="storytext"]',
  "usatoday.com" => 'div[id="mainstory"]',
  "content.usatoday.com" => 'div[id="mainstory"]',
  "guardian.co.uk" => 'div[id="article-body-blocks"]',
  "nytimes.com" => 'div[id="article"]',
  "bloomberg.com" => 'div[id="story_content"]',
  "online.wsj.com" => 'div[id="article_story_body"]',
  "asia.wsj.com" => 'div[id="article_story_body"]',
  "businessweek.com" => 'div[id="story-body"]',
  "cnn.com" => 'div[id="cnnContentContainer"]',
  "edition.cnn.com" => 'cnn_storyarea[id="cnnContentContainer"]',
  "money.cnn.com" => 'div[id="storytext"]',
  "abcnews.go.com" => 'div[id="innerbody"]',
  "foxnews.com" => 'div[id="introduction"]',
  "businessweek.com" => 'div[id="story-body"]',
  "entertainment.msnbc.msn.com" => 'div[id="vine-t"] article',
  "washingtonpost.com" => 'div[id="article_body"]',
  #    "bbc.co.uk" => 'div[id="main-content"]',
  "huffingtonpost.com" => 'div[id="entry_12345"]',
  "telegraph.co.uk" => 'div[id="mainBodyArea"]',
  "chicagotribune.com" => 'div[id="story-body-text"]',
  "foxbusiness.com" => 'div[id="introduction"]',
  "thedailybeast.com" => 'div[id="main"] article',
  "economictimes.indiatimes.com" => 'div[id="storydiv"]',
  "forbes.com" => 'div[id="leftRail"]',
  "arstechnica.com" => 'div[id="story"]',
  "theregister.co.uk"=> 'div[id="body"]',
  "ingame.msnbc.msn.com"=> 'div[id="vine-t"] article',
  "informationweek.com"=> 'span[id="articleBody"]',
  "newyorker.com"=> 'div[id="articletext"]',
  "kotaku.com"=> 'div[id="page"]',
  "slashgear.com"=> 'span[id="intelliTxt"]',
  "pcworld.com"=> 'div[id="articleText"]',
  "news.cnet.com"=> 'div[id="article"]',
  "english.aljazeera.net"=> 'td[id="tdTextContent"]',
  "dailymail.co.uk"=> 'div[id="js-article-text"]',
  "rttnews.com"=> 'div[id=""]',
  "ft.com"=> 'div[id="storyContent"]',
  "politico.com"=> 'div[id="mainContent"]',
  "boston.com"=> 'div[id="page1"]',
  "sfgate.com"=> 'div[id="fontprefs_bottom"]',
  "oregonlive.com"=> 'div[id="article"]'
  #""=> 'div[id=""]',

  #    "wired.com"=> 'div[id=""]'?
  #http://latimesblogs.latimes.com ?
}
end

.main ⇒ Object 



160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
 
# File 'lib/webminer.rb', line 160

def self.main
  prng = Random.new
  topics = Hash.new(0)
  topics['Top Stories'] = 'http://news.google.com/news?pz=1&cf=all&ned=us&hl=en&output=rss'
  topics['U.S.'] = 'http://news.google.com/news?pz=1&cf=all&ned=us&hl=en&topic=n&output=rss'
  topics['Health'] = 'http://news.google.com/news?pz=1&cf=all&ned=us&hl=en&topic=m&output=rss'
  topics['Business'] = 'http://news.google.com/news?pz=1&cf=all&ned=us&hl=en&topic=b&output=rss'
  topics['Technology'] = 'http://news.google.com/news?pz=1&cf=all&ned=us&hl=en&topic=tc&output=rss'
  topics['World'] = 'http://news.google.com/news?pz=1&cf=all&ned=us&hl=en&topic=w&output=rss'

  topic_threads = []
  $options = {}
  $options['User-Agent']='Mozilla/5.0 (X11; Linux x86_64; rv:7.0.1) Gecko/20100101 Firefox/7.0.1'
  $options['Accept']='text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8'
  $options['From']='[email protected]'
  topics.keys.each do |topic|
    url = topics[topic]
    sleep 20+prng.rand*2
    topic_threads << Thread.new(url) do
      #uri = URI.parse(url)

      #http = Net::HTTP.new(uri.host, uri.port)
      #request = Net::HTTP::Get.new(uri.request_uri)
      print "I'm working on #{url}\n"    

      hash = 0

      while true
        #response = http.request(request)
        #code = response.code
        #body = response.body
        content = ""
        begin
          open(url,$options) do |s| content = s.read end
        rescue OpenURI::HTTPError => ex
          puts "Google caught us again! "+url
#          rescue => ex
#            puts "Unhandled exception:"+ex.message
        rescue SocketError => ex
          puts ex.message
        rescue Timeout::Error => ex
          puts ex.message
        end
        new_hash = content.hash

        # Headers are lowercased
        #response["cache-control"] # => public, max-age=2592000
        


        #print "Got code: #{code}\n"             # => 301
        print "Got content: #{content.length}, hash: #{new_hash}\n"             # => The body (HTML, XML, blob, whatever)

        if new_hash != hash
          print "Parsing\n"
          self.parse(content, topic)
        else
          print "Skipping\n"
        end
        hash = new_hash
        sleep 180+prng.rand*10
      end #while
    end #thread
  end #each


  topic_threads.each do |thr|
    thr.join
  end #each

end

.parse(content, topic) ⇒ Object

def



74
75
76
77
78
79
80
81
82
83
84
85
86
 
# File 'lib/webminer.rb', line 74

def self.parse(content, topic)
  rss = RSS::Parser::parse(content, false)
  if rss != nil
    items = rss.items
    if items != nil
      for item in items
        story = self.create_story(item, topic)
        #print "Created story: #{story}\n"
        #story.save
      end #for
    end
  end
end