Class: Yahoo

Inherits:
Object
  • Object
show all
Includes:
Searchy
Defined in:
lib/esearchy/SearchEngines/yahoo.rb

Instance Attribute Summary collapse

Instance Method Summary collapse

Methods included from Searchy

#clean, #fix, #hash_url, #maxhits=, #print_emails, #search_depth, #search_docs, #search_emails, #search_office_xml, #search_pdfs, #search_txts

Constructor Details

#initialize(maxhits = 0, appid = nil, start = 0) ⇒ Yahoo

Returns a new instance of Yahoo.



8
9
10
11
12
13
14
15
16
17
18
19
20
 
# File 'lib/esearchy/SearchEngines/yahoo.rb', line 8

def initialize(maxhits=0, appid = nil, start=0)
  @appid = appid || Keys::YAHOO_APP_KEY
  @start = start
  @totalhits = maxhits
  @emails = []
  @r_urls = Queue.new
  @r_docs = Queue.new
  @r_pdfs = Queue.new
  @r_officexs = Queue.new
  @r_txts = Queue.new
  @threads = []
  @lock = Mutex.new
end

Instance Attribute Details

#appid ⇒ Object

Returns the value of attribute appid.



21
22
23
 
# File 'lib/esearchy/SearchEngines/yahoo.rb', line 21

def appid
  @appid
end

#emails ⇒ Object

Returns the value of attribute emails.



21
22
23
 
# File 'lib/esearchy/SearchEngines/yahoo.rb', line 21

def emails
  @emails
end

Instance Method Details

#parse(json) ⇒ Object 



56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
 
# File 'lib/esearchy/SearchEngines/yahoo.rb', line 56

def parse(json)
  doc = JSON.parse(json) 
  @totalhits = doc["ysearchresponse"]["totalhits"].to_i if @totalhits == 0
  doc["ysearchresponse"]["resultset_web"].each do |result|
    case result["url"]
    when /.pdf$/i
      @r_pdfs << result["url"]
    when /.docx$|.xlsx$|.pptx$|.odt$|.odp$|.ods$|.odb$/i
      @r_officexs << result["url"]
    when /.doc$/i
      @r_docs << result["url"]
    when /.txt$|.rtf$|ans$/i
      @r_txts << result["url"]
    else
      @r_urls << result["url"]
    end
  end
end

#search(query) ⇒ Object 



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
 
# File 'lib/esearchy/SearchEngines/yahoo.rb', line 23

def search(query)
  @query = query
  begin
    http = Net::HTTP.new("boss.yahooapis.com",80)
    http.start do |http|
      request = Net::HTTP::Get.new("/ysearch/web/v1/" + CGI.escape(query) + 
                                   "?appid="+ @appid + 
                                   "&format=json&count=50"+ 
                                   "&start=#{@start}", {'User-Agent' => UserAgent::fetch} )
      response = http.request(request)
      case response
      when Net::HTTPSuccess, Net::HTTPRedirection
        parse(response.body)
        @start = @start + 50
        if @totalhits > @start
          ESearchy::LOG.puts "Searching #{self.class} from #{@start-50} to #{@start}"
          search_emails(response.body)
          sleep(ESearchy::DELAY)
          search(@query)
        else
          ESearchy::LOG.puts "Searching #{self.class} from #{@start-50} to #{@start}"
          search_emails(response.body)
        end
      else
        return response.error!
      end
    end
  rescue Net::HTTPFatalError
    ESearchy::LOG.puts "Error: Something went wrong with the HTTP request"
  end
  @start = 0
end