Class: Baiduserp::Client

Inherits:
Object
  • Object
show all
Includes:
HTTParty
Defined in:
lib/baiduserp/client.rb

Constant Summary collapse

AllUserAgents = 
YAML.load(open(File.expand_path('../user_agents.yml',__FILE__)))

Class Method Summary collapse

Instance Method Summary collapse

Class Method Details

.get_rank_url(url) ⇒ Object 



20
21
22
 
# File 'lib/baiduserp/client.rb', line 20

def self.get_rank_url(url)
  self.new.get_rank_url(url)
end

.get_serp(url, retries = 3) ⇒ Object 



16
17
18
 
# File 'lib/baiduserp/client.rb', line 16

def self.get_serp(url,retries = 3)
  self.new.get_serp(url,retries)
end

.rand_ua ⇒ Object 



7
8
9
 
# File 'lib/baiduserp/client.rb', line 7

def self.rand_ua
  AllUserAgents[rand(AllUserAgents.size)]
end

Instance Method Details

#get_rank_url(url) ⇒ Object 



24
25
26
27
28
29
30
31
32
33
34
 
# File 'lib/baiduserp/client.rb', line 24

def get_rank_url(url)
  begin
    response = self.class.get(url)
  rescue StandardError => e
    puts e.class
    puts e.message
    sleep(10)
    retry
  end
  response
end

#get_serp(url, retries = 3) ⇒ Object 



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
73
74
75
76
77
78
79
80
81
 
# File 'lib/baiduserp/client.rb', line 36

def get_serp(url, retries = 3)
  if retries > 0
    begin
      response = self.class.get(url)
    rescue StandardError => e
      puts e.class
      puts e.message
      sleep(10)
      retry
    end

    if response.code != 200
      puts response
      puts "Retry on URL: #{url}"
      sleep(rand(60)+1200)
      response = self.class.get_serp(url,retries - 1)
    end

    if response.nil?
      puts "Still error after 3 tries, sleep 3600s now."
      sleep(3600)
      response = self.class.get_serp(url)
    end

    ##Baidu Stopped response Content-Length in headers...
    #if response.headers['Content-Length'].nil?
    #  puts "Can't read Content-Length from response, retry."
    #  response = self.class.get_serp(url,retries-1)
    #end
    #
    #if response.headers['Content-Length'].to_i != response.body.bytesize
    #  issue_file = "/tmp/baiduserp_crawler_issue_#{Time.now.strftime("%Y%m%d%H%M%S")}.html"
    #  open(issue_file,'w').puts(response.body)
    #  puts "Notice:"
    #  puts "Baiduserp get an error when crawl SERP: response size (#{response.headers['Content-Length']}) not match body size."
    #  puts "Please see file #{issue_file} for body content."
    #  puts "Sleep 10s and retry"
    #  sleep(10)
    #  response = self.class.get_serp(url)
    #end

    response
  else
    nil
  end
end