Class: Crawler

Inherits:
SiteContext show all
Includes:
CrawlerHelper, Magick
Defined in:
lib/site_context.rb

Defined Under Namespace

Classes: Progress

Instance Attribute Summary collapse

Attributes inherited from SiteContext

#driver

Instance Method Summary collapse

Methods included from CrawlerHelper

#do_not_ignore?, #fix_scheme, #sanitize, #test

Methods inherited from SiteContext

#login_to_as, #set_driver

Constructor Details

#initialize(site, test, base, browser = :firefox) ⇒ Crawler

Returns a new instance of Crawler.



89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
 
# File 'lib/site_context.rb', line 89

def initialize(site, test, base, browser = :firefox)
  @site = site.to_s
  $config.SCREEN_RESOLUTION.keys.each do |key|
    FileUtils::mkdir_p "#{ENV['HOME']}/DoubleTake_data/#{@site}/#{key}"
  end
  puts "* Screenshots are saved in #{ENV['HOME']}/DoubleTake_data/#{@site}"
  $config.to_be_scraped << test
  @test_domain_length = test.length
  puts "Crawler initialized"
  @driver1 = SiteContext.new
  @driver1 = @driver1.set_driver(browser)
  @driver1.get(test)
  @driver1.manage.window.resize_to($config.SCREEN_RESOLUTION.first[1][0], $config.SCREEN_RESOLUTION.first[1][1]) #First resolution
  unless $cf == "scrape"
    @driver2 = SiteContext.new
    @driver2 = @driver2.set_driver(browser)
    @driver2.get(base)
    @driver2.manage.window.resize_to($config.SCREEN_RESOLUTION.first[1][0], $config.SCREEN_RESOLUTION.first[1][1]) #First resolution
  end
end

Instance Attribute Details

#driver1 ⇒ Object

Returns the value of attribute driver1.



70
71
72
 
# File 'lib/site_context.rb', line 70

def driver1
  @driver1
end

#driver2 ⇒ Object

Returns the value of attribute driver2.



70
71
72
 
# File 'lib/site_context.rb', line 70

def driver2
  @driver2
end

#progress ⇒ Object

Returns the value of attribute progress.



70
71
72
 
# File 'lib/site_context.rb', line 70

def progress
  @progress
end

#site ⇒ Object

Returns the value of attribute site.



70
71
72
 
# File 'lib/site_context.rb', line 70

def site
  @site
end

Instance Method Details

#bad_link?(link) ⇒ Boolean

Returns:

  • (Boolean) 


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/site_context.rb', line 116

def bad_link?(link)
  # This should populate bad_links[] and return bool
  # regarding the link being passed in.
  # $config.scraped << link
  # Bad link could be parameterised URLs(?, #) or
  # External domains such as facebook, twitter etc. or
  # link is non http Ex: mailto: ftp: file: etc.
  # puts $config.scraped
  if link.include?("$") || link.include?("#") || link.include?(".png") || link.include?(".js")
    puts "Bad Link: "+link
    $config.scraped << link if ($config.scraped.include?(link) == false)
    $config.bad_links << link if ($config.bad_links.include?(link) == false)
    return true
  elsif link.include?(".pdf") || link.include?(".jpeg") || link.include?(".css") || link.include?(".jpg")
    puts "Bad Link: "+link
    $config.scraped << link if ($config.scraped.include?(link) == false)
    $config.bad_links << link if ($config.bad_links.include?(link) == false)
    return true
  elsif link.include?("video/pop") || link.include?("?") || link.include?("/user/logout")
    puts "Bad Link: "+link
    $config.scraped << link if ($config.scraped.include?(link) == false)
    $config.bad_links << link if ($config.bad_links.include?(link) == false)
    return true
  elsif link[0..3] != "http" #TODO: This doesn't seem to work.
    puts "Bad Link: "+link
    $config.scraped << link if ($config.scraped.include?(link) == false)
    $config.bad_links << link if ($config.bad_links.include?(link) == false)
    return true
  elsif link[0..@test_domain_length-1] != $config.stage || link.include?("%")
    puts "Out of Scope: "+link
    $config.scraped << link if ($config.scraped.include?(link) == false)
    $config.bad_links << link if ($config.bad_links.include?(link) == false)
    return true
  elsif $config.scraped.include?(link)
    puts "Already scraped: "+link
    $config.bad_links << link if ($config.bad_links.include?(link) == false)
    return true
  else
    puts "Good Link: "+link
    return false
  end #End of `if`
end

#clean_up ⇒ Object 



110
111
112
113
114
 
# File 'lib/site_context.rb', line 110

def clean_up
  @driver1.quit
  @driver2.quit unless $cf == "scrape"
  puts "Destroyed WebDriver instances."
end

#crawl ⇒ Object 



159
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
231
 
# File 'lib/site_context.rb', line 159

def crawl
  # unless $config.to_be_scraped.empty?
  loop do
    puts "length of progress.to_be_scraped: #{$config.to_be_scraped.length.to_s}"
    break if $config.to_be_scraped.length < 1
    puts "length of @to_be_scrapped: #{$config.to_be_scraped.length.to_s}"
    $config.to_be_scraped.each do |each_link|
      puts "*  length of @to_be_scrapped: #{$config.to_be_scraped.length.to_s}"
      puts "** length of $config.scraped: #{$config.scraped.length.to_s}"
      puts "Working on: #{each_link}"
      begin
        @driver1.get(each_link)
        #### This Code block collects New Links and cleans
        #     $config.to_be_scraped Array.
        all_a_objs = @driver1.find_elements(:xpath, '//a')
        all_a_objs.each do |each_a_obj|
          if each_a_obj.attribute("href") != nil #Why? Cause some link obj are dicks and don't have a href
            # TODO: ^^ This if should be changed to begin - rescue
            $config.to_be_scraped << each_a_obj.attribute("href") if (each_a_obj.attribute("href").include?("http") && bad_link?(each_a_obj.attribute("href")) == false)
          end
        end #all_a_objs.each do |each_a_obj|
        $config.to_be_scraped.uniq! # Remove duplicate links.
        $config.to_be_scraped.each do |each_new_link|
          #This code block cleans the  $config.to_be_scraped Array
          $config.to_be_scraped =  $config.to_be_scraped - [each_new_link] if ($config.scraped.include?(each_new_link) || bad_link?(each_new_link))
        end   #$config.to_be_scraped.each do |each_new_link|
        if $config.scraped.include?(each_link)
          # In case a bad link makes it into the loop
          # This code-block will skip over it.
          # It will also delte it from the $config.to_be_scraped Array
          $config.to_be_scraped =  $config.to_be_scraped - [each_link]
          puts "Already Scrapped linked creeped in: #{each_link}"
          # next
        end

        #
        ### End of Code Block to collect URL's to be scraped
        stage_uri = each_link[@test_domain_length..-1]
        prod_link = $config.prod + stage_uri
        # *****************************************
        if $cf == "scrape"
          $config.SCREEN_RESOLUTION.each do |type, res|
            name = sanitize(stage_uri)
            @driver1.manage.window.resize_to(res[0], res[1])
            @driver1.save_screenshot("#{ENV['HOME']}/DoubleTake_data/#{@site}/#{type}/#{name}_stage.png")
          end
          @driver1.manage.window.resize_to($config.SCREEN_RESOLUTION.first[1][0], $config.SCREEN_RESOLUTION.first[1][1])
        else
          @driver2.get(prod_link)
          image_stuff(stage_uri)
        end
        # *****************************************
        $config.scraped << each_link # Last Step: Marking the URL as scraped!
        $config.scraped.uniq! # bad_link? may add duplicate entries
        $config.to_be_scraped = $config.to_be_scraped - [nil] # This was issue when .delete() was used which resulted in element replaced by nil
        $config.to_be_scraped = $config.to_be_scraped - [each_link]
        $config.to_be_scraped.uniq!
        File.open("#{ENV['HOME']}/DoubleTake_data/progress_#{@site}.yml", "w") {|f| f.write($config.to_yaml)}
      rescue Selenium::WebDriver::Error::StaleElementReferenceError => e
        puts "Stale element error occured moving to next link: #{stage_uri}"
        puts e
        next
      rescue Exception => e
        puts "Generic Exception occured"
        #binding.pry
        puts e.backtrace
        next
      end #End of begin
    end   #$config.to_be_scraped.each do |each_link|
  end # Loop do
  puts "to_be_scraped: " + $config.to_be_scraped.to_s
  puts "scraped      : " + $config.scraped.to_s
end

#image_stuff(stage_uri) ⇒ Object

Crawl Ending



233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
 
# File 'lib/site_context.rb', line 233

def image_stuff(stage_uri)
  name = sanitize(stage_uri)
  $config.SCREEN_RESOLUTION.each do |type, res|
    @driver1.manage.window.resize_to(res[0], res[1])
    @driver2.manage.window.resize_to(res[0], res[1])
    @driver1.save_screenshot("#{ENV['HOME']}/DoubleTake_data/#{@site}/#{type}/#{name}_stage.png")
    @driver2.save_screenshot("#{ENV['HOME']}/DoubleTake_data/#{@site}/#{type}/#{name}_prod.png")
    # a, b = IO.read("#{ENV['HOME']}/DoubleTake_data/desktop/stage_"+@site+"/"+name+".png")[0x10..0x18].unpack('NN')
    img1 = ImageList.new("#{ENV['HOME']}/DoubleTake_data/#{@site}/#{type}/#{name}_stage.png")
    img2 = ImageList.new("#{ENV['HOME']}/DoubleTake_data/#{@site}/#{type}/#{name}_prod.png")
    diff_img, diff_metric  = img1[0].compare_channel( img2[0], Magick::MeanSquaredErrorMetric )
    if diff_metric > $config.IMAGE_THRESHOLD
      diff_img.write("#{ENV['HOME']}/DoubleTake_data/#{@site}/#{type}/"+name+"_diff.png")
    else
      File.delete("#{ENV['HOME']}/DoubleTake_data/#{@site}/#{type}/#{name}_stage.png")
      File.delete("#{ENV['HOME']}/DoubleTake_data/#{@site}/#{type}/#{name}_prod.png")
    end # if diff_metric > $IMAGE_THRESHOLD
  end
end