Class: BlackStack::Bots::Scraper

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

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(init_domain, timeout, h) ⇒ Scraper

Returns a new instance of Scraper.



8
9
10
11
12
13
14
15
16
17
18
 
# File 'lib/scraper.rb', line 8

def initialize(init_domain, timeout, h)
    self.domain = init_domain
    self.timeout = timeout || 10
    self.load_wait_time = 3
    self.stop_scraping_at_page_number = 25
    self.stop_scraping_at_match_number = 1
    #self.agent.agent.http.verify_mode = OpenSSL::SSL::VERIFY_NONE
    self.links = []
    self.links_processed = []
    self.browser = BlackStack::Bots::Browser.new()
end

Instance Attribute Details

#browser ⇒ Object

Returns the value of attribute browser.



4
5
6
 
# File 'lib/scraper.rb', line 4

def browser
  @browser
end

#domain ⇒ Object

Returns the value of attribute domain.



4
5
6
 
# File 'lib/scraper.rb', line 4

def domain
  @domain
end

#links ⇒ Object

Returns the value of attribute links.



4
5
6
 
# File 'lib/scraper.rb', line 4

def links
  @links
end

#links_processed ⇒ Object

auxiliar array of links that I have extracted links from



6
7
8
 
# File 'lib/scraper.rb', line 6

def links_processed
  @links_processed
end

#load_wait_time ⇒ Object

Returns the value of attribute load_wait_time.



4
5
6
 
# File 'lib/scraper.rb', line 4

def load_wait_time
  @load_wait_time
end

#stop_scraping_at_match_number ⇒ Object

Returns the value of attribute stop_scraping_at_match_number.



4
5
6
 
# File 'lib/scraper.rb', line 4

def stop_scraping_at_match_number
  @stop_scraping_at_match_number
end

#stop_scraping_at_page_number ⇒ Object

Returns the value of attribute stop_scraping_at_page_number.



4
5
6
 
# File 'lib/scraper.rb', line 4

def stop_scraping_at_page_number
  @stop_scraping_at_page_number
end

#timeout ⇒ Object

Returns the value of attribute timeout.



4
5
6
 
# File 'lib/scraper.rb', line 4

def timeout
  @timeout
end

Instance Method Details

#find_keywords(a, stop_at = 25, stop_on_first_link_found = false, l = nil) ⇒ Object

def get_links



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
 
# File 'lib/scraper.rb', line 180

def find_keywords(a, stop_at=25, stop_on_first_link_found=false, l=nil)
    pages = []
    l = BlackStack::DummyLogger.new(nil) if l.nil?
    # iterate the links
    j = 0
    self.links.reject { |link| link =~ /\.pdf$/i || link =~ /\.jpg$/i || link =~ /\.jpeg$/i || link =~ /\.gif$/i }.each { |link|
        j += 1
        break if j > stop_at
        l.logs "#{j.to_s}. find_keywords (#{link})... "
        begin
            # get the page
            browser.goto link
            sleep(self.load_wait_time) # wait 10 seconds for javascript content to load
            # get page body content in plain text
            title = browser.title
            s = browser.body.text
            # add the link to the results of no-keyword
            hpage = { 'page_url' => link.downcase, 'page_title' => title, 'page_html' => browser.body.html, 'keywords' => [] }
            pages << hpage
            # iterate the keywords
            i = 0
            match = false
            a.each { |k|
                # find the keyword
                match = ( s =~ /#{Regexp.escape(k)}/i )
                hpage['keywords'] << k if match
                # count the number of links with match
                # break if only 1 link is needed
                if match 
                    i += 1
                    break if stop_on_first_link_found
                end # if
            } # each
            break if match && stop_on_first_link_found
            l.logf i == 0 ? 'no keywords found'.yellow : "#{i} keywords found".green # find_keywords

        rescue Net::ReadTimeout => e
            l.logf "Timeout Error: #{e.message}".red

            l.logs "Restarting browser..."
            browser.close if browser    
            self.browser = BlackStack::Bots::Browser.new()
            l.done             
        rescue => e
            l.logf "Error: #{e.message.split("\n").first[0..100]}".red # get_links
        end
    } # each
    # return
    pages
end

#get_links(stop_at = 100, l = nil) ⇒ Object

def get_links_from_url



158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
 
# File 'lib/scraper.rb', line 158

def get_links(stop_at=100, l=nil)
    l = BlackStack::DummyLogger.new(nil) if l.nil?

    # get links from the sitemap
    self.get_links_from_sitemap(stop_at, l)
=begin
    # working with root url
    url = "http://#{self.domain}/"
    self.links << url if self.links.select { |link| link == url }.empty?
    # iterate until I have discovered all the links
    while self.links.size != self.links_processed.size && stop_at >= self.links.size
        # iterate the links who are not in links_processed
        self.links.select { |link| !self.links_processed.include?(link) }.each { |link|
            # get the links from the url
            self.get_links_from_url(link, l)
            # add the link to the list of processed links
            self.links_processed << link
        }
    end # while
=end

end

#get_links_from_sitemap(stop_at = 100, l = nil) ⇒ Object

def initialize



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
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
 
# File 'lib/scraper.rb', line 20

def get_links_from_sitemap(stop_at=100, l=nil)
    max_allowed_timeout_errors = 3
    timeout_errors = 0
    max_links = self.links.size + stop_at

    l.logs "Scrape sitemaps... "
    begin
        l.logs "get_sitemaps from #{self.domain}... "
        
        # download the robots.txt
        url = "http://#{domain}/robots.txt"
        
        # get the content of robots.txt from url
        browser.goto url
        s = browser.text
        
        # get the sitemap
        sitemaps = s.split("\n").select { |line| line =~ /^sitemap:/i }.map { |a| a.downcase.split('sitemap:').last.strip }.uniq
        processed = []
        to_process = sitemaps - processed
        l.logf sitemaps.size == 0 ? 'no sitemap found'.yellow : "#{sitemaps.size} sitemaps found".green # get_links

        # while there are sitemaps to process
        while to_process.size > 0 && timeout_errors < max_allowed_timeout_errors && max_links >= self.links.size
            to_process.each { |b|
                l.logs "go to #{b}... "
                begin
                    browser.goto b
                    l.done

                    l.logs "parsing #{b}... "
                    s = browser.text
                    # extract all URLs
                    doc = Nokogiri::HTML(s)
                    l.done

                    # get the value of all <loc> tags with .xml extension
                    l.logs "get_sitemaps from #{b}... "
                    sitemaps += doc.xpath('//loc').map(&:text).select { |s| s =~ /\.xml$/ }.map { |s| s.downcase }
                    sitemaps.uniq!
                    l.logf sitemaps.size == 0 ? 'no sitemap found'.yellow : "#{sitemaps.size} sitemaps found".green # get_links
                    
                    # get the value of all <loc> tags without .xml extension
                    l.logs "get_links from #{b}..."
                    self.links += doc.xpath('//loc').map(&:text).select { |s| s !~ /\.xml$/ }.map { |s| s.downcase }
                    self.links.uniq!
                    l.logf self.links.size == 0 ? 'no links found'.yellow : "#{self.links.size} links found".green # get_links
                    
                    # add the sitemap to the list of processed sitemaps
                    processed << b

                    # reset timeout errors
                    timeout_errors = 0

                    # break if I exceeded the limit of links
                    break if max_links <= self.links.size

                rescue Net::ReadTimeout => e
                    l.logf "Timeout Error: #{e.message}".red            
                
                    l.logs "Restarting browser..."
                    browser.close if browser    
                    self.browser = BlackStack::Bots::Browser.new()
                    l.done 

                    timeout_errors += 1
                    break if timeout_errors >= max_allowed_timeout_errors

                rescue => e
                    l.logf "Error: #{e.message.split("\n").first[0..100]}".red # get_links
                end            
            }
            # update the list of sitemaps to process
            processed.uniq!
            to_process = sitemaps - processed
        end
        l.logf sitemaps.size == 0 ? 'no sitemap found'.yellow : "#{sitemaps.size} sitemaps found".green # get_links

    rescue Net::ReadTimeout => e
        l.logf "Timeout Error: #{e.message}".red            

        l.logs "Restarting browser..."
        browser.close if browser    
        self.browser = BlackStack::Bots::Browser.new()
        l.done    
    rescue => e
        l.logf "Error: #{e.message.split("\n").first[0..100]}".red # get_links
    end
end

#get_links_from_url(url, l = nil) ⇒ Object

internal use only



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
 
# File 'lib/scraper.rb', line 111

def get_links_from_url(url, l=nil)
    l = BlackStack::DummyLogger.new(nil) if l.nil?
    l.logs "get_links (#{url})... "
    aux = []
    begin       
        # trim url
        url = url.strip
        # get domain of the url using open-uri
        domain = URI.parse(url).host
        # visit the main page of the website
        browser.goto url
        sleep(self.load_wait_time) # wait 10 seconds for javascript content to load
        # get the self.links to the pages of the website
        aux = browser.links.map(&:href)
        # remove non-string elements
        aux = aux.select { |link| link.is_a?(String) }
        # remove # from the self.links
        aux = aux.map { |link| !link.nil? && link.split('#').first }
        # remove querystring from the self.links
        aux = aux.map { |link| !link.nil? && link.split('?').first }
        # remove the self.links that are not http:// or https://
        aux = aux.select { |link| !link.nil? && link =~ /^https?:\/\// }
        # remove the self.links that are not from the same domain
        aux = aux.select { |link| !link.nil? && link =~ /#{domain}/ }
        # remove nil values
        aux = aux.compact
        # remove duplications
        aux = aux.uniq
        # filter links who already are in the list
        a = aux.size
        aux = aux.select { |link| !self.links.include?(link) }
        b = aux.size
        # add new links to self.links
        l.logf "done".green + " (#{a} links found, #{b} new, #{self.links.size} total)" # get_links
    rescue Net::ReadTimeout => e
        l.logf "Timeout Error: #{e.message}".red            

        l.logs "Restarting browser..."
        browser.close if browser    
        self.browser = BlackStack::Bots::Browser.new()
        l.done  
    rescue => e
        l.logf "Error: #{e.message.split("\n").first[0..100]}".red # get_links
    end
    self.links += aux
end