Class: TheFox::Sengi::Crawler

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

Instance Method Summary collapse

Constructor Details

#initialize(url, options) ⇒ Crawler

Returns a new instance of Crawler.



21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
 
# File 'lib/sengi/crawler.rb', line 21

def initialize(url, options)
	@url = url
	@options = options
	
	@options['serial'] = false if !@options.has_key?('serial')
	@options['relative'] = false if !@options.has_key?('relative')
	@options['force'] = false if !@options.has_key?('force')
	@options['debug'] = false if !@options.has_key?('debug')
	
	@options['parent_id'] = 0 if !@options.has_key?('parent_id')
	@options['level'] = 0 if !@options.has_key?('level')
	#pp @options
	
	@redis = nil
	@uri = nil
	@request = nil
	@response = nil
	@html_doc = nil
	@url_delay = nil
	@url_separate_delay = nil
	@url_reschedule = nil
end

Instance Method Details

#go ⇒ Object 



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
 
# File 'lib/sengi/crawler.rb', line 44

def go
	redis_setup
	
	uri_setup
	puts "#{Time.now.strftime('%F %T')} perform: #{@options['parent_id']} #{@options['level']} #{@options['relative'] ? 'y' : 'n'} #{@uri}"
	
	check_blacklist
	puts "\t" + "blacklisted: #{@uri.is_blacklisted ? 'YES' : 'no'}"
	return if @uri.is_blacklisted
	
	insert_url
	puts "\t" + "url: #{@uri.id}"
	if @uri.is_ignored && !@options['debug'] && !@options['force']
		puts "\t" + "ignored reason: #{@uri.is_ignored_reason}"
		return
	end
	
	insert_domain
	puts "\t" + "domain id: #{@uri.domain_id}"
	
	insert_request
	puts "\t" + "request id: #{@uri.request_id}"
	
	make_http_request
	puts "\t" + "http response: #{@response.nil? ? 'FAILED' : 'ok'}"
	return if @response.nil?
	
	insert_response
	puts "\t" + "response: #{@uri.response_id} #{@uri.response_size}"
	
	puts "\t" + 'process http response'
	process_http_response
	puts "\t" + "http response"
	if @uri.is_ignored && !@options['force']
		puts "\t" + "ignored reason: #{@uri.is_ignored_reason}"
		return
	end
	if @html_doc.nil?
		puts "\t" + 'HTML INVALID'
		return
	end
	
	puts "\t" + 'process html links'
	process_html_links
	
	puts "\t" + 'process html meta'
	process_html_meta
	
	puts "\t" + 'url done'
end