cobweb 0.0.2 → 0.0.5
Sign up to get free protection for your applications and to get access to all the features.
- data/README.textile +1 -1
- data/lib/cobweb.rb +189 -39
- data/lib/{content_process_job.rb → cobweb_process_job.rb} +2 -2
- data/lib/crawl_job.rb +9 -13
- data/spec/cobweb/cobweb_spec.rb +4 -1
- metadata +19 -5
data/README.textile
CHANGED
data/lib/cobweb.rb
CHANGED
@@ -1,6 +1,7 @@
|
|
1
|
-
require 'rubygems'
|
1
|
+
require 'rubygems'
|
2
2
|
require 'uri'
|
3
3
|
require 'resque'
|
4
|
+
require "addressable/uri"
|
4
5
|
require 'digest/sha1'
|
5
6
|
|
6
7
|
Dir[File.dirname(__FILE__) + '/*.rb'].each do |file|
|
@@ -13,9 +14,12 @@ class CobWeb
|
|
13
14
|
@options = options
|
14
15
|
@options[:follow_redirects] = true unless @options.has_key?(:follow_redirects)
|
15
16
|
@options[:redirect_limit] = 10 unless @options.has_key?(:redirect_limit)
|
16
|
-
@options[:processing_queue] =
|
17
|
+
@options[:processing_queue] = CobwebProcessJob unless @options.has_key?(:processing_queue)
|
18
|
+
@options[:quiet] = true unless @options.has_key?(:quiet)
|
17
19
|
@options[:debug] = false unless @options.has_key?(:debug)
|
18
20
|
@options[:cache] = 300 unless @options.has_key?(:cache)
|
21
|
+
@options[:timeout] = 10 unless @options.has_key?(:timeout)
|
22
|
+
@options[:redis_options] = {} unless @options.has_key?(:redis_options)
|
19
23
|
|
20
24
|
end
|
21
25
|
|
@@ -27,7 +31,7 @@ class CobWeb
|
|
27
31
|
}
|
28
32
|
|
29
33
|
request.merge!(@options)
|
30
|
-
|
34
|
+
|
31
35
|
Resque.enqueue(CrawlJob, request)
|
32
36
|
end
|
33
37
|
|
@@ -35,11 +39,13 @@ class CobWeb
|
|
35
39
|
|
36
40
|
raise "url cannot be nil" if url.nil?
|
37
41
|
|
42
|
+
absolutize = Absolutize.new(url, :output_debug => false, :raise_exceptions => false, :force_escaping => false, :remove_anchors => true)
|
43
|
+
|
38
44
|
# get the unique id for this request
|
39
45
|
unique_id = Digest::SHA1.hexdigest(url)
|
40
46
|
|
41
47
|
# connect to redis
|
42
|
-
redis = NamespacedRedis.new(Redis.new, "cobweb")
|
48
|
+
redis = NamespacedRedis.new(Redis.new(@options[:redis_options]), "cobweb")
|
43
49
|
|
44
50
|
content = {}
|
45
51
|
|
@@ -47,61 +53,205 @@ class CobWeb
|
|
47
53
|
if redis.get(unique_id) and @options[:cache]
|
48
54
|
puts "Cache hit for #{url}" unless @options[:quiet]
|
49
55
|
content = JSON.parse(redis.get(unique_id)).deep_symbolize_keys
|
50
|
-
|
56
|
+
content[:body] = Base64.decode64(content[:body]) unless content[:body].nil? or content[:mime_type].include?("text/html") or content[:mime_type].include?("application/xhtml+xml")
|
57
|
+
|
51
58
|
content
|
52
59
|
else
|
53
60
|
# this url is valid for processing so lets get on with it
|
54
61
|
print "Retrieving #{url }... " unless @options[:quiet]
|
55
|
-
uri = URI.parse(url)
|
62
|
+
uri = Addressable::URI.parse(url.strip)
|
56
63
|
|
57
64
|
# retrieve data
|
58
65
|
http = Net::HTTP.new(uri.host, uri.port)
|
59
66
|
if uri.scheme == "https"
|
60
67
|
http.use_ssl = true
|
61
68
|
http.verify_mode = OpenSSL::SSL::VERIFY_NONE
|
62
|
-
end
|
69
|
+
end
|
70
|
+
|
63
71
|
request_time = Time.now.to_f
|
64
|
-
|
65
|
-
|
66
|
-
|
67
|
-
|
68
|
-
|
69
|
-
|
70
|
-
|
71
|
-
|
72
|
+
http.read_timeout = @options[:timeout].to_i
|
73
|
+
http.open_timeout = @options[:timeout].to_i
|
74
|
+
begin
|
75
|
+
response = http.start() {|http|
|
76
|
+
response = http.get(uri.request_uri)
|
77
|
+
}
|
78
|
+
|
79
|
+
if @options[:follow_redirects] and response.code.to_i >= 300 and response.code.to_i < 400
|
80
|
+
puts "redirected... " unless @options[:quiet]
|
81
|
+
url = absolutize.url(response['location']).to_s
|
82
|
+
ap redirect_limit
|
83
|
+
redirect_limit = redirect_limit - 1
|
84
|
+
content = get(url, redirect_limit)
|
85
|
+
content[:url] = uri.to_s
|
86
|
+
content[:redirect_through] = [] if content[:redirect_through].nil?
|
87
|
+
content[:redirect_through].insert(0, url)
|
88
|
+
|
89
|
+
content[:response_time] = Time.now.to_f - request_time
|
90
|
+
else
|
91
|
+
content[:response_time] = Time.now.to_f - request_time
|
92
|
+
|
93
|
+
puts "Retrieved." unless @options[:quiet]
|
94
|
+
|
95
|
+
# create the content container
|
96
|
+
content[:url] = uri.to_s
|
97
|
+
content[:status_code] = response.code.to_i
|
98
|
+
content[:mime_type] = response.content_type.split(";")[0].strip
|
99
|
+
charset = response["Content-Type"][response["Content-Type"].index(";")+2..-1] if !response["Content-Type"].nil? and response["Content-Type"].include?(";")
|
100
|
+
charset = charset[charset.index("=")+1..-1] if charset and charset.include?("=")
|
101
|
+
content[:character_set] = charset
|
102
|
+
content[:length] = response.content_length
|
103
|
+
if content[:mime_type].include?("text/html") or content[:mime_type].include?("application/xhtml+xml")
|
104
|
+
content[:body] = response.body
|
105
|
+
else
|
106
|
+
content[:body] = Base64.encode64(response.body)
|
107
|
+
end
|
108
|
+
content[:location] = response["location"]
|
109
|
+
content[:headers] = response.to_hash.symbolize_keys
|
110
|
+
# parse data for links
|
111
|
+
link_parser = ContentLinkParser.new(content[:url], content[:body])
|
112
|
+
content[:links] = link_parser.link_data
|
113
|
+
|
114
|
+
# add content to cache if required
|
115
|
+
if @options[:cache]
|
116
|
+
content[:body] = Base64.encode64(content[:body]) unless content[:body].nil? or content[:mime_type].include?("text/html") or content[:mime_type].include?("application/xhtml+xml")
|
117
|
+
redis.set(unique_id, content.to_json)
|
118
|
+
redis.expire unique_id, @options[:cache].to_i
|
119
|
+
end
|
120
|
+
end
|
121
|
+
rescue SocketError => e
|
122
|
+
puts "ERROR: #{e.message}"
|
123
|
+
|
124
|
+
## generate a blank content
|
125
|
+
content = {}
|
72
126
|
content[:url] = uri.to_s
|
73
|
-
content[:
|
74
|
-
content[:
|
127
|
+
content[:response_time] = Time.now.to_f - request_time
|
128
|
+
content[:status_code] = 0
|
129
|
+
content[:length] = 0
|
130
|
+
content[:body] = ""
|
131
|
+
content[:error] = e.message
|
132
|
+
content[:mime_type] = "error/dnslookup"
|
133
|
+
content[:headers] = {}
|
134
|
+
content[:links] = {}
|
135
|
+
|
136
|
+
rescue Timeout::Error => e
|
137
|
+
puts "ERROR: #{e.message}"
|
75
138
|
|
139
|
+
## generate a blank content
|
140
|
+
content = {}
|
141
|
+
content[:url] = uri.to_s
|
76
142
|
content[:response_time] = Time.now.to_f - request_time
|
143
|
+
content[:status_code] = 0
|
144
|
+
content[:length] = 0
|
145
|
+
content[:body] = ""
|
146
|
+
content[:error] = e.message
|
147
|
+
content[:mime_type] = "error/serverdown"
|
148
|
+
content[:headers] = {}
|
149
|
+
content[:links] = {}
|
150
|
+
end
|
151
|
+
end
|
152
|
+
content
|
153
|
+
end
|
154
|
+
|
155
|
+
def head(url, redirect_limit = @options[:redirect_limit])
|
156
|
+
raise "url cannot be nil" if url.nil?
|
157
|
+
|
158
|
+
absolutize = Absolutize.new(url, :output_debug => false, :raise_exceptions => false, :force_escaping => false, :remove_anchors => true)
|
159
|
+
|
160
|
+
# get the unique id for this request
|
161
|
+
unique_id = Digest::SHA1.hexdigest(url)
|
162
|
+
|
163
|
+
# connect to redis
|
164
|
+
redis = NamespacedRedis.new(Redis.new(@options[:redis_options]), "cobweb")
|
165
|
+
|
166
|
+
content = {}
|
167
|
+
|
168
|
+
# check if it has already been cached
|
169
|
+
if (redis.get(unique_id) or redis.get("head-#{unique_id}")) and @options[:cache]
|
170
|
+
puts "Cache hit for #{url}" unless @options[:quiet]
|
171
|
+
if redis.get(unique_id)
|
172
|
+
content = JSON.parse(redis.get(unique_id)).deep_symbolize_keys
|
77
173
|
else
|
78
|
-
content
|
174
|
+
content = JSON.parse(redis.get("head-#{unique_id}")).deep_symbolize_keys
|
175
|
+
end
|
176
|
+
content
|
177
|
+
else
|
178
|
+
print "Retrieving #{url }... " unless @options[:quiet]
|
179
|
+
uri = Addressable::URI.parse(url.strip)
|
180
|
+
|
181
|
+
# retrieve data
|
182
|
+
http = Net::HTTP.new(uri.host, uri.port)
|
183
|
+
if uri.scheme == "https"
|
184
|
+
http.use_ssl = true
|
185
|
+
http.verify_mode = OpenSSL::SSL::VERIFY_NONE
|
186
|
+
end
|
187
|
+
|
188
|
+
request_time = Time.now.to_f
|
189
|
+
http.read_timeout = @options[:timeout].to_i
|
190
|
+
http.open_timeout = @options[:timeout].to_i
|
191
|
+
|
192
|
+
begin
|
193
|
+
response = http.head(uri.to_s)
|
79
194
|
|
80
|
-
|
195
|
+
if @options[:follow_redirects] and response.code.to_i >= 300 and response.code.to_i < 400
|
196
|
+
puts "redirected... " unless @options[:quiet]
|
197
|
+
url = absolutize.url(response['location']).to_s
|
198
|
+
redirect_limit = redirect_limit - 1
|
199
|
+
content = head(url, redirect_limit)
|
200
|
+
content[:url] = uri.to_s
|
201
|
+
content[:redirect_through] = [] if content[:redirect_through].nil?
|
202
|
+
content[:redirect_through].insert(0, url)
|
203
|
+
else
|
204
|
+
content[:url] = uri.to_s
|
205
|
+
content[:status_code] = response.code.to_i
|
206
|
+
unless response.content_type.nil?
|
207
|
+
content[:mime_type] = response.content_type.split(";")[0].strip
|
208
|
+
charset = response["Content-Type"][response["Content-Type"].index(";")+2..-1] if !response["Content-Type"].nil? and response["Content-Type"].include?(";")
|
209
|
+
charset = charset[charset.index("=")+1..-1] if charset and charset.include?("=")
|
210
|
+
content[:character_set] = charset
|
211
|
+
end
|
212
|
+
|
213
|
+
# add content to cache if required
|
214
|
+
if @options[:cache]
|
215
|
+
puts "Stored in cache [head-#{unique_id}]" if @options[:debug]
|
216
|
+
redis.set("head-#{unique_id}", content.to_json)
|
217
|
+
redis.expire "head-#{unique_id}", @options[:cache].to_i
|
218
|
+
else
|
219
|
+
puts "Not storing in cache as cache disabled" if @options[:debug]
|
220
|
+
end
|
221
|
+
end
|
222
|
+
rescue SocketError => e
|
223
|
+
puts "ERROR: #{e.message}"
|
81
224
|
|
82
|
-
|
225
|
+
## generate a blank content
|
226
|
+
content = {}
|
83
227
|
content[:url] = uri.to_s
|
84
|
-
content[:
|
85
|
-
content[:
|
86
|
-
|
87
|
-
|
88
|
-
content[:
|
89
|
-
content[:
|
90
|
-
content[:
|
91
|
-
content[:
|
92
|
-
|
93
|
-
|
94
|
-
|
95
|
-
content[:links] = link_parser.link_data
|
228
|
+
content[:response_time] = Time.now.to_f - request_time
|
229
|
+
content[:status_code] = 0
|
230
|
+
content[:length] = 0
|
231
|
+
content[:body] = ""
|
232
|
+
content[:error] = e.message
|
233
|
+
content[:mime_type] = "error/dnslookup"
|
234
|
+
content[:headers] = {}
|
235
|
+
content[:links] = {}
|
236
|
+
|
237
|
+
rescue Timeout::Error => e
|
238
|
+
puts "ERROR: #{e.message}"
|
96
239
|
|
97
|
-
|
98
|
-
|
99
|
-
|
100
|
-
|
101
|
-
|
240
|
+
## generate a blank content
|
241
|
+
content = {}
|
242
|
+
content[:url] = uri.to_s
|
243
|
+
content[:response_time] = Time.now.to_f - request_time
|
244
|
+
content[:status_code] = 0
|
245
|
+
content[:length] = 0
|
246
|
+
content[:body] = ""
|
247
|
+
content[:error] = e.message
|
248
|
+
content[:mime_type] = "error/serverdown"
|
249
|
+
content[:headers] = {}
|
250
|
+
content[:links] = {}
|
102
251
|
end
|
252
|
+
|
253
|
+
content
|
103
254
|
end
|
104
|
-
content
|
105
255
|
end
|
106
256
|
end
|
107
257
|
|
@@ -126,4 +276,4 @@ class Hash
|
|
126
276
|
end
|
127
277
|
self
|
128
278
|
end
|
129
|
-
end
|
279
|
+
end
|
data/lib/crawl_job.rb
CHANGED
@@ -9,7 +9,7 @@ class CrawlJob
|
|
9
9
|
def self.perform(content_request)
|
10
10
|
# change all hash keys to symbols
|
11
11
|
content_request.deep_symbolize_keys
|
12
|
-
redis = NamespacedRedis.new(Redis.new, "cobweb-#{content_request[:crawl_id]}")
|
12
|
+
redis = NamespacedRedis.new(Redis.new(content_request[:redis_options]), "cobweb-#{content_request[:crawl_id]}")
|
13
13
|
@absolutize = Absolutize.new(content_request[:url], :output_debug => false, :raise_exceptions => false, :force_escaping => false, :remove_anchors => true)
|
14
14
|
|
15
15
|
# check we haven't crawled this url before
|
@@ -19,18 +19,14 @@ class CrawlJob
|
|
19
19
|
redis.incr "crawl-counter"
|
20
20
|
crawl_counter = redis.get("crawl-counter").to_i
|
21
21
|
queue_counter = redis.get("queue-counter").to_i
|
22
|
-
if crawl_counter <= content_request[:crawl_limit]
|
23
|
-
content = CobWeb.get(content_request)
|
24
|
-
redis.sadd "crawled", content_request[:url]
|
22
|
+
if crawl_counter <= content_request[:crawl_limit].to_i
|
23
|
+
content = CobWeb.new(content_request).get(content_request[:url])
|
24
|
+
redis.sadd "crawled", content_request[:url]
|
25
25
|
set_base_url redis, content, content_request[:base_url]
|
26
|
-
if queue_counter <= content_request[:crawl_limit]
|
27
|
-
ap content[:links]
|
26
|
+
if queue_counter <= content_request[:crawl_limit].to_i
|
28
27
|
content[:links].keys.map{|key| content[:links][key]}.flatten.each do |link|
|
29
|
-
ap link
|
30
28
|
unless redis.sismember "crawled", link
|
31
|
-
|
32
|
-
puts "---------------------------------"
|
33
|
-
if link.match(Regexp.new("^#{redis.get("base_url")}"))
|
29
|
+
if link.to_s.match(Regexp.new("^#{redis.get("base_url")}"))
|
34
30
|
new_request = content_request.clone
|
35
31
|
new_request[:url] = link
|
36
32
|
new_request[:parent] = content_request[:url]
|
@@ -42,12 +38,12 @@ class CrawlJob
|
|
42
38
|
end
|
43
39
|
|
44
40
|
# enqueue to processing queue
|
45
|
-
Resque.enqueue(const_get(content_request[:processing_queue]), content)
|
41
|
+
Resque.enqueue(const_get(content_request[:processing_queue]), content.merge({:source_id => content_request[:source_id]}))
|
46
42
|
puts "#{content_request[:url]} has been sent for processing." if content_request[:debug]
|
47
43
|
puts "Crawled: #{crawl_counter} Limit: #{content_request[:crawl_limit]} Queued: #{queue_counter}" if content_request[:debug]
|
48
44
|
|
49
45
|
else
|
50
|
-
puts "Crawl Limit Exceeded by #{crawl_counter - content_request[:crawl_limit]} objects" if content_request[:debug]
|
46
|
+
puts "Crawl Limit Exceeded by #{crawl_counter - content_request[:crawl_limit].to_i} objects" if content_request[:debug]
|
51
47
|
end
|
52
48
|
else
|
53
49
|
puts "Already crawled #{content_request[:url]}" if content_request[:debug]
|
@@ -60,7 +56,7 @@ class CrawlJob
|
|
60
56
|
if content[:status_code] >= 300 and content[:status_code] < 400
|
61
57
|
#redirect received for first url
|
62
58
|
redis.set("base_url", @absolutize.url(content[:location]).to_s)
|
63
|
-
puts "
|
59
|
+
puts "WARNING: base_url given redirects to another location, setting base_url to #{@absolutize.url(content[:location]).to_s}"
|
64
60
|
else
|
65
61
|
redis.set("base_url", base_url)
|
66
62
|
end
|
data/spec/cobweb/cobweb_spec.rb
CHANGED
@@ -36,7 +36,10 @@ describe CobWeb do
|
|
36
36
|
|
37
37
|
@mock_http_client.stub!(:request).with(@mock_http_request).and_return(@mock_http_response)
|
38
38
|
@mock_http_client.stub!(:request).with(@mock_http_redirect_request).and_return(@mock_http_redirect_response)
|
39
|
-
@mock_http_client.stub!(:request).with(@mock_http_redirect_request2).and_return(@mock_http_redirect_response2)
|
39
|
+
@mock_http_client.stub!(:request).with(@mock_http_redirect_request2).and_return(@mock_http_redirect_response2)
|
40
|
+
@mock_http_client.stub!(:read_timeout=).and_return(nil)
|
41
|
+
@mock_http_client.stub!(:open_timeout=).and_return(nil)
|
42
|
+
@mock_http_client.stub!(:start).and_return(@mock_http_response)
|
40
43
|
|
41
44
|
@mock_http_response.stub!(:code).and_return(200)
|
42
45
|
@mock_http_response.stub!(:content_type).and_return("text/html")
|
metadata
CHANGED
@@ -1,13 +1,13 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: cobweb
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
hash:
|
4
|
+
hash: 21
|
5
5
|
prerelease: false
|
6
6
|
segments:
|
7
7
|
- 0
|
8
8
|
- 0
|
9
|
-
-
|
10
|
-
version: 0.0.
|
9
|
+
- 5
|
10
|
+
version: 0.0.5
|
11
11
|
platform: ruby
|
12
12
|
authors:
|
13
13
|
- Stewart McKee
|
@@ -15,7 +15,7 @@ autorequire:
|
|
15
15
|
bindir: bin
|
16
16
|
cert_chain: []
|
17
17
|
|
18
|
-
date:
|
18
|
+
date: 2011-01-28 00:00:00 +00:00
|
19
19
|
default_executable:
|
20
20
|
dependencies:
|
21
21
|
- !ruby/object:Gem::Dependency
|
@@ -74,6 +74,20 @@ dependencies:
|
|
74
74
|
version: "0"
|
75
75
|
type: :runtime
|
76
76
|
version_requirements: *id004
|
77
|
+
- !ruby/object:Gem::Dependency
|
78
|
+
name: addressable
|
79
|
+
prerelease: false
|
80
|
+
requirement: &id005 !ruby/object:Gem::Requirement
|
81
|
+
none: false
|
82
|
+
requirements:
|
83
|
+
- - ">="
|
84
|
+
- !ruby/object:Gem::Version
|
85
|
+
hash: 3
|
86
|
+
segments:
|
87
|
+
- 0
|
88
|
+
version: "0"
|
89
|
+
type: :runtime
|
90
|
+
version_requirements: *id005
|
77
91
|
description:
|
78
92
|
email: stewart@rockwellcottage.com
|
79
93
|
executables: []
|
@@ -90,8 +104,8 @@ files:
|
|
90
104
|
- spec/cobweb/cobweb_spec.rb
|
91
105
|
- lib/namespaced_redis.rb
|
92
106
|
- lib/cobweb.rb
|
93
|
-
- lib/content_process_job.rb
|
94
107
|
- lib/content_link_parser.rb
|
108
|
+
- lib/cobweb_process_job.rb
|
95
109
|
- lib/crawl_job.rb
|
96
110
|
- README.textile
|
97
111
|
has_rdoc: false
|