首页 探索 帮助
注册 登录
J1X
/
huginn
关注 2
点赞 0
派生 0

浏览代码
Modified website_agent to take an array of urls

Maximilian Clarke 11 年之前
父节点
b661bf36da
当前提交
19c005fe45
共有 2 个文件被更改，包括 96 次插入 和 60 次删除
显示文件统计
  1. 72 60
      app/models/agents/website_agent.rb
  2. 24 0
      spec/models/agents/website_agent_spec.rb

+ 72 - 60
app/models/agents/website_agent.rb
查看文件 

@@ -107,85 +107,97 @@ module Agents
107 107 
       log "Fetching #{options['url']}"
108 108 
       request_opts = { :followlocation => true }
109 109 
       request_opts[:userpwd] = options['basic_auth'] if options['basic_auth'].present?
110 
-      request = Typhoeus::Request.new(options['url'], request_opts)
111 110
112 
-      request.on_failure do |response|
113 
-        error "Failed: #{response.inspect}"
111 
+      requests = []
112 
+
113 
+      if options['url'].kind_of?(Array)
114 
+        options['url'].each do |url|
115 
+           requests.push(Typhoeus::Request.new(url, request_opts))
116 
+        end
117 
+      else
118 
+        requests.push(Typhoeus::Request.new(options['url'], request_opts))
114 119 
       end
115 120
116 
-      request.on_success do |response|
117 
-        body = response.body
118 
-        if (encoding = options['force_encoding']).present?
119 
-          body = body.encode(Encoding::UTF_8, encoding)
121 
+      requests.each do |request|
122 
+        request.on_failure do |response|
123 
+          error "Failed: #{response.inspect}"
120 124 
         end
121 
-        doc = parse(body)
122 125
123 
-        if extract_full_json?
124 
-          if store_payload!(previous_payloads(1), doc)
125 
-            log "Storing new result for '#{name}': #{doc.inspect}"
126 
-            create_event :payload => doc
126 
+        request.on_success do |response|
127 
+          body = response.body
128 
+          if (encoding = options['force_encoding']).present?
129 
+            body = body.encode(Encoding::UTF_8, encoding)
127 130 
           end
128 
-        else
129 
-          output = {}
130 
-          options['extract'].each do |name, extraction_details|
131 
-            if extraction_type == "json"
132 
-              result = Utils.values_at(doc, extraction_details['path'])
133 
-              log "Extracting #{extraction_type} at #{extraction_details['path']}: #{result}"
134 
-            else
135 
-              case
136 
-              when css = extraction_details['css']
137 
-                nodes = doc.css(css)
138 
-              when xpath = extraction_details['xpath']
139 
-                nodes = doc.xpath(xpath)
131 
+          doc = parse(body)
132 
+
133 
+          if extract_full_json?
134 
+            if store_payload!(previous_payloads(1), doc)
135 
+              log "Storing new result for '#{name}': #{doc.inspect}"
136 
+              create_event :payload => doc
137 
+            end
138 
+          else
139 
+            output = {}
140 
+            options['extract'].each do |name, extraction_details|
141 
+              if extraction_type == "json"
142 
+                result = Utils.values_at(doc, extraction_details['path'])
143 
+                log "Extracting #{extraction_type} at #{extraction_details['path']}: #{result}"
140 144 
               else
141 
-                error "'css' or 'xpath' is required for HTML or XML extraction"
142 
-                return
143 
-              end
144 
-              unless Nokogiri::XML::NodeSet === nodes
145 
-                error "The result of HTML/XML extraction was not a NodeSet"
146 
-                return
147 
-              end
148 
-              result = nodes.map { |node|
149 
-                if extraction_details['attr']
150 
-                  node.attr(extraction_details['attr'])
151 
-                elsif extraction_details['text']
152 
-                  node.text()
145 
+                case
146 
+                when css = extraction_details['css']
147 
+                  nodes = doc.css(css)
148 
+                when xpath = extraction_details['xpath']
149 
+                  nodes = doc.xpath(xpath)
153 150 
                 else
154 
-                  error "'attr' or 'text' is required on HTML or XML extraction patterns"
151 
+                  error "'css' or 'xpath' is required for HTML or XML extraction"
152 
+                  return
153 
+                end
154 
+                unless Nokogiri::XML::NodeSet === nodes
155 
+                  error "The result of HTML/XML extraction was not a NodeSet"
155 156 
                   return
156 157 
                 end
157 
-              }
158 
-              log "Extracting #{extraction_type} at #{xpath || css}: #{result}"
158 
+                result = nodes.map { |node|
159 
+                  if extraction_details['attr']
160 
+                    node.attr(extraction_details['attr'])
161 
+                  elsif extraction_details['text']
162 
+                    node.text()
163 
+                  else
164 
+                    error "'attr' or 'text' is required on HTML or XML extraction patterns"
165 
+                    return
166 
+                  end
167 
+                }
168 
+                log "Extracting #{extraction_type} at #{xpath || css}: #{result}"
169 
+              end
170 
+              output[name] = result
159 171 
             end
160 
-            output[name] = result
161 
-          end
162 172
163 
-          num_unique_lengths = options['extract'].keys.map { |name| output[name].length }.uniq
173 
+            num_unique_lengths = options['extract'].keys.map { |name| output[name].length }.uniq
164 174
165 
-          if num_unique_lengths.length != 1
166 
-            error "Got an uneven number of matches for #{options['name']}: #{options['extract'].inspect}"
167 
-            return
168 
-          end
169 
-
170 
-          old_events = previous_payloads num_unique_lengths.first
171 
-          num_unique_lengths.first.times do |index|
172 
-            result = {}
173 
-            options['extract'].keys.each do |name|
174 
-              result[name] = output[name][index]
175 
-              if name.to_s == 'url'
176 
-                result[name] = URI.join(options['url'], result[name]).to_s if (result[name] =~ URI::DEFAULT_PARSER.regexp[:ABS_URI]).nil?
177 
-              end
175 
+            if num_unique_lengths.length != 1
176 
+              error "Got an uneven number of matches for #{options['name']}: #{options['extract'].inspect}"
177 
+              return
178 178 
             end
179 
+
180 
+            old_events = previous_payloads num_unique_lengths.first
181 
+            num_unique_lengths.first.times do |index|
182 
+              result = {}
183 
+              options['extract'].keys.each do |name|
184 
+                result[name] = output[name][index]
185 
+                if name.to_s == 'url'
186 
+                  result[name] = URI.join(options['url'], result[name]).to_s if (result[name] =~ URI::DEFAULT_PARSER.regexp[:ABS_URI]).nil?
187 
+                end
188 
+              end
179 189
180 
-            if store_payload!(old_events, result)
181 
-              log "Storing new parsed result for '#{name}': #{result.inspect}"
182 
-              create_event :payload => result
190 
+              if store_payload!(old_events, result)
191 
+                log "Storing new parsed result for '#{name}': #{result.inspect}"
192 
+                create_event :payload => result
193 
+              end
183 194 
             end
184 195 
           end
185 196 
         end
197 
+
198 
+        hydra.queue request
199 
+        hydra.run
186 200 
       end
187 
-      hydra.queue request
188 
-      hydra.run
189 201 
     end
190 202
191 203 
     private

+ 24 - 0
spec/models/agents/website_agent_spec.rb
查看文件 

@@ -91,6 +91,30 @@ describe Agents::WebsiteAgent do
91 91 
         @checker.check
92 92 
         @checker.logs.first.message.should =~ /Got an uneven number of matches/
93 93 
       end
94 
+
95 
+      it "should accept an array for url" do
96 
+        @site['url'] = ["http://xkcd.com/1/", "http://xkcd.com/2/"]
97 
+        @checker.options = @site
98 
+        lambda { @checker.save! }.should_not raise_error;
99 
+        lambda { @checker.check }.should_not raise_error;
100 
+      end
101 
+
102 
+      it "should parse events from all urls in array" do
103 
+        lambda {
104 
+          @site['url'] = ["http://xkcd.com/", "http://xkcd.com/"]
105 
+          @site['mode'] = 'all'
106 
+          @checker.options = @site
107 
+          @checker.check
108 
+        }.should change { Event.count }.by(2)
109 
+      end
110 
+
111 
+      it "should follow unique rules when parsing array of urls" do
112 
+        lambda {
113 
+          @site['url'] = ["http://xkcd.com/", "http://xkcd.com/"]
114 
+          @checker.options = @site
115 
+          @checker.check
116 
+        }.should change { Event.count }.by(1)
117 
+      end
94 118 
     end
95 119
96 120 
     describe 'encoding' do