Home Explore Help
Register Sign In
J1X
/
huginn
Watch 2
Star 0
Fork 0

Browse Source
Add event_data_path option to the WebsiteAgent to handle data directly in incoming Events

Andrew Cantino 10 years ago
parent
e624815eb7
commit
5eeca709c7
2 changed files with 232 additions and 112 deletions
Show Diff Stats
  1. 67 41
      app/models/agents/website_agent.rb
  2. 165 71
      spec/models/agents/website_agent_spec.rb

+ 67 - 41
app/models/agents/website_agent.rb
View File 

@@ -20,7 +20,12 @@ module Agents
20 20
21 21 
       `url` can be a single url, or an array of urls (for example, for multiple pages with the exact same structure but different content to scrape)
22 22
23 
-      The WebsiteAgent can also scrape based on incoming events. It will scrape the url contained in the `url` key of the incoming event payload, or if you set `url_from_event` it is used as a Liquid template to generate the url to access. If you specify `merge` as the `mode`, it will retain the old payload and update it with the new values.
23 
+      The WebsiteAgent can also scrape based on incoming events.
24 
+
25 
+      * If the Event contains a `url` key, that URL will be fetched.
26 
+      * For more control, you can set the `url_from_event` option and it will be used as a Liquid template to generate the url to access based on the Event.
27 
+      * If you set `event_data_path` to the [JSONPath](http://goessner.net/articles/JsonPath/) of content in the Event, that will be used directly without fetching any URL.
28 
+      * If you specify `merge` for the `mode` option, Huginn will retain the old payload and update it with the new values.
24 29
25 30 
       # Supported Document Types
26 31 
 
@@ -140,7 +145,7 @@ module Agents
140 145
141 146 
     def validate_options
142 147 
       # Check for required fields
143 
-      errors.add(:base, "either url or url_from_event is required") unless options['url'].present? || options['url_from_event'].present?
148 
+      errors.add(:base, "either url, url_from_event, or event_data_path are required") unless options['url'].present? || options['url_from_event'].present? || options['event_data_path'].present?
144 149 
       errors.add(:base, "expected_update_period_in_days is required") unless options['expected_update_period_in_days'].present?
145 150 
       validate_extract_options!
146 151 
 
@@ -251,15 +256,15 @@ module Agents
251 256 
       check_urls(interpolated['url'])
252 257 
     end
253 258
254 
-    def check_urls(in_url, payload = {})
259 
+    def check_urls(in_url, existing_payload = {})
255 260 
       return unless in_url.present?
256 261
257 262 
       Array(in_url).each do |url|
258 
-        check_url(url, payload)
263 
+        check_url(url, existing_payload)
259 264 
       end
260 265 
     end
261 266
262 
-    def check_url(url, payload = {})
267 
+    def check_url(url, existing_payload = {})
263 268 
       unless /\Ahttps?:\/\//i === url
264 269 
         error "Ignoring a non-HTTP url: #{url.inspect}"
265 270 
         return
 
@@ -271,70 +276,91 @@ module Agents
271 276
272 277 
       interpolation_context.stack {
273 278 
         interpolation_context['_response_'] = ResponseDrop.new(response)
274 
-        body = response.body
275 
-        doc = parse(body)
279 
+        handle_data(response.body, response.env[:url], existing_payload)
280 
+      }
281 
+    rescue => e
282 
+      error "Error when fetching url: #{e.message}\n#{e.backtrace.join("\n")}"
283 
+    end
276 284
277 
-        if extract_full_json?
278 
-          if store_payload!(previous_payloads(1), doc)
279 
-            log "Storing new result for '#{name}': #{doc.inspect}"
280 
-            create_event payload: payload.merge(doc)
281 
-          end
282 
-          return
285 
+    def handle_data(body, url, existing_payload)
286 
+      doc = parse(body)
287 
+
288 
+      if extract_full_json?
289 
+        if store_payload!(previous_payloads(1), doc)
290 
+          log "Storing new result for '#{name}': #{doc.inspect}"
291 
+          create_event payload: existing_payload.merge(doc)
283 292 
         end
293 
+        return
294 
+      end
284 295
285 
-        output =
286 
-          case extraction_type
296 
+      output =
297 
+        case extraction_type
287 298 
           when 'json'
288 299 
             extract_json(doc)
289 300 
           when 'text'
290 301 
             extract_text(doc)
291 302 
           else
292 303 
             extract_xml(doc)
293 
-          end
304 
+        end
294 305
295 
-        num_unique_lengths = interpolated['extract'].keys.map { |name| output[name].length }.uniq
306 
+      num_unique_lengths = interpolated['extract'].keys.map { |name| output[name].length }.uniq
296 307
297 
-        if num_unique_lengths.length != 1
298 
-          raise "Got an uneven number of matches for #{interpolated['name']}: #{interpolated['extract'].inspect}"
299 
-        end
308 
+      if num_unique_lengths.length != 1
309 
+        raise "Got an uneven number of matches for #{interpolated['name']}: #{interpolated['extract'].inspect}"
310 
+      end
300 311
301 
-        old_events = previous_payloads num_unique_lengths.first
302 
-        num_unique_lengths.first.times do |index|
303 
-          result = {}
304 
-          interpolated['extract'].keys.each do |name|
305 
-            result[name] = output[name][index]
306 
-            if name.to_s == 'url'
307 
-              result[name] = (response.env[:url] + Utils.normalize_uri(result[name])).to_s
308 
-            end
312 
+      old_events = previous_payloads num_unique_lengths.first
313 
+      num_unique_lengths.first.times do |index|
314 
+        result = {}
315 
+        interpolated['extract'].keys.each do |name|
316 
+          result[name] = output[name][index]
317 
+          if name.to_s == 'url' && url.present?
318 
+            result[name] = (url + Utils.normalize_uri(result[name])).to_s
309 319 
           end
320 
+        end
310 321
311 
-          if store_payload!(old_events, result)
312 
-            log "Storing new parsed result for '#{name}': #{result.inspect}"
313 
-            create_event payload: payload.merge(result)
314 
-          end
322 
+        if store_payload!(old_events, result)
323 
+          log "Storing new parsed result for '#{name}': #{result.inspect}"
324 
+          create_event payload: existing_payload.merge(result)
315 325 
         end
316 
-      }
317 
-    rescue => e
318 
-      error "Error when fetching url: #{e.message}\n#{e.backtrace.join("\n")}"
326 
+      end
319 327 
     end
320 328
321 329 
     def receive(incoming_events)
322 330 
       incoming_events.each do |event|
323 331 
         interpolate_with(event) do
324 
-          url_to_scrape =
325 
-            if url_template = options['url_from_event'].presence
326 
-              interpolate_options(url_template)
332 
+          existing_payload = interpolated['mode'].to_s == "merge" ? event.payload : {}
333 
+
334 
+          if event_data_path = options['event_data_path'].presence
335 
+            data = Utils.value_at(event.payload, interpolate_options(event_data_path))
336 
+            if data.present?
337 
+              handle_event_data(data, event, existing_payload)
327 338 
             else
328 
-              event.payload['url']
339 
+              error "No data was found in the Event payload at the JSONPath #{interpolate_options(event_data_path)}", inbound_event: event
329 340 
             end
330 
-          check_urls(url_to_scrape,
331 
-                    interpolated['mode'].to_s == "merge" ? event.payload : {})
341 
+          else
342 
+            url_to_scrape =
343 
+              if event_data_path = options['event_data_path'].presence
344 
+                interpolate_options(event_data_path)
345 
+              elsif url_template = options['url_from_event'].presence
346 
+                interpolate_options(url_template)
347 
+              else
348 
+                event.payload['url']
349 
+              end
350 
+            check_urls(url_to_scrape, existing_payload)
351 
+          end
332 352 
         end
333 353 
       end
334 354 
     end
335 355
336 356 
     private
337 357
358 
+    def handle_event_data(data, event, existing_payload)
359 
+      handle_data(data, event.payload['url'], existing_payload)
360 
+    rescue => e
361 
+      error "Error when handling event data: #{e.message}\n#{e.backtrace.join("\n")}", inbound_event: event
362 
+    end
363 
+
338 364 
     # This method returns true if the result should be stored as a new event.
339 365 
     # If mode is set to 'on_change', this method may return false and update an existing
340 366 
     # event to expire further in the future.

+ 165 - 71
spec/models/agents/website_agent_spec.rb
View File 

@@ -763,92 +763,186 @@ fire: hot
763 763 
     end
764 764
765 765 
     describe "#receive" do
766 
-      before do
767 
-        @event = Event.new
768 
-        @event.agent = agents(:bob_rain_notifier_agent)
769 
-        @event.payload = {
770 
-          'url' => 'http://xkcd.com',
771 
-          'link' => 'Random',
772 
-        }
773 
-      end
766 
+      describe "with a url or url_from_event" do
767 
+        before do
768 
+          @event = Event.new
769 
+          @event.agent = agents(:bob_rain_notifier_agent)
770 
+          @event.payload = {
771 
+            'url' => 'http://xkcd.com',
772 
+            'link' => 'Random',
773 
+          }
774 
+        end
774 775
775 
-      it "should scrape from the url element in incoming event payload" do
776 
-        expect {
777 
-          @checker.options = @valid_options
776 
+        it "should scrape from the url element in incoming event payload" do
777 
+          expect {
778 
+            @checker.options = @valid_options
779 
+            @checker.receive([@event])
780 
+          }.to change { Event.count }.by(1)
781 
+        end
782 
+
783 
+        it "should use url_from_event as url to scrape if it exists when receiving an event" do
784 
+          stub = stub_request(:any, 'http://example.org/?url=http%3A%2F%2Fxkcd.com')
785 
+
786 
+          @checker.options = @valid_options.merge(
787 
+            'url_from_event' => 'http://example.org/?url={{url | uri_escape}}'
788 
+          )
778 789 
           @checker.receive([@event])
779 
-        }.to change { Event.count }.by(1)
780 
-      end
781 790
782 
-      it "should use url_from_event as url to scrape if it exists when receiving an event" do
783 
-        stub = stub_request(:any, 'http://example.org/?url=http%3A%2F%2Fxkcd.com')
791 
+          expect(stub).to have_been_requested
792 
+        end
784 793
785 
-        @checker.options = @valid_options.merge(
786 
-          'url_from_event' => 'http://example.org/?url={{url | uri_escape}}'
787 
-        )
788 
-        @checker.receive([@event])
794 
+        it "should allow url_from_event to be an array of urls" do
795 
+          stub1 = stub_request(:any, 'http://example.org/?url=http%3A%2F%2Fxkcd.com')
796 
+          stub2 = stub_request(:any, 'http://google.org/?url=http%3A%2F%2Fxkcd.com')
789 797
790 
-        expect(stub).to have_been_requested
791 
-      end
798 
+          @checker.options = @valid_options.merge(
799 
+            'url_from_event' => ['http://example.org/?url={{url | uri_escape}}', 'http://google.org/?url={{url | uri_escape}}']
800 
+          )
801 
+          @checker.receive([@event])
792 802
793 
-      it "should allow url_from_event to be an array of urls" do
794 
-        stub1 = stub_request(:any, 'http://example.org/?url=http%3A%2F%2Fxkcd.com')
795 
-        stub2 = stub_request(:any, 'http://google.org/?url=http%3A%2F%2Fxkcd.com')
803 
+          expect(stub1).to have_been_requested
804 
+          expect(stub2).to have_been_requested
805 
+        end
796 806
797 
-        @checker.options = @valid_options.merge(
798 
-          'url_from_event' => ['http://example.org/?url={{url | uri_escape}}', 'http://google.org/?url={{url | uri_escape}}']
799 
-        )
800 
-        @checker.receive([@event])
807 
+        it "should interpolate values from incoming event payload" do
808 
+          expect {
809 
+            @valid_options['extract'] = {
810 
+              'from' => {
811 
+                'xpath' => '*[1]',
812 
+                'value' => '{{url | to_xpath}}'
813 
+              },
814 
+              'to' => {
815 
+                'xpath' => '(//a[@href and text()={{link | to_xpath}}])[1]',
816 
+                'value' => '@href'
817 
+              },
818 
+            }
819 
+            @checker.options = @valid_options
820 
+            @checker.receive([@event])
821 
+          }.to change { Event.count }.by(1)
801 822
802 
-        expect(stub1).to have_been_requested
803 
-        expect(stub2).to have_been_requested
804 
-      end
823 
+          expect(Event.last.payload).to eq({
824 
+            'from' => 'http://xkcd.com',
825 
+            'to' => 'http://dynamic.xkcd.com/random/comic/',
826 
+          })
827 
+        end
805 828
806 
-      it "should interpolate values from incoming event payload" do
807 
-        expect {
808 
-          @valid_options['extract'] = {
809 
-            'from' => {
810 
-              'xpath' => '*[1]',
811 
-              'value' => '{{url | to_xpath}}'
812 
-            },
813 
-            'to' => {
814 
-              'xpath' => '(//a[@href and text()={{link | to_xpath}}])[1]',
815 
-              'value' => '@href'
816 
-            },
817 
-          }
818 
-          @checker.options = @valid_options
819 
-          @checker.receive([@event])
820 
-        }.to change { Event.count }.by(1)
829 
+        it "should interpolate values from incoming event payload and _response_" do
830 
+          @event.payload['title'] = 'XKCD'
821 831
822 
-        expect(Event.last.payload).to eq({
823 
-          'from' => 'http://xkcd.com',
824 
-          'to' => 'http://dynamic.xkcd.com/random/comic/',
825 
-        })
826 
-      end
832 
+          expect {
833 
+            @valid_options['extract'] = {
834 
+              'response_info' => @valid_options['extract']['url'].merge(
835 
+                'value' => '{% capture sentence %}The reponse from {{title}} was {{_response_.status}} {{_response_.headers.X-Status-Message}}.{% endcapture %}{{sentence | to_xpath}}'
836 
+              )
837 
+            }
838 
+            @checker.options = @valid_options
839 
+            @checker.receive([@event])
840 
+          }.to change { Event.count }.by(1)
827 841
828 
-      it "should interpolate values from incoming event payload and _response_" do
829 
-        @event.payload['title'] = 'XKCD'
842 
+          expect(Event.last.payload['response_info']).to eq('The reponse from XKCD was 200 OK.')
843 
+        end
830 844
831 
-        expect {
832 
-          @valid_options['extract'] = {
833 
-            'response_info' => @valid_options['extract']['url'].merge(
834 
-              'value' => '{% capture sentence %}The reponse from {{title}} was {{_response_.status}} {{_response_.headers.X-Status-Message}}.{% endcapture %}{{sentence | to_xpath}}'
845 
+        it "should support merging of events" do
846 
+          expect {
847 
+            @checker.options = @valid_options
848 
+            @checker.options[:mode] = "merge"
849 
+            @checker.receive([@event])
850 
+          }.to change { Event.count }.by(1)
851 
+          last_payload = Event.last.payload
852 
+          expect(last_payload['link']).to eq('Random')
853 
+        end
854 
+      end
855 
+
856 
+      describe "with a event_data_path" do
857 
+        describe "with json data" do
858 
+          before do
859 
+            @event = Event.new
860 
+            @event.agent = agents(:bob_rain_notifier_agent)
861 
+            @event.payload = {
862 
+              'something' => 'some value',
863 
+              'some_object' => {
864 
+                'some_data' => { hello: 'world' }.to_json
865 
+              }
866 
+            }
867 
+            @event.save!
868 
+
869 
+            @checker.options = @valid_options.merge(
870 
+              'type' => 'json',
871 
+              'event_data_path' => 'some_object.some_data',
872 
+              'extract' => {
873 
+                'value' => { 'path' => 'hello' }
874 
+              }
875 
+            )
876 
+          end
877 
+
878 
+          it "should extract from the event data in the incoming event payload" do
879 
+            expect {
880 
+              @checker.receive([@event])
881 
+            }.to change { Event.count }.by(1)
882 
+            expect(@checker.events.last.payload).to eq({ 'value' => 'world' })
883 
+          end
884 
+
885 
+          it "should support merge mode" do
886 
+            @checker.options['mode'] = "merge"
887 
+
888 
+            expect {
889 
+              @checker.receive([@event])
890 
+            }.to change { Event.count }.by(1)
891 
+            expect(@checker.events.last.payload).to eq(@event.payload.merge('value' => 'world'))
892 
+          end
893 
+
894 
+          it "should output an error when nothing can be found at the path" do
895 
+            @checker.options = @checker.options.merge(
896 
+              'event_data_path' => 'some_object.mistake'
835 897 
             )
836 
-          }
837 
-          @checker.options = @valid_options
838 
-          @checker.receive([@event])
839 
-        }.to change { Event.count }.by(1)
840 898
841 
-        expect(Event.last.payload['response_info']).to eq('The reponse from XKCD was 200 OK.')
842 
-      end
899 
+            expect {
900 
+              @checker.receive([@event])
901 
+            }.to_not change { Event.count }
843 902
844 
-      it "should support merging of events" do
845 
-        expect {
846 
-          @checker.options = @valid_options
847 
-          @checker.options[:mode] = "merge"
848 
-          @checker.receive([@event])
849 
-        }.to change { Event.count }.by(1)
850 
-        last_payload = Event.last.payload
851 
-        expect(last_payload['link']).to eq('Random')
903 
+            expect(@checker.logs.last.message).to match(/No data was found in the Event payload at the JSONPath some_object.mistake/)
904 
+          end
905 
+
906 
+          it "should output an error when the data cannot be parsed" do
907 
+            @event.update_attribute :payload, @event.payload.merge('some_object' => { 'some_data' => '{invalid json' })
908 
+
909 
+            expect {
910 
+              @checker.receive([@event])
911 
+            }.to_not change { Event.count }
912 
+
913 
+            expect(@checker.logs.last.message).to match(/Error when handling event data:/)
914 
+          end
915 
+        end
916 
+
917 
+        describe "with HTML data" do
918 
+          before do
919 
+            @event = Event.new
920 
+            @event.agent = agents(:bob_rain_notifier_agent)
921 
+            @event.payload = {
922 
+              'url' => 'http://xkcd.com',
923 
+              'some_object' => {
924 
+                'some_data' => "<div><span class='title'>Title!</span><span class='body'>Body!</span></div>"
925 
+              }
926 
+            }
927 
+            @event.save!
928 
+
929 
+            @checker.options = @valid_options.merge(
930 
+              'type' => 'html',
931 
+              'event_data_path' => 'some_object.some_data',
932 
+              'extract' => {
933 
+                'title' => { 'css' => ".title", 'value' => ".//text()" },
934 
+                'body' => { 'css' => "div span.body", 'value' => ".//text()" }
935 
+              }
936 
+            )
937 
+          end
938 
+
939 
+          it "should extract from the event data in the incoming event payload" do
940 
+            expect {
941 
+              @checker.receive([@event])
942 
+            }.to change { Event.count }.by(1)
943 
+            expect(@checker.events.last.payload).to eq({ 'title' => 'Title!', 'body' => 'Body!' })
944 
+          end
945 
+        end
852 946 
       end
853 947 
     end
854 948 
   end