onebox 1.4.9 → 1.5.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: acf0205db8c9ea7c0a4c0ded94674fae2c2a7647
-  data.tar.gz: fd639f567d646668041ae8c28a5af22d39512fe8
+  metadata.gz: 5b5e72349ed5dcca5148875d8e669b3850de0311
+  data.tar.gz: b1d032a771038719f1c51064fee5942c50f94353
 SHA512:
-  metadata.gz: 3f4fcf781c785773901b5507e06d746d76e3adc1975074524c8cbb3f687e4d836b7352c4618e4e55a24a42f70550df964bc507ac0d98957f84ffa41f00afacdd
-  data.tar.gz: 2972de535e55661b17473efb5df139d5d7ef2eae4993de8683897960a4edaeae470608b0715adb1496a8e52fa090b8531cbffe06c53cbf9aaececb305072db28
+  metadata.gz: cd575f2a3a215a63a5c9359e788cb5c3740b8b7b1d53ad59cbdf9db2e120fc63c95e6c96432f11b38f5dbd8d9a7fff41ce722fe51c70e3843bbad2b5362be229
+  data.tar.gz: 6cb0cc9aef0081ba65b35221261492b44cf6ca161161d50044c1c108d7732a4f9c840867907d8ca1d9ea308430378cfe5f45909a82a31461fd0939ac85399d96

data/lib/onebox/engine/wikipedia_onebox.rb

@@ -10,26 +10,66 @@ module Onebox
       private
 
       def data
-        # get all the paras
-        paras = raw.search("p")
+        paras = []
         text = ""
 
+        # Detect section Hash in the url and retrive the related paragraphs.  if no hash provided the first few paragraphs will be used
+        # Author Lidlanca
+        # Date 9/8/2014
+        if ( m_url_hash = @url.match /#([^\/?]+)/ ) #extract url hash
+          m_url_hash_name= m_url_hash[1]
+        end
+
+        unless m_url_hash.nil?
+          section_header_title = raw.xpath("//span[@id='#{m_url_hash_name}']") 
+
+          if section_header_title.empty? 
+            paras = raw.search("p") #default get all the paras
+          else 
+            section_title_text = section_header_title.inner_text
+            section_header = section_header_title[0].parent #parent element of the section span element should be an <h3> node
+            cur_element = section_header
+            
+            # p|text|div covers the general case. We assume presence of atleast 1 P node. if section has no P node we may end up with a P node from the next section.
+            # div tag is commonly used as an assets wraper in an article section. often as the first element holding an image.
+            # ul support will imporve the output generated for a section with a list as the main content (for example: an Author Bibliography, A musician Discography, etc)
+            first_p_found = nil 
+            while ( ((next_sibling = cur_element.next_sibling).name =~ /p|text|div|ul/) || first_p_found.nil? ) do  #from section header get the next sibling until it is a breaker tag
+              cur_element = next_sibling
+              if (cur_element.name == "p" || cur_element.name == "ul") #we treat a list as we detect a p to avoid showing 
+                first_p_found = true 
+                paras.push(cur_element)
+              end
+            end
+          end
+        else # no hash found in url
+          paras = raw.search("p") #default get all the paras
+        end
+
         unless paras.empty?
           cnt = 0
           while text.length < Onebox::LayoutSupport.max_text && cnt <= 3
             break if cnt >= paras.size
             text << " " unless cnt == 0
-            paragraph = paras[cnt].inner_text[0..Onebox::LayoutSupport.max_text]
+            
+            if paras[cnt].name =="ul" #Handle UL tag. Generate a textual ordered list (1.item | 2.item | 3.item). Unfourtently no newline allowed in output
+              li_index=1
+              list_items = []
+              paras[cnt].children.css("li").each {|li| list_items.push "#{li_index}." + li.inner_text ; li_index+=1}
+              paragraph =  (list_items.join " |\n ")[0..Onebox::LayoutSupport.max_text]
+            else
+              paragraph = paras[cnt].inner_text[0..Onebox::LayoutSupport.max_text]  
+            end
+            
             paragraph.gsub!(/\[\d+\]/mi, "")
             text << paragraph
             cnt += 1
           end
         end
-
         text = "#{text[0..Onebox::LayoutSupport.max_text]}..." if text.length > Onebox::LayoutSupport.max_text
         result = {
           link: link,
-          title: raw.css("html body h1").inner_text,
+          title: raw.css("html body h1").inner_text + (section_title_text ? " | " + section_title_text : ""),  #if a section sub title exists add it to the main article title
           description: text
         }
         img = raw.css(".image img")

data/lib/onebox/version.rb

@@ -1,3 +1,3 @@
 module Onebox
-  VERSION = "1.4.9"
+  VERSION = "1.5.0"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: onebox
 version: !ruby/object:Gem::Version
-  version: 1.4.9
+  version: 1.5.0
 platform: ruby
 authors:
 - Joanna Zeta
@@ -10,7 +10,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-08-29 00:00:00.000000000 Z
+date: 2014-09-09 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: multi_json
@@ -342,7 +342,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.1.11
+rubygems_version: 2.2.2
 signing_key: 
 specification_version: 4
 summary: A gem for turning URLs into previews.