Replace Hpricot with Nokogiri

lib/cooked_post_processor.rb

@@ -9,7 +9,7 @@ class CookedPostProcessor
     @dirty = false
     @opts = opts
     @post = post
-    @doc = Hpricot(post.cooked)
+    @doc = Nokogiri::HTML(post.cooked)
   end
 
   def dirty?

lib/oneboxer.rb

@@ -34,7 +34,7 @@ module Oneboxer
     if Whitelist.allowed?(url)
       page_html = open(url).read
       if page_html.present?
-        doc = Hpricot(page_html)
+        doc = Nokogiri::HTML(page_html)
 
         # See if if it has an oembed thing we can use
         (doc/"link[@type='application/json+oembed']").each do |oembed|
@@ -56,7 +56,7 @@ module Oneboxer
   # Parse URLs out of HTML, returning the document when finished.
   def self.each_onebox_link(string_or_doc)
     doc = string_or_doc
-    doc = Hpricot(doc) if doc.is_a?(String)
+    doc = Nokogiri::HTML(doc) if doc.is_a?(String)
 
     onebox_links = doc.search("a.onebox")
     if onebox_links.present?

lib/oneboxer/amazon_onebox.rb

@@ -22,19 +22,19 @@ module Oneboxer
     end
 
     def parse(data)
-      hp = Hpricot(data)
+      html_doc = Nokogiri::HTML(data)
 
       result = {}
-      result[:title] = hp.at("h1")
+      result[:title] = html_doc.at("h1")
       result[:title] = result[:title].inner_html if result[:title].present?
 
-      image = hp.at(".main-image img")
+      image = html_doc.at(".main-image img")
       result[:image] = image['src'] if image
 
-      result[:by_info] = hp.at("#by-line")
+      result[:by_info] = html_doc.at("#by-line")
       result[:by_info] = BaseOnebox.remove_whitespace(result[:by_info].inner_html) if result[:by_info].present?
 
-      summary = hp.at("#description-and-details-content") 
+      summary = html_doc.at("#description-and-details-content") 
       result[:text] = summary.inner_html if summary.present?
 
       result

lib/oneboxer/android_app_store_onebox.rb

@@ -12,20 +12,20 @@ module Oneboxer
     
     def parse(data)
 
-      hp = Hpricot(data)
+      html_doc = Nokogiri::HTML(data)
 
       result = {}
 
-      m = hp.at("h1.doc-banner-title")
+      m = html_doc.at("h1.doc-banner-title")
       result[:title] = m.inner_text if m
 
-      m = hp.at("div#doc-original-text")
+      m = html_doc.at("div#doc-original-text")
       if m
         result[:text] = BaseOnebox.replace_tags_with_spaces(m.inner_html)
         result[:text] = result[:text][0..MAX_TEXT]
       end
 
-      m = hp.at("div.doc-banner-icon img")
+      m = html_doc.at("div.doc-banner-icon img")
       result[:image] = m['src'] if m
 
       result

lib/oneboxer/apple_app_onebox.rb

@@ -17,17 +17,17 @@ module Oneboxer
     
     def parse(data)
 
-      hp = Hpricot(data)
+      html_doc = Nokogiri::HTML(data)
 
       result = {}
 
-      m = hp.at("h1")
+      m = html_doc.at("h1")
       result[:title] = m.inner_text if m
 
-      m = hp.at("h4 ~ p")
+      m = html_doc.at("h4 ~ p")
       result[:text] = m.inner_text[0..MAX_TEXT] if m
 
-      m = hp.at(".product img.artwork")
+      m = html_doc.at(".product img.artwork")
       result[:image] = m['src'] if m
 
       result

lib/oneboxer/flickr_onebox.rb

@@ -9,7 +9,7 @@ module Oneboxer
 
       page_html = open(@url).read
       return nil if page_html.blank?
-      doc = Hpricot(page_html)
+      doc = Nokogiri::HTML(page_html)
 
       # Flikrs oembed just stopped returning images for no reason. Let's use opengraph instead.
       open_graph = Oneboxer.parse_open_graph(doc)

lib/oneboxer/wikipedia_onebox.rb

@@ -20,23 +20,23 @@ module Oneboxer
 
     def parse(data)
 
-      hp = Hpricot(data)
+      html_doc = Nokogiri::HTML(data)
 
       result = {}
 
-      title = hp.at('title').inner_html
+      title = html_doc.at('title').inner_html
       result[:title] = title.gsub!(/ - Wikipedia, the free encyclopedia/, '') if title.present?
 
       # get the first image > 150 pix high
-      images = hp.search("img").select { |img| img['height'].to_i > 150 }
+      images = html_doc.search("img").select { |img| img['height'].to_i > 150 }
       
       result[:image] = "http:#{images[0]["src"]}" unless images.empty?
 
       # remove the table from mobile layout, as it can contain paras in some rare cases
-      hp.search("table").remove
+      html_doc.search("table").remove
 
       # get all the paras
-      paras = hp.search("p")
+      paras = html_doc.search("p")
       text = ""
 
       unless paras.empty?