onebox 1.1.0 → 1.2.0

data/lib/onebox/engine/github_commit_onebox.rb

@@ -2,6 +2,7 @@ module Onebox
   module Engine
     class GithubCommitOnebox
       include Engine
+      include LayoutSupport
       include JSON
 
       matches do
@@ -14,29 +15,21 @@ module Onebox
       end
 
       def url
-        "https://api.github.com/repos/#{match[:owner]}/#{match[:repo]}/commits/#{match[:number]}"
+        "https://api.github.com/repos/#{match[:owner]}/#{match[:repository]}/commits/#{match[:sha]}"
       end
 
       private
 
       def match
-        @url.match(/github\.com\/(?<owner>[^\/]+)\/(?<repo>[^\/]+)\/commit\/(?<number>[^\/]+)/)
+        @match ||= @url.match(%{github\.com/(?<owner>[^/]+)/(?<repository>[^/]+)/commit/(?<sha>[^/]+)})
       end
 
       def data
-        {
-          url: @url,
-          owner: match[:owner],
-          repo: match[:repo],
-          sha: raw["sha"],
-          gravatar: raw["author"]["avatar_url"],
-          message: raw["commit"]["message"],
-          author: raw["author"]["login"],
-          time_date: raw["commit"]["committer"]["date"],
-          files_changed: raw["files"].length,
-          additions: raw["stats"]["additions"],
-          deletions: raw["stats"]["deletions"]
-        }
+        result = raw.clone
+        result['link'] = link
+        result['title'] = result['commit']['message']
+        result['commit_date'] = Time.parse(result['commit']['author']['date']).strftime("%I:%M%p - %d %b %y")
+        result
       end
     end
   end

data/lib/onebox/engine/github_gist_onebox.rb

@@ -2,7 +2,6 @@ module Onebox
   module Engine
     class GithubGistOnebox
       include Engine
-      include JSON
 
       matches do
         http
@@ -12,22 +11,23 @@ module Onebox
       end
 
       def url
-        "https://api.github.com/gists/#{match[:number]}"
+        "https://api.github.com/gists/#{match[:sha]}"
+      end
+
+      def to_html
+        "<script src=\"http://gist.github.com/#{match[:sha]}.js\"></script>"
       end
 
       private
 
-      def match
-        @url.match(/gist\.github\.com\/([^\/]+\/)?(?<number>[0-9a-f]+)/)
+      def data
+        { sha: match[:sha], title: match[:sha], link: @url }
       end
 
-      def data
-        {
-          url: @url,
-          content: raw["files"].first[1]["content"],
-          author: raw["user"]["login"]
-        }
+      def match
+        @match ||= @url.match(%r{gist\.github\.com/([^/]+/)?(?<sha>[0-9a-f]+)})
       end
+
     end
   end
 end

data/lib/onebox/engine/github_pullrequest_onebox.rb

@@ -2,6 +2,7 @@ module Onebox
   module Engine
     class GithubPullRequestOnebox
       include Engine
+      include LayoutSupport
       include JSON
 
       matches do
@@ -14,29 +15,20 @@ module Onebox
       end
 
       def url
-        "https://api.github.com/repos/#{match[:owner]}/#{match[:repo]}/pulls/#{match[:number]}"
+        "https://api.github.com/repos/#{match[:owner]}/#{match[:repository]}/pulls/#{match[:number]}"
       end
 
       private
 
       def match
-        @url.match(/github\.com\/(?<owner>[^\/]+)\/(?<repo>[^\/]+)\/pull\/(?<number>[^\/]+)/)
+        @match ||= @url.match(%r{github\.com/(?<owner>[^/]+)/(?<repository>[^/]+)/pull/(?<number>[^/]+)})
       end
 
       def data
-        {
-          url: @url,
-          author: raw["user"]["login"],
-          gravatar: raw["user"]["avatar_url"],
-          title: raw["title"],
-          repo: raw["base"]["repo"]["full_name"],
-          time_date: raw["created_at"],
-          commits: raw["commits"],
-          additions: raw["additions"],
-          deletions: raw["deletions"],
-          changed_files: raw["changed_files"],
-          description: raw["body"]
-        }
+        result = raw.clone
+        result['link'] = link
+        result['created_at'] = Time.parse(result['created_at']).strftime("%I:%M%p - %d %b %y")
+        result
       end
     end
   end

data/lib/onebox/engine/google_play_app_onebox.rb

@@ -0,0 +1,30 @@
+module Onebox
+  module Engine
+    class GooglePlayAppOnebox
+      include Engine
+      include LayoutSupport
+      include HTML
+
+      matches do
+        http
+        with("play.")
+        domain("google")
+        tld("com")
+        with("/store/apps/")
+      end
+
+      private
+
+      def data
+        {
+          link: link,
+          title: raw.css(".document-title div").inner_text,
+          developer: raw.css(".document-subtitle.primary").inner_text,
+          image: raw.css(".cover-image").first["src"],
+          description: raw.css(".text-body div").inner_text,
+          price: raw.css(".price.buy meta[itemprop=price]").first["content"]
+        }
+      end
+    end
+  end
+end

data/lib/onebox/engine/html.rb

@@ -3,8 +3,17 @@ module Onebox
     module HTML
       private
 
+      # Overwrite for any custom headers
+      def http_params
+        {}
+      end
+
       def raw
-        @raw ||= Nokogiri::HTML(open(url, read_timeout: timeout))
+        @raw ||= Nokogiri::HTML(open(url, {read_timeout: timeout}.merge(http_params)).read)
+      end
+
+      def html?
+        raw.respond_to(:css)
       end
     end
   end

data/lib/onebox/engine/image_onebox.rb

@@ -0,0 +1,13 @@
+module Onebox
+  module Engine
+    class ImageOnebox
+      include Engine
+
+      matches_regexp /^(https?:)?\/\/.+\.(png|jpg|jpeg|gif|bmp|tif|tiff)$/i
+
+      def to_html
+        "<a href='#{@url}' target='_blank'><img src='#{@url}'></a>"
+      end
+    end
+  end
+end

data/lib/onebox/engine/stack_exchange_onebox.rb

@@ -2,23 +2,37 @@ module Onebox
   module Engine
     class StackExchangeOnebox
       include Engine
-      include HTML
+      include LayoutSupport
+      include JSON
 
-      matches do
-        http
-        domain("stackoverflow")
-        tld("com")
-        with("/questions/")
+      def self.domains
+        %w(stackexchange stackoverflow superuser serverfault askubuntu)
       end
 
+      matches_regexp /^http:\/\/(?:(?:(?<subsubdomain>\w*)\.)?(?<subdomain>\w*)\.)?(?<domain>#{domains.join('|')})\.com\/(?:questions|q)\/(?<question>\d*)/
+
       private
 
+      def match
+        @match ||= @url.match(@@matcher)
+      end
+
+      def url
+        domain = match[:subsubdomain] || match[:subdomain] || match[:domain]
+        "http://api.stackexchange.com/2.1/questions/#{match[:question]}?site=#{domain}"
+      end
+
       def data
-        {
-          url: @url,
-          title: raw.css(".question-hyperlink").inner_text,
-          question: raw.css(".question .post-text p").first.inner_text
-        }
+        return @data if @data
+
+        result = raw['items'][0]
+        if result
+          result['creation_date'] =
+            Time.at(result['creation_date'].to_i).strftime("%I:%M%p - %d %b %y")
+
+          result['tags'] = result['tags'].take(4).join(', ')
+        end
+        @data = result
       end
     end
   end

data/lib/onebox/engine/standard_embed.rb

@@ -0,0 +1,48 @@
+module Onebox
+  module Engine
+    module StandardEmbed
+
+      def raw
+        return @raw if @raw
+        response = fetch_response(url)
+
+        html_doc = Nokogiri::HTML(response.body)
+
+        # Determine if we should use OEmbed or OpenGraph
+        oembed_alternate = html_doc.at("//link[@type='application/json+oembed']") || html_doc.at("//link[@type='text/json+oembed']")
+        if oembed_alternate
+          # If the oembed request fails, we can still try the opengraph below.
+          begin
+            @raw = Onebox::Helpers.symbolize_keys(::MultiJson.load(fetch_response(oembed_alternate['href']).body))
+          rescue Errno::ECONNREFUSED, Net::HTTPError, MultiJson::LoadError
+            @raw = nil
+          end
+        end
+
+        open_graph = OpenGraph.new(response.body, false)
+        if @raw
+          @raw[:image] = open_graph.images.first if @raw[:image].nil? && open_graph && open_graph.images
+
+          return @raw
+        end
+
+        @raw = open_graph
+      end
+
+      private
+
+      def fetch_response(location, limit = 3)
+        raise Net::HTTPError.new('HTTP redirect too deep', location) if limit == 0
+
+        response = Net::HTTP.get_response(URI(location))
+        case response
+        when Net::HTTPSuccess     then response
+        when Net::HTTPRedirection then fetch_response(response['location'], limit - 1)
+        else
+          response.error!
+        end
+      end
+
+    end
+  end
+end

data/lib/onebox/engine/twitter_status_onebox.rb

@@ -0,0 +1,100 @@
+module Onebox
+  module Engine
+    class TwitterStatusOnebox
+      include Engine
+      include LayoutSupport
+      include HTML
+
+      matches do
+        http
+        maybe("www.")
+        domain("twitter")
+        tld("com")
+        anything
+        has("/status/")
+      end
+
+      private
+
+      def match
+        @match ||= @url.match(%r{twitter\.com/.+?/status/(?<id>\d+)})
+      end
+
+      def client
+        Onebox.options.twitter_client
+      end
+
+      def raw
+        if client
+          @raw ||= OpenStruct.new(client.status(match[:id]).to_hash)
+        else
+          super
+        end
+      end
+
+      def access(*keys)
+        keys.reduce(raw) do |memo, key|
+          memo[key] || memo[key.to_s]
+        end
+      end
+
+      def tweet
+        if raw.html?
+          raw.css(".tweet-text")[0].inner_text
+        else
+          access(:text)
+        end
+      end
+
+      def timestamp
+        if raw.html?
+          raw.css(".metadata span").inner_text
+        else
+          access(:created_at)
+        end
+      end
+
+      def title
+        if raw.html?
+          raw.css(".stream-item-header .username").inner_text
+        else
+          access(:user, :screen_name)
+        end
+      end
+
+      def avatar
+        if raw.html?
+          raw.css(".avatar")[2]["src"]
+        else
+          access(:user, :profile_image_url)
+        end
+      end
+
+      def favorites
+        if raw.html?
+          raw.css(".stats li .request-favorited-popup").inner_text
+        else
+          access(:favorite_count)
+        end
+      end
+
+      def retweets
+        if raw.html?
+          raw.css(".stats li .request-retweeted-popup").inner_text
+        else
+          access(:retweet_count)
+        end
+      end
+
+      def data
+        { link: link,
+          tweet: tweet,
+          timestamp: timestamp,
+          title: title,
+          avatar: avatar,
+          favorites: favorites,
+          retweets: retweets }
+      end
+    end
+  end
+end

data/lib/onebox/engine/whitelisted_generic_onebox.rb

@@ -0,0 +1,223 @@
+module Onebox
+  module Engine
+    class WhitelistedGenericOnebox
+      include Engine
+      include StandardEmbed
+      include LayoutSupport
+
+      def self.whitelist=(list)
+        @whitelist = list
+      end
+
+      def self.whitelist
+        @whitelist ||= default_whitelist.dup
+      end
+
+      def self.default_whitelist
+        %w(500px.com
+          about.com
+          answers.com
+          ask.com
+          battle.net
+          bbc.co.uk
+          bbs.boingboing.net
+          bestbuy.ca
+          bestbuy.com
+          blip.tv
+          bloomberg.com
+          businessinsider.com
+          clikthrough.com
+          cnet.com
+          cnn.com
+          collegehumor.com
+          coursera.org
+          cracked.com
+          dailymail.co.uk
+          dailymotion.com
+          deadline.com
+          dell.com
+          digg.com
+          dotsub.com
+          ebay.ca
+          ebay.co.uk
+          ebay.com
+          ehow.com
+          espn.go.com
+          etsy.com
+          findery.com
+          flickr.com
+          forbes.com
+          foxnews.com
+          funnyordie.com
+          groupon.com
+          howtogeek.com
+          huffingtonpost.com
+          huffingtonpost.ca
+          hulu.com
+          ign.com
+          ikea.com
+          imgur.com
+          indiatimes.com
+          instagr.am
+          instagram.com
+          itunes.apple.com
+          khanacademy.org
+          kickstarter.com
+          kinomap.com
+          mashable.com
+          mlb.com
+          myspace.com
+          nba.com
+          npr.org
+          photobucket.com
+          pinterest.com
+          reference.com
+          revision3.com
+          rottentomatoes.com
+          samsung.com
+          screenr.com
+          scribd.com
+          slideshare.net
+          soundcloud.com
+          sourceforge.net
+          spotify.com
+          squidoo.com
+          techcrunch.com
+          ted.com
+          thefreedictionary.com
+          theglobeandmail.com
+          theonion.com
+          thestar.com
+          thesun.co.uk
+          thinkgeek.com
+          tmz.com
+          torontosun.com
+          tumblr.com
+          twitpic.com
+          usatoday.com
+          vimeo.com
+          walmart.com
+          washingtonpost.com
+          wikia.com
+          wikihow.com
+          wired.com
+          wistia.com
+          wi.st
+          wonderhowto.com
+          wsj.com
+          zappos.com
+          zillow.com)
+      end
+
+      # A re-written URL coverts https:// -> // - it is useful on HTTPS sites that embed
+      # youtube for example
+      def self.rewrites
+        @rewrites ||= default_rewrites.dup
+      end
+
+      def self.rewrites=(new_list)
+        @rewrites = new_list
+      end
+
+      def self.default_rewrites
+        %w()
+      end
+
+      def self.host_matches(uri, list)
+        !!list.find {|h| %r((^|\.)#{Regexp.escape(h)}$).match(uri.host) }
+      end
+
+      def self.===(other)
+        if other.kind_of?(URI)
+          return WhitelistedGenericOnebox.host_matches(other, WhitelistedGenericOnebox.whitelist)
+        else
+          super
+        end
+      end
+
+      # Generates the HTML for the embedded content
+      def photo_type?
+        data[:type] =~ /photo/ || data[:type] =~ /image/
+      end
+
+      def rewrite_agnostic(html)
+        return html unless html
+        uri = URI(@url)
+        if WhitelistedGenericOnebox.host_matches(uri, WhitelistedGenericOnebox.rewrites)
+          html.gsub!(/https?:\/\//, '//')
+        end
+        html
+      end
+
+      def generic_html
+        return data[:html] if data[:html]
+        return html_for_video(data[:video]) if data[:video]
+        return image_html if photo_type?
+        return nil unless data[:title]
+        layout.to_html
+      end
+
+      def to_html
+        rewrite_agnostic(generic_html)
+      end
+
+      def placeholder_html
+        result = nil
+        result = image_html if data[:html] || data[:video] || photo_type?
+        result || to_html
+      end
+
+      def data
+        return raw if raw.is_a?(Hash)
+
+        data_hash = { link: link, title: raw.title, description: raw.description }
+        data_hash[:image] = raw.images.first if raw.images && raw.images.first
+        data_hash[:type] = raw.type if raw.type
+
+        if raw.metadata && raw.metadata[:video] && raw.metadata[:video].first
+          data_hash[:video] = raw.metadata[:video].first 
+        end
+
+        data_hash
+      end
+
+      private
+
+      def image_html
+        return @image_html if @image_html
+
+        return @image_html = "<img src=\"#{data[:image]}\">" if data[:image]
+
+        if data[:thumbnail_url]
+          @image_html = "<img src=\"#{data[:thumbnail_url]}\""
+          @image_html << " width=\"#{data[:thumbnail_width]}\"" if data[:thumbnail_width]
+          @image_html << " height=\"#{data[:thumbnail_height]}\"" if data[:thumbnail_height]
+          @image_html << ">"
+        end
+
+        @image_html
+      end
+
+      def html_for_video(video)
+        video_url = video[:_value]
+
+        if video_url
+          html = "<iframe src=\"#{video_url}\" frameborder=\"0\" title=\"#{data[:title]}\""
+
+          append_attribute(:width, html, video)
+          append_attribute(:height, html, video)
+
+          html << "></iframe>"
+          return html
+        end
+      end
+
+      def append_attribute(attribute, html, video)
+        if video[attribute] && video[attribute].first
+          val = video[attribute].first[:_value]
+          html << " #{attribute.to_s}=\"#{val}\""
+        end
+      end
+    end
+  end
+end