Implement OpenGraph library

Author: huyha85

Gemfile

@@ -6,9 +6,9 @@ source "http://rubygems.org"
 # Add dependencies to develop your gem here.
 # Include everything needed to run rake, tests, features, etc.
 group :development do
-  gem "rspec", "~> 2.8.0"
-  gem "rdoc", "~> 3.12"
-  gem "bundler", "~> 1.0.0"
-  gem "jeweler", "~> 1.8.4"
-  gem "rcov", ">= 0"
+  gem "rspec"
+  gem "rdoc"
+  gem "bundler"
+  gem "jeweler"
+  gem "nokogiri"
 end

Gemfile.lock

@@ -0,0 +1,33 @@
+GEM
+  remote: http://rubygems.org/
+  specs:
+    diff-lcs (1.1.3)
+    git (1.2.5)
+    jeweler (1.8.4)
+      bundler (~> 1.0)
+      git (>= 1.2.5)
+      rake
+      rdoc
+    json (1.7.5)
+    nokogiri (1.5.5)
+    rake (0.9.2.2)
+    rdoc (3.12)
+      json (~> 1.4)
+    rspec (2.11.0)
+      rspec-core (~> 2.11.0)
+      rspec-expectations (~> 2.11.0)
+      rspec-mocks (~> 2.11.0)
+    rspec-core (2.11.1)
+    rspec-expectations (2.11.3)
+      diff-lcs (~> 1.1.3)
+    rspec-mocks (2.11.3)
+
+PLATFORMS
+  ruby
+
+DEPENDENCIES
+  bundler
+  jeweler
+  nokogiri
+  rdoc
+  rspec

README.rdoc

@@ -1,9 +1,36 @@
-= opengraph_parser
+= OpengraphParser
 
-Description goes here.
+OpengraphParser is a simple Ruby library for parsing Open Graph protocol information from a web site. Learn more about the protocol at:
+http://ogp.me
+
+== Installation
+  gem install opengraph_parser
+
+or add to Gemfile
+
+  gem "opengraph_parser"
+
+== Usage
+  og = OpenGraph.new("http://ogp.me")
+  og.title # => "Open Graph protocol"
+  og.type # => "website"
+  og.url # => "http://ogp.me/"
+  og.description # => "The Open Graph protocol enables any web page to become a rich object in a social graph."
+  og.images # => ["http://ogp.me/logo.png"]
+
+You can also get other Open Graph metadata as:
+  og.metadata # => {"og:image:type"=>"image/png", "og:image:width"=>"300", "og:image:height"=>"300"}
+
+If you try to parse Open Graph information for a website that doesn’t have any Open Graph metadata, the library will try to find other information in the website as the following rules:
+  <title> for title
+  <meta name="description"> for description
+  <link rel="image_src"> or all <img> tags for images
+
+You can disable this fallback lookup by passing false to init method:
+  og = OpenGraph.new("http://ogp.me", false)
 
 == Contributing to opengraph_parser
- 
+
 * Check out the latest master to make sure the feature hasn't been implemented or the bug hasn't been fixed yet.
 * Check out the issue tracker to make sure someone already hasn't requested it and/or contributed it.
 * Fork the project.
@@ -16,4 +43,3 @@ Description goes here.
 
 Copyright (c) 2012 Huy Ha. See LICENSE.txt for
 further details.
-

Rakefile

@@ -17,11 +17,12 @@ Jeweler::Tasks.new do |gem|
   gem.name = "opengraph_parser"
   gem.homepage = "http://github.com/huyha85/opengraph_parser"
   gem.license = "MIT"
-  gem.summary = %Q{TODO: one-line summary of your gem}
-  gem.description = %Q{TODO: longer description of your gem}
+  gem.summary = %Q{A simple Ruby library for parsing Open Graph Protocol information from a website.}
+  gem.description = %Q{A simple Ruby library for parsing Open Graph Protocol information from a website. It also includes a fallback solution when the website has no Open Graph information.}
   gem.email = "[email protected]"
-  gem.authors = ["Huy Ha"]
+  gem.authors = ["Huy Ha", "Duc Trinh"]
   # dependencies defined in Gemfile
+  gem.files = Dir.glob('lib/**/*.rb')
 end
 Jeweler::RubygemsDotOrgTasks.new
 

VERSION

@@ -0,0 +1 @@
+0.1.0

lib/open_graph.rb

@@ -0,0 +1,85 @@
+require 'nokogiri'
+require 'redirect_follower'
+
+class OpenGraph
+  attr_accessor :src, :url, :type, :title, :description, :images, :metadata, :response
+
+  def initialize(src, fallback = true)
+    @src = src
+    @images = []
+    @metadata = {}
+    parse_opengraph
+    load_fallback if fallback
+    check_images_path
+  end
+
+  def parse_opengraph
+    begin
+      @response = RedirectFollower.new(@src).resolve
+    rescue
+      @title = @url = @src
+      return
+    end
+
+    if @response && @response.body
+      attrs_list = %w(title url type description)
+      doc = Nokogiri.parse(@response.body)
+      doc.css('meta').each do |m|
+        if m.attribute('property') && m.attribute('property').to_s.match(/^og:(.+)$/i)
+          m_content = m.attribute('content').to_s.strip
+          case metadata_name = m.attribute('property').to_s.gsub("og:", "")
+            when *attrs_list
+              self.instance_variable_set("@#{metadata_name}", m_content) unless m_content.empty?
+            when "image"
+              add_image(m_content)
+            else
+              @metadata[m.attribute('property').to_s] = m_content
+          end
+        end
+      end
+    end
+  end
+
+  def load_fallback
+    if @response && @response.body
+      doc = Nokogiri.parse(@response.body)
+
+      if @title.to_s.empty? && doc.xpath("//head/title").size > 0
+        @title = doc.xpath("//head/title").first.text.to_s.strip
+      end
+
+      @url = @src if @url.to_s.empty?
+
+      if @description.to_s.empty? && description_meta = doc.xpath("//head/meta[@name='description']").first
+        @description = description_meta.attribute("content").to_s.strip
+      end
+
+      fetch_images(doc, "//head/link[@rel='image_src']", "href") if @images.empty?
+      fetch_images(doc, "//img", "src") if @images.empty?
+    end
+  end
+
+  def check_images_path
+    uri = URI.parse(URI.escape(@src))
+    imgs = @images.dup
+    @images = []
+    imgs.each do |img|
+      if URI.parse(URI.escape(img)).host.nil?
+        add_image("#{uri.scheme}://#{uri.host}:#{uri.port}#{img}")
+      else
+        add_image(img)
+      end
+    end
+  end
+
+  private
+  def add_image(image_url)
+    @images << image_url unless @images.include?(image_url) || image_url.to_s.empty?
+  end
+
+  def fetch_images(doc, xpath_str, attr)
+    doc.xpath(xpath_str).each do |link|
+      add_image(link.attribute(attr).to_s.strip)
+    end
+  end
+end

lib/opengraph_parser.rb

@@ -0,0 +1 @@
+require 'open_graph'

lib/redirect_follower.rb

@@ -0,0 +1,34 @@
+require 'net/http'
+
+class RedirectFollower
+  class TooManyRedirects < StandardError; end
+
+  attr_accessor :url, :body, :redirect_limit, :response
+
+  def initialize(url, limit = 5)
+    @url, @redirect_limit = url, limit
+  end
+
+  def resolve
+    raise TooManyRedirects if redirect_limit < 0
+
+    self.response = Net::HTTP.get_response(URI.parse(URI.escape(url)))
+
+    if response.kind_of?(Net::HTTPRedirection)
+      self.url = redirect_url
+      self.redirect_limit -= 1
+      resolve
+    end
+
+    self.body = response.body
+    self
+  end
+
+  def redirect_url
+    if response['location'].nil?
+      response.body.match(/<a href=\"([^>]+)\">/i)[1]
+    else
+      response['location']
+    end
+  end
+end

spec/lib/open_graph_spec.rb

@@ -0,0 +1,61 @@
+require File.expand_path(File.dirname(__FILE__) + '/../spec_helper')
+
+describe OpenGraph do
+  describe "#initialize" do
+    context "with invalid src" do
+      it "should set title and url the same as src" do
+        og = OpenGraph.new("invalid")
+        og.src.should == "invalid"
+        og.title.should == "invalid"
+        og.url.should == "invalid"
+      end
+    end
+
+    context "with no fallback" do
+      it "should get values from opengraph metadata" do
+        response = double(body: File.open("#{File.dirname(__FILE__)}/../view/opengraph.html", 'r') { |f| f.read })
+        RedirectFollower.stub(:new) { double(resolve: response) }
+
+        og = OpenGraph.new("http://test.host", false)
+        og.src.should == "http://test.host"
+        og.title.should == "OpenGraph Title"
+        og.type.should == "article"
+        og.url.should == "http://test.host"
+        og.description.should == "My OpenGraph sample site for Rspec"
+        og.images.should == ["http://test.host/images/rock1.jpg", "http://test.host/images/rock2.jpg"]
+      end
+    end
+
+    context "with fallback" do
+      context "when website has opengraph metadata" do
+        it "should get values from opengraph metadata" do
+          response = double(body: File.open("#{File.dirname(__FILE__)}/../view/opengraph.html", 'r') { |f| f.read })
+          RedirectFollower.stub(:new) { double(resolve: response) }
+
+          og = OpenGraph.new("http://test.host")
+          og.src.should == "http://test.host"
+          og.title.should == "OpenGraph Title"
+          og.type.should == "article"
+          og.url.should == "http://test.host"
+          og.description.should == "My OpenGraph sample site for Rspec"
+          og.images.should == ["http://test.host/images/rock1.jpg", "http://test.host/images/rock2.jpg"]
+        end
+      end
+
+      context "when website has no opengraph metadata" do
+        it "should lookup for other data from website" do
+          response = double(body: File.open("#{File.dirname(__FILE__)}/../view/opengraph_no_metadata.html", 'r') { |f| f.read })
+          RedirectFollower.stub(:new) { double(resolve: response) }
+
+          og = OpenGraph.new("http://test.host")
+          og.src.should == "http://test.host"
+          og.title.should == "OpenGraph Title Fallback"
+          og.type.should be_nil
+          og.url.should == "http://test.host"
+          og.description.should == "Short Description Fallback"
+          og.images.should == ["http://test.host:80/images/wall1.jpg", "http://test.host:80/images/wall2.jpg"]
+        end
+      end
+    end
+  end
+end

spec/lib/redirect_follower_spec.rb

@@ -0,0 +1,43 @@
+require File.expand_path(File.dirname(__FILE__) + '/../spec_helper')
+
+describe RedirectFollower do
+  describe "#resolve" do
+    let(:url) { "http://test.host" }
+    let(:mock_res) { double(body: "Body is here.") }
+    let(:mock_redirect) {
+      m = double(body: %Q{<body><a href="http://new.test.host"></a></body>}, kind_of?: Net::HTTPRedirection)
+      m.stub(:[]).and_return(nil)
+      m
+    }
+
+    context "with no redirection" do
+      it "should return the response" do
+        Net::HTTP.should_receive(:get_response).and_return(mock_res)
+
+        res = RedirectFollower.new(url).resolve
+        res.body.should == "Body is here."
+        res.redirect_limit.should == 5
+      end
+    end
+
+    context "with redirection" do
+      it "should follow the link in redirection" do
+        Net::HTTP.should_receive(:get_response).with(URI.parse(URI.escape(url))).and_return(mock_redirect)
+        Net::HTTP.should_receive(:get_response).with(URI.parse(URI.escape("http://new.test.host"))).and_return(mock_res)
+
+        res = RedirectFollower.new(url).resolve
+        res.body.should == "Body is here."
+        res.redirect_limit.should == 4
+      end
+    end
+
+    context "with unlimited redirection" do
+      it "should raise TooManyRedirects error" do
+        Net::HTTP.stub(:get_response).and_return(mock_redirect)
+        lambda {
+          RedirectFollower.new(url).resolve
+        }.should raise_error(RedirectFollower::TooManyRedirects)
+      end
+    end
+  end
+end