embedify 0.2.0 → 0.3.0

data/Gemfile

@@ -13,7 +13,5 @@ group :development do
   gem 'rspec'
 end
 
-gem 'nokogiri'
-gem 'httparty'
-gem 'opengraph', git: 'git://github.com/parasquid/opengraph.git'
-#gem 'opengraph', path: '../opengraph'
+gem 'nokogiri', '~> 1.5.0'
+gem 'faraday'

data/Gemfile.lock

@@ -1,36 +1,25 @@
-GIT
-  remote: git://github.com/parasquid/opengraph.git
-  revision: f750abbdb50d36f32e2b390cfec58a76d23ede49
-  specs:
-    opengraph (0.0.4)
-      hashie
-      nokogiri (~> 1.4.0)
-      rest-client (~> 1.6.0)
-
 GEM
   remote: http://rubygems.org/
   specs:
+    addressable (2.2.7)
     diff-lcs (1.1.3)
+    faraday (0.7.6)
+      addressable (~> 2.2)
+      multipart-post (~> 1.1)
+      rack (~> 1.1)
     git (1.2.5)
-    hashie (1.2.0)
-    httparty (0.8.1)
-      multi_json
-      multi_xml
     jeweler (1.8.3)
       bundler (~> 1.0)
       git (>= 1.2.5)
       rake
       rdoc
     json (1.6.5)
-    mime-types (1.17.2)
-    multi_json (1.1.0)
-    multi_xml (0.4.1)
-    nokogiri (1.4.7)
+    multipart-post (1.1.5)
+    nokogiri (1.5.0)
+    rack (1.4.1)
     rake (0.9.2.2)
     rdoc (3.12)
       json (~> 1.4)
-    rest-client (1.6.7)
-      mime-types (>= 1.16)
     rspec (2.8.0)
       rspec-core (~> 2.8.0)
       rspec-expectations (~> 2.8.0)
@@ -46,10 +35,9 @@ PLATFORMS
 
 DEPENDENCIES
   bundler (~> 1.0.0)
-  httparty
+  faraday
   jeweler (~> 1.8.3)
-  nokogiri
-  opengraph!
+  nokogiri (~> 1.5.0)
   rdoc (~> 3.12)
   rspec
   shoulda

data/LICENSE.txt

@@ -163,3 +163,5 @@ whether future versions of the GNU Lesser General Public License shall
 apply, that proxy's public statement of acceptance of any version is
 permanent authorization for you to choose that version for the
 Library.
+
+Original opengraph module Copyright (c) 2009 Intridea, Inc. and Michael Bleigh

data/VERSION

@@ -1 +1 @@
-0.2.0
+0.3.0

data/embedify.gemspec

@@ -5,11 +5,11 @@
 
 Gem::Specification.new do |s|
   s.name = "embedify"
-  s.version = "0.2.0"
+  s.version = "0.3.0"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.authors = ["Tristan Gomez"]
-  s.date = "2012-02-20"
+  s.date = "2012-03-09"
   s.description = "Embedify is a ruby gem that parses and returns an\n    embeddable version of a page, similar to how Facebook fetches and retrieves\n    page data when posting a link."
   s.email = "tristan.gomez+embedify@gmail.com"
   s.extra_rdoc_files = [
@@ -26,6 +26,12 @@ Gem::Specification.new do |s|
     "Rakefile",
     "VERSION",
     "embedify.gemspec",
+    "examples/embedify_server/Gemfile",
+    "examples/embedify_server/Gemfile.lock",
+    "examples/embedify_server/Rakefile",
+    "examples/embedify_server/config.ru",
+    "examples/embedify_server/lib/embedify_server.rb",
+    "examples/embedify_server/lib/views/index.erb",
     "lib/embedify.rb",
     "test/helper.rb",
     "test/test_embedify.rb"
@@ -40,18 +46,16 @@ Gem::Specification.new do |s|
     s.specification_version = 3
 
     if Gem::Version.new(Gem::VERSION) >= Gem::Version.new('1.2.0') then
-      s.add_runtime_dependency(%q<nokogiri>, [">= 0"])
-      s.add_runtime_dependency(%q<httparty>, [">= 0"])
-      s.add_runtime_dependency(%q<opengraph>, [">= 0"])
+      s.add_runtime_dependency(%q<nokogiri>, ["~> 1.5.0"])
+      s.add_runtime_dependency(%q<faraday>, [">= 0"])
       s.add_development_dependency(%q<shoulda>, [">= 0"])
       s.add_development_dependency(%q<rdoc>, ["~> 3.12"])
       s.add_development_dependency(%q<bundler>, ["~> 1.0.0"])
       s.add_development_dependency(%q<jeweler>, ["~> 1.8.3"])
       s.add_development_dependency(%q<rspec>, [">= 0"])
     else
-      s.add_dependency(%q<nokogiri>, [">= 0"])
-      s.add_dependency(%q<httparty>, [">= 0"])
-      s.add_dependency(%q<opengraph>, [">= 0"])
+      s.add_dependency(%q<nokogiri>, ["~> 1.5.0"])
+      s.add_dependency(%q<faraday>, [">= 0"])
       s.add_dependency(%q<shoulda>, [">= 0"])
       s.add_dependency(%q<rdoc>, ["~> 3.12"])
       s.add_dependency(%q<bundler>, ["~> 1.0.0"])
@@ -59,9 +63,8 @@ Gem::Specification.new do |s|
       s.add_dependency(%q<rspec>, [">= 0"])
     end
   else
-    s.add_dependency(%q<nokogiri>, [">= 0"])
-    s.add_dependency(%q<httparty>, [">= 0"])
-    s.add_dependency(%q<opengraph>, [">= 0"])
+    s.add_dependency(%q<nokogiri>, ["~> 1.5.0"])
+    s.add_dependency(%q<faraday>, [">= 0"])
     s.add_dependency(%q<shoulda>, [">= 0"])
     s.add_dependency(%q<rdoc>, ["~> 3.12"])
     s.add_dependency(%q<bundler>, ["~> 1.0.0"])

data/examples/embedify_server/Gemfile

@@ -0,0 +1,7 @@
+source "http://rubygems.org"
+
+gem 'sinatra'
+gem 'embedify', git: 'git://github.com/parasquid/embedify.git'
+gem 'shotgun'
+gem 'thin'
+gem 'sinatra-jsonp', :require => 'sinatra/jsonp'

data/examples/embedify_server/Gemfile.lock

@@ -0,0 +1,61 @@
+GIT
+  remote: git://github.com/parasquid/embedify.git
+  revision: 419b8f4d84975bfcb69e7b3804ab8135e8a585e7
+  specs:
+    embedify (0.2.0)
+      httparty
+      nokogiri
+      opengraph
+
+GIT
+  remote: git://github.com/parasquid/opengraph.git
+  revision: dbcfc27a40be6b4681bfd55c31b767ecf685ff96
+  specs:
+    opengraph (0.0.4)
+      hashie
+      httparty
+      httparty (~> 0.8.1)
+      nokogiri
+      nokogiri (~> 1.5.0)
+
+GEM
+  remote: http://rubygems.org/
+  specs:
+    daemons (1.1.8)
+    eventmachine (0.12.10)
+    hashie (1.2.0)
+    httparty (0.8.1)
+      multi_json
+      multi_xml
+    json_pure (1.6.5)
+    multi_json (1.1.0)
+    multi_xml (0.4.1)
+    nokogiri (1.5.0)
+    rack (1.4.1)
+    rack-protection (1.2.0)
+      rack
+    shotgun (0.9)
+      rack (>= 1.0)
+    sinatra (1.3.2)
+      rack (~> 1.3, >= 1.3.6)
+      rack-protection (~> 1.2)
+      tilt (~> 1.3, >= 1.3.3)
+    sinatra-jsonp (0.3.2)
+      json_pure (~> 1.4)
+      sinatra (~> 1.0)
+    thin (1.3.1)
+      daemons (>= 1.0.9)
+      eventmachine (>= 0.12.6)
+      rack (>= 1.0.0)
+    tilt (1.3.3)
+
+PLATFORMS
+  ruby
+
+DEPENDENCIES
+  embedify!
+  opengraph!
+  shotgun
+  sinatra
+  sinatra-jsonp
+  thin

data/examples/embedify_server/Rakefile

@@ -0,0 +1,45 @@
+# 
+# To change this template, choose Tools | Templates
+# and open the template in the editor.
+ 
+
+require 'rubygems'
+require 'rake'
+require 'rake/clean'
+require 'rake/gempackagetask'
+require 'rake/rdoctask'
+require 'rake/testtask'
+
+spec = Gem::Specification.new do |s|
+  s.name = 'embedify_server'
+  s.version = '0.0.1'
+  s.has_rdoc = true
+  s.extra_rdoc_files = ['README', 'LICENSE']
+  s.summary = 'Your summary here'
+  s.description = s.summary
+  s.author = ''
+  s.email = ''
+  # s.executables = ['your_executable_here']
+  s.files = %w(LICENSE README Rakefile) + Dir.glob("{bin,lib,spec}/**/*")
+  s.require_path = "lib"
+  s.bindir = "bin"
+end
+
+Rake::GemPackageTask.new(spec) do |p|
+  p.gem_spec = spec
+  p.need_tar = true
+  p.need_zip = true
+end
+
+Rake::RDocTask.new do |rdoc|
+  files =['README', 'LICENSE', 'lib/**/*.rb']
+  rdoc.rdoc_files.add(files)
+  rdoc.main = "README" # page to start on
+  rdoc.title = "embedify_server Docs"
+  rdoc.rdoc_dir = 'doc/rdoc' # rdoc output folder
+  rdoc.options << '--line-numbers'
+end
+
+Rake::TestTask.new do |t|
+  t.test_files = FileList['test/**/*.rb']
+end

data/examples/embedify_server/config.ru

@@ -0,0 +1,4 @@
+$LOAD_PATH << File.expand_path(File.dirname(__FILE__) + "/lib")
+
+require 'embedify_server'
+run EmbedifyServer

data/examples/embedify_server/lib/embedify_server.rb

@@ -0,0 +1,43 @@
+require 'sinatra/base'
+require "sinatra/jsonp"
+require 'json'
+require 'cgi'
+
+$LOAD_PATH << File.expand_path('../../lib')
+
+require 'embedify'
+require 'sinatra/jsonp'
+
+class EmbedifyServer < Sinatra::Base
+  helpers Sinatra::Jsonp
+  enable :sessions
+  set :session_secret, "My session secret"
+
+  def self.get_or_post(url,&block)
+    get(url,&block)
+    post(url,&block)
+  end
+
+  get_or_post '/' do
+    puts "processing #{params[:url]}"
+
+    properties = Embedify.fetch(CGI.unescapeHTML(params[:url]))
+
+    # we don't want to return the nokogiri doc
+    properties.delete :nokogiri_parsed_document
+
+    session[:properties] = properties
+
+    puts "returning #{properties.to_json}"
+
+    headers 'Access-Control-Allow-Origin' => '*',
+      'Access-Control-Allow-Methods' => '*',
+      'Access-Control-Allow-Headers' => '*'
+    content_type 'application/json'
+    jsonp properties.to_json, params[:callback] || :callback
+  end
+
+  # start the server if ruby file executed directly
+  run! if app_file == $0
+
+end

data/examples/embedify_server/lib/views/index.erb

@@ -0,0 +1,17 @@
+<html>
+  <head></head>
+  <body>
+    <% if @properties %>
+      <ul>
+        <% @properties.each do |key, value| %>
+          <li><%= key %> : <%= value %></li>
+        <% end %>
+      </ul>
+    <% end %>
+
+    <form action="/" method="post">
+      <input type="url" name="url"/>
+      <input type="submit" />
+    </form>
+  </body>
+</html>

data/lib/embedify.rb

@@ -1,14 +1,67 @@
+$:.unshift(File.expand_path(File.dirname(__FILE__))) unless $:.include?(File.dirname(__FILE__)) || $:.include?(File.expand_path(File.dirname(__FILE__)))
+
 require 'nokogiri'
-require 'httparty'
-require 'opengraph'
+require 'faraday'
+require 'hashie'
 
 module Embedify
 
-  OPENGRAPH_REGEX = /^og:(.+)$/i
+  TYPES = {
+    'activity' => %w(activity sport),
+    'business' => %w(bar company cafe hotel restaurant),
+    'group' => %w(cause sports_league sports_team),
+    'organization' => %w(band government non_profit school university),
+    'person' => %w(actor athlete author director musician politician public_figure),
+    'place' => %w(city country landmark state_province),
+    'product' => %w(album book drink food game movie product song tv_show),
+    'website' => %w(blog website)
+  }
+
+  # The OpenGraph::Object is a Hash with method accessors for
+  # all detected Open Graph attributes.
+  class Object < Hashie::Mash
+    MANDATORY_ATTRIBUTES = %w(title type image url)
+
+    # The object type.
+    def type
+      self['type']
+    end
+
+    # The schema under which this particular object lies. May be any of
+    # the keys of the TYPES constant.
+    def schema
+      Embedify::TYPES.each_pair do |schema, types|
+        return schema if types.include?(self.type)
+      end
+      nil
+    end
+
+    Embedify::TYPES.values.flatten.each do |type|
+      define_method "#{type}?" do
+        self.type == type
+      end
+    end
+
+    Embedify::TYPES.keys.each do |scheme|
+      define_method "#{scheme}?" do
+        self.type == scheme || OpenGraph::TYPES[scheme].include?(self.type)
+      end
+    end
+
+    # If the Open Graph information for this object doesn't contain
+    # the mandatory attributes, this will be <tt>false</tt>.
+    def valid?
+      MANDATORY_ATTRIBUTES.each{|a| return false unless (self[a] && !self[a].empty?)}
+      true
+    end
+  end
 
   def self.fetch(uri, options = {})
-    begin
-      opengraph = OpenGraph.fetch(uri, false)
+    opengraph = begin
+      html = Faraday.get(uri)
+      page = parse(html.body, false)
+      page['url'] = html.request.last_uri.to_s unless page.include? 'url'
+      page
     rescue SocketError => e
       raise e
     end
@@ -16,27 +69,76 @@ module Embedify
     # check if the opengraph object is complete
     unless opengraph.valid?
       # the opengraph object is lacking some mandatory attributes
-      attributes = OpenGraph::Object::MANDATORY_ATTRIBUTES
+      attributes = Embedify::Object::MANDATORY_ATTRIBUTES
       attributes.delete 'url'
       attributes.each do |attribute|
         self.send(attribute.to_sym, opengraph)
       end
     end
+
+    # fill in extra attributes
+    unless opengraph.include? 'description'
+      self.description(opengraph)
+    end
     opengraph
   end
 
   private
 
+  def self.parse(html, strict = true)
+    doc = Nokogiri::HTML.parse(html)
+    page = Embedify::Object.new
+    doc.css('meta').each do |m|
+      if m.attribute('property') && m.attribute('property').to_s.match(/^og:(.+)$/i)
+        page[$1.gsub('-','_')] = m.attribute('content').to_s
+      end
+    end
+    page[:nokogiri_parsed_document] = doc
+    page
+  end
+
   def self.title(document)
-    document['title'] = document[:nokogiri_parsed_document].css('title').first.inner_text
+    document['title'] ||= document[:nokogiri_parsed_document].css('title').first.inner_text
   end
 
   def self.type(document)
-    document['type'] = 'website'
+    document['type'] ||= 'website'
   end
 
   def self.image(document)
+    # TODO: loop through all images and bring back up to 30 good candidates
+    # Good candidates: at least 50x50, max aspect ratio of 3:1, png/jpeg/gif format
     puts "image"
   end
 
-end
+  def self.description(document)
+    meta_tags = document[:nokogiri_parsed_document].css('meta')
+    meta_tags.each do |meta_tag|
+      if meta_tag.attribute('name').to_s.match(/^description$/i)
+        document['description'] = meta_tag.attribute('content').to_s
+        return
+      end
+    end
+
+    # TODO: make the description be the first few words of the first <p></p>
+    document['description'] = ''
+  end
+
+end
+
+
+
+
+require 'nokogiri'
+require 'faraday'
+
+module OpenGraph
+  # Fetch Open Graph data from the specified URI. Makes an
+  # HTTP GET request and returns an OpenGraph::Object if there
+  # is data to be found or <tt>false</tt> if there isn't.
+  #
+  # Pass <tt>false</tt> for the second argument if you want to
+  # see invalid (i.e. missing a required attribute) data.
+
+
+end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: embedify
 version: !ruby/object:Gem::Version
-  version: 0.2.0
+  version: 0.3.0
   prerelease: 
 platform: ruby
 authors:
@@ -9,33 +9,22 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-02-20 00:00:00.000000000 Z
+date: 2012-03-09 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokogiri
-  requirement: &75319620 !ruby/object:Gem::Requirement
+  requirement: &71264200 !ruby/object:Gem::Requirement
     none: false
     requirements:
-    - - ! '>='
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :runtime
-  prerelease: false
-  version_requirements: *75319620
-- !ruby/object:Gem::Dependency
-  name: httparty
-  requirement: &75318830 !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ! '>='
+    - - ~>
       - !ruby/object:Gem::Version
-        version: '0'
+        version: 1.5.0
   type: :runtime
   prerelease: false
-  version_requirements: *75318830
+  version_requirements: *71264200
 - !ruby/object:Gem::Dependency
-  name: opengraph
-  requirement: &75318120 !ruby/object:Gem::Requirement
+  name: faraday
+  requirement: &71262990 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -43,10 +32,10 @@ dependencies:
         version: '0'
   type: :runtime
   prerelease: false
-  version_requirements: *75318120
+  version_requirements: *71262990
 - !ruby/object:Gem::Dependency
   name: shoulda
-  requirement: &75317400 !ruby/object:Gem::Requirement
+  requirement: &71259700 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -54,10 +43,10 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *75317400
+  version_requirements: *71259700
 - !ruby/object:Gem::Dependency
   name: rdoc
-  requirement: &75316510 !ruby/object:Gem::Requirement
+  requirement: &71258770 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -65,10 +54,10 @@ dependencies:
         version: '3.12'
   type: :development
   prerelease: false
-  version_requirements: *75316510
+  version_requirements: *71258770
 - !ruby/object:Gem::Dependency
   name: bundler
-  requirement: &75315750 !ruby/object:Gem::Requirement
+  requirement: &71257430 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -76,10 +65,10 @@ dependencies:
         version: 1.0.0
   type: :development
   prerelease: false
-  version_requirements: *75315750
+  version_requirements: *71257430
 - !ruby/object:Gem::Dependency
   name: jeweler
-  requirement: &75315260 !ruby/object:Gem::Requirement
+  requirement: &71295870 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -87,10 +76,10 @@ dependencies:
         version: 1.8.3
   type: :development
   prerelease: false
-  version_requirements: *75315260
+  version_requirements: *71295870
 - !ruby/object:Gem::Dependency
   name: rspec
-  requirement: &75314780 !ruby/object:Gem::Requirement
+  requirement: &71293630 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -98,7 +87,7 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *75314780
+  version_requirements: *71293630
 description: ! "Embedify is a ruby gem that parses and returns an\n    embeddable
   version of a page, similar to how Facebook fetches and retrieves\n    page data
   when posting a link."
@@ -118,6 +107,12 @@ files:
 - Rakefile
 - VERSION
 - embedify.gemspec
+- examples/embedify_server/Gemfile
+- examples/embedify_server/Gemfile.lock
+- examples/embedify_server/Rakefile
+- examples/embedify_server/config.ru
+- examples/embedify_server/lib/embedify_server.rb
+- examples/embedify_server/lib/views/index.erb
 - lib/embedify.rb
 - test/helper.rb
 - test/test_embedify.rb
@@ -136,7 +131,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: 700225327
+      hash: -523818223
 required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
   requirements: