mida 0.3.2 → 0.3.3

data/CHANGELOG.rdoc

@@ -1,3 +1,11 @@
+== 0.3.3 (31st July 2011)
+* Removed validation from +new+ for +Item+ and +Document+
+* Created <tt>vocabularies/</tt> to put the vocabularies in
+* Added +URL+ and +Enumeration+ <tt>DataType</tt>s
+* Converted <tt>DataTypes</tt>s to classes derived from +Generic+
+* Now supports circular dependencies in vocabularies
+* Added all the <tt>Schema.org</tt> vocabularies
+
 == 0.3.2 (9th July 2011)
 * Ensure <tt>bin/mida</tt> only searches if type regexp passed
 * Allow vocabularies to be included into other vocabularies

data/{LICENSE.rdoc → LICENCE.rdoc}

@@ -1,4 +1,4 @@
-= The MIT License
+= The MIT Licence
 
 Copyright (c) 2011 Lawrence Woodman
 

data/README.rdoc

@@ -131,6 +131,6 @@ If you find a bug or want to make a feature request, please report it at the
 Mida project's {issues tracker}[https://github.com/LawrenceWoodman/mida/issues]
 on github.
 
-== License
+== Licence
 Copyright (c) 2011 Lawrence Woodman.
-This software is licensed under the MIT License.  Please see the file, LICENSE.rdoc, for details.
+This software is licensed under the MIT Licence.  Please see the file, LICENCE.rdoc, for details.

data/Rakefile

@@ -5,8 +5,8 @@ require 'rubygems/package_task'
 spec = Gem::Specification.new do |s|
   s.name        = "mida"
   s.summary     = "A Microdata parser/extractor library"
-  s.description = "A Microdata parser and extractor library, based on the latest published version of the Microdata Specification, dated 5th April 2011."
-  s.version     = "0.3.2"
+  s.description = "A Microdata parser and extractor library which includes support for the schema.org vocabularies"
+  s.version     = "0.3.3"
   s.author      = "Lawrence Woodman"
   s.email       = "lwoodman@vlifesystems.com"
   s.homepage    = %q{http://lawrencewoodman.github.com/mida/}
@@ -14,9 +14,10 @@ spec = Gem::Specification.new do |s|
   s.required_ruby_version = '>=1.9'
   s.files       = Dir['lib/**/*.rb'] + Dir['spec/**/*.rb'] + Dir['*.rdoc'] + Dir['Rakefile']
   s.executables = ['mida']
-  s.extra_rdoc_files = ['README.rdoc', 'LICENSE.rdoc', 'CHANGELOG.rdoc']
+  s.extra_rdoc_files = ['README.rdoc', 'LICENCE.rdoc', 'CHANGELOG.rdoc']
   s.rdoc_options << '--main' << 'README.rdoc'
   s.add_dependency('nokogiri', '>= 1.5')
+  s.add_dependency('blankslate')
   s.add_development_dependency('rspec', '>= 2.0' )
   s.add_development_dependency('bundler')
 end

data/bin/mida

@@ -32,7 +32,17 @@ rescue LoadError
   raise
 end
 
-options = {sourcename: true, validate: true}
+# Turn off validation by unregistering all the vocabularies apart from
+# GenericVocabulary
+def turnOffValidation
+  Mida::Vocabulary.vocabularies.each do |vocabulary|
+    unless vocabulary == Mida::GenericVocabulary
+      Mida::Vocabulary.unregister(vocabulary)
+    end
+  end
+end
+
+options = {sourcename: true}
 ARGV.options do |option|
   option.banner = banner
   option.on('-c','--count', 'Display the counts of each Microdata Type') do
@@ -47,7 +57,7 @@ ARGV.options do |option|
   end
   option.on('-v','--no-validate',
             "Don't validate the items against known Vocabularies") do
-    options[:validate] = false
+    turnOffValidation
   end
   option.on_tail('-h','--help', 'This help message') {puts option; exit}
 
@@ -84,7 +94,7 @@ def parse_source(source, options)
   url = get_url  
   begin
     open(source) do |f|
-      doc = Mida::Document.new(f, url, options[:validate])
+      doc = Mida::Document.new(f, url)
       items = if options[:type]
         doc.search(options[:type])
       else

data/lib/mida.rb

@@ -6,4 +6,5 @@ module Mida
 
 end
 
-require 'mida/genericvocabulary'
+require 'mida/vocabularies/genericvocabulary'
+require 'mida/vocabularies/schemaorg'

data/lib/mida/datatype.rb

@@ -8,8 +8,10 @@ module Mida
 end
 
 require 'mida/datatype/boolean'
+require 'mida/datatype/enumeration'
 require 'mida/datatype/float'
 require 'mida/datatype/integer'
 require 'mida/datatype/iso8601date'
 require 'mida/datatype/number'
 require 'mida/datatype/text'
+require 'mida/datatype/url'

data/lib/mida/datatype/boolean.rb

@@ -1,18 +1,30 @@
+require 'mida/datatype/generic'
+
 module Mida
   module DataType
 
     # Boolean data type
-    module Boolean
+    # Provides access to TrueClass/FalseClass methods
+    class Boolean < Generic
+
+      # Raises +ArgumentError+ if value not valid boolean
+      def initialize(value)
+        @parsedValue = case value.downcase
+                       when 'true' then true
+                       when 'false' then false
+                       else raise ::ArgumentError, 'Invalid value'
+                       end
+      end
 
-      # Returns the +value+ as a boolean
-      # or raises ArgumentError if not valid
-      def self.extract(value)
-        case value.downcase
-        when 'true' then true
-        when 'false' then false
-        else raise ArgumentError, 'Invalid value'
-        end
+      def to_s
+        @parsedValue.to_s.capitalize
       end
+
+      def !@
+        !@parsedValue
+      end
+
     end
+
   end
 end

data/lib/mida/datatype/enumeration.rb

@@ -0,0 +1,28 @@
+require 'mida/datatype/generic'
+require 'mida/datatype/url'
+require 'uri'
+
+module Mida
+  module DataType
+
+    # Enumeration data type
+    # Provides access to underly DataType methods
+    # Subclasses should implement VALID_VALUES as an array of the form:
+    # [[DataType, Regexp], [DataType, Regexp]]
+    class Enumeration < Generic
+
+      Enumeration.reveal(:class)
+
+      # Raises +ArgumentError+ if value not valid value
+      def initialize(value)
+        value_is_valid = self.class::VALID_VALUES.any? do |valid_value|
+          @parsedValue = valid_value[0].parse(value)
+          @parsedValue.to_s =~ valid_value[1]
+        end
+        raise ArgumentError unless value_is_valid 
+      end
+
+    end
+
+  end
+end

data/lib/mida/datatype/float.rb

@@ -1,15 +1,18 @@
+require 'mida/datatype/generic'
+
 module Mida
   module DataType
 
     # Float data type
-    module Float
+    # Provides access to Float methods
+    class Float < Generic
 
-      # Returns the +value+ as a floating point number
-      # Relies on +Float+ to raise +ArgumentError+ if not valid
-      def self.extract(value)
-        Float(value)
+      # Raises +ArgumentError+ if value not valid
+      def initialize(value)
+        @parsedValue = ::Kernel.Float(value)
       end
 
     end
+
   end
 end

data/lib/mida/datatype/generic.rb

@@ -0,0 +1,33 @@
+require 'blankslate'
+
+module Mida
+  module DataType
+
+    # The base DataType Parser
+    class Generic < BlankSlate
+      
+      # Convenience method, same as +new+
+      def self.parse(value)
+        self.new(value)
+      end
+
+      def method_missing(name, *args, &block)
+        @parsedValue.send(name, *args, &block)
+      end
+
+      def to_s
+        @parsedValue.to_s
+      end
+
+      def to_yaml(options={})
+        to_s.to_yaml(options)
+      end
+
+      def ==(other)
+        @parsedValue.to_s.downcase == other.to_s.downcase
+      end
+
+    end
+
+  end
+end

data/lib/mida/datatype/integer.rb

@@ -2,12 +2,12 @@ module Mida
   module DataType
 
     # Integer data type
-    module Integer
+    # Provides access to Integer methods
+    class Integer < Generic
 
-      # Returns the +value+ as an integer
-      # Relies on +Integer+ to raise +ArgumentError+ if not valid
-      def self.extract(value)
-        Integer(value)
+      # Raises +ArgumentError+ if value not valid
+      def initialize(value)
+        @parsedValue = ::Kernel.Integer(value)
       end
 
     end

data/lib/mida/datatype/iso8601date.rb

@@ -1,17 +1,23 @@
+require 'mida/datatype/generic'
 require 'date'
 
 module Mida
   module DataType
 
     # ISO 8601 Date data type
-    module ISO8601Date
+    # Provides access to DateTime methods
+    class ISO8601Date < Generic
 
-      # Returns the +value+ as a +DateTime+ instance
-      # Relies on <tt>DateTime#iso8601</tt> to raise
-      # +ArgumentError+ if not valid
-      def self.extract(value)
-        DateTime.iso8601(value)
+      # Raises +ArgumentError+ if value not valid
+      def initialize(value)
+        @parsedValue = ::DateTime.iso8601(value)
       end
+
+      def to_s
+        @parsedValue.rfc822
+      end
+
     end
+
   end
 end

data/lib/mida/datatype/number.rb

@@ -2,14 +2,15 @@ module Mida
   module DataType
 
     # Number data type
-    module Number
+    # Provides access to Float methods
+    class Number < Generic
 
-      # Returns the +value+ as a number
-      # Relies on +Float+ to raise +ArgumentError+ if not valid
-      def self.extract(value)
-        Float(value)
+      # Raises +ArgumentError+ if value not valid
+      def initialize(value)
+        @parsedValue = ::Kernel.Float(value)
       end
 
     end
+
   end
 end

data/lib/mida/datatype/text.rb

@@ -2,12 +2,14 @@ module Mida
   module DataType
 
     # Text data type
-    module Text
+    class Text < Generic
 
       # Returns the value extracted
-      def self.extract(value)
-        value
+      def initialize(value)
+        @parsedValue = value
       end
+
     end
+
   end
 end

data/lib/mida/datatype/url.rb

@@ -0,0 +1,20 @@
+require 'mida/datatype/generic'
+require 'uri'
+
+module Mida
+  module DataType
+
+    # URL data type
+    # Provides access to URI methods
+    class URL < Generic
+
+      # Raises +ArgumentError+ if value not valid url
+      def initialize(value)
+        raise ::ArgumentError unless value =~ ::URI::DEFAULT_PARSER.regexp[:ABS_URI]
+        @parsedValue = ::URI.parse(value)
+      end
+
+    end
+
+  end
+end

data/lib/mida/document.rb

@@ -15,11 +15,10 @@ module Mida
     # [target] The string containing the html that you want to parse.
     # [page_url] The url of target used for form absolute urls. This must
     #            include the filename, e.g. index.html.
-    # [validate] Whether to validate the items against known vocabularies.
-    def initialize(target, page_url=nil, validate=true)
+    def initialize(target, page_url=nil)
       @doc = Nokogiri(target)
       @page_url = page_url
-      @items = extract_items(validate)
+      @items = extract_items
     end
 
     # Implements method for Enumerable
@@ -46,13 +45,13 @@ module Mida
     end
 
   private
-    def extract_items(validate)
+    def extract_items
       itemscopes = @doc.search('//*[@itemscope and not(@itemprop)]')
       return nil unless itemscopes
 
       itemscopes.collect do |itemscope|
         itemscope = Itemscope.new(itemscope, @page_url)
-        Item.new(itemscope, validate)
+        Item.new(itemscope)
       end
     end
 

data/lib/mida/item.rb

@@ -24,13 +24,12 @@ module Mida
     # its +properties+
     #
     # [itemscope] The itemscope that has been parsed by +Itemscope+
-    # [validate] Whether to validate the item against known vocabularies
-    def initialize(itemscope, validate=true)
+    def initialize(itemscope)
       @type = itemscope.type
       @id = itemscope.id
       @vocabulary = Mida::Vocabulary.find(@type)
       @properties = itemscope.properties
-      validate_properties if validate
+      validate_properties
     end
 
     # Return a Hash representation
@@ -149,7 +148,7 @@ module Mida
     def datatype_extract(valid_types, value)
       valid_types.find do |type|
         begin
-          return type.extract(value) if type.respond_to?(:extract)
+          return type.parse(value) if type.respond_to?(:parse)
         rescue ArgumentError
         end
       end

data/lib/mida/vocabularies/schemaorg.rb

@@ -0,0 +1 @@
+Dir.glob(File.dirname(__FILE__) + '/schemaorg/*.rb') {|file| require file}

data/lib/mida/vocabularies/schemaorg/aboutpage.rb

@@ -0,0 +1,19 @@
+require 'mida/vocabulary'
+
+module Mida
+  module SchemaOrg
+
+    autoload :Thing, 'mida/vocabularies/schemaorg/thing'
+    autoload :CreativeWork, 'mida/vocabularies/schemaorg/creativework'
+    autoload :WebPage, 'mida/vocabularies/schemaorg/webpage'
+
+    # Web page type: About page.
+    class AboutPage < Mida::Vocabulary
+      itemtype %r{http://schema.org/AboutPage}i
+      include_vocabulary Mida::SchemaOrg::Thing
+      include_vocabulary Mida::SchemaOrg::CreativeWork
+      include_vocabulary Mida::SchemaOrg::WebPage
+    end
+
+  end
+end

data/lib/mida/vocabularies/schemaorg/accountingservice.rb

@@ -0,0 +1,21 @@
+require 'mida/vocabulary'
+
+module Mida
+  module SchemaOrg
+
+    autoload :Thing, 'mida/vocabularies/schemaorg/thing'
+    autoload :Place, 'mida/vocabularies/schemaorg/place'
+    autoload :Organization, 'mida/vocabularies/schemaorg/organization'
+    autoload :LocalBusiness, 'mida/vocabularies/schemaorg/localbusiness'
+
+    # Accountancy business.
+    class AccountingService < Mida::Vocabulary
+      itemtype %r{http://schema.org/AccountingService}i
+      include_vocabulary Mida::SchemaOrg::Thing
+      include_vocabulary Mida::SchemaOrg::Place
+      include_vocabulary Mida::SchemaOrg::Organization
+      include_vocabulary Mida::SchemaOrg::LocalBusiness
+    end
+
+  end
+end