epub-parser 0.0.5 → 0.0.6

data/README.markdown

@@ -0,0 +1,29 @@
+EPUB Parser
+===========
+
+INSTALL
+-------
+    gem install epub-parser  
+
+USAGE
+-----
+    require 'epub/parser'
+    
+    book = EPUB::Parser.parse 'book.epub', 'working_directory'
+    book.each_page_on_spine do |page|
+      # do somethong...
+    end
+
+See [Wiki][] for more info.
+
+[Wiki]:https://gitorious.org/epub/pages/EpubParser
+
+REQUIREMENTS
+------------
+* UNIX like `unzip` command
+* libxml2 and libxslt for Nokogiri gem
+
+LICENSE
+-------
+This library is distribuetd under the term of the MIT License.
+See MIT-LICENSE file for more info.

data/bin/epubinfo

@@ -0,0 +1,13 @@
+require 'epub/parser'
+require 'tmpdir'
+
+# Use optparse or slop or Thor
+# Validat ARGV
+
+file = ARGV[0]
+dir = Dir.mktmpdir('epub-parser') do |dir|
+
+  book = EPUB::Parser.parse file, dir
+  puts book.package.metadata.to_json
+
+end

data/epub-parser.gemspec

@@ -30,6 +30,8 @@ Gem::Specification.new do |s|
   s.add_development_dependency 'epzip'
   s.add_development_dependency 'epubcheck'
   s.add_development_dependency 'epub_validator'
+  s.add_development_dependency 'zipruby'
+  s.add_development_dependency 'rubyzip'
 
   s.add_runtime_dependency 'enumerabler'
   s.add_runtime_dependency 'nokogiri'

data/lib/epub.rb

@@ -1,6 +1,7 @@
 require 'epub/ocf'
 require 'epub/publication'
 require 'epub/content_document'
+require 'epub/parser'
 
 module EPUB
   modules = [ :ocf, :package, :content_document ]
@@ -12,7 +13,13 @@ module EPUB
     end
   end
 
-  def each_page_by_spine(&blk)
+  def parse(file, dir, options = {})
+    options = options.merge({:book => self})
+    Parser.parse(file, dir, options)
+  end
+  module_function :parse
+
+  def each_page_on_spine(&blk)
     enum = @package.spine.items
     if block_given?
       enum.each &blk
@@ -21,7 +28,7 @@ module EPUB
     end
   end
 
-  def each_page_by_toc(&blk)
+  def each_page_on_toc(&blk)
   end
 
   def each_content(&blk)

data/lib/epub/constants.rb

@@ -17,7 +17,7 @@ module EPUB
     end
 
     module MediaType
-      class NotSupportedError < StandardError; end
+      class UnsupportedError < StandardError; end
 
       IMAGE = %w[
         image/gif

data/lib/epub/parser.rb

@@ -27,7 +27,6 @@ module EPUB
       unzip_cmd = options['unzip-command'] || 'unzip'
       unzip_cmd << " #{@filepath.to_s.shellescape} -d #{@dir.to_s.shellescape}"
       system unzip_cmd
-puts unzip_cmd
     end
 
     def parse

data/lib/epub/parser/ocf.rb

@@ -21,7 +21,7 @@ module EPUB
 
       def parse
         EPUB::OCF::MODULES.each do |m|
-          @ocf.send "#{m}=", send("parse_#{m}")
+          @ocf.__send__ "#{m}=", __send__("parse_#{m}")
         end
         @ocf
       end
@@ -33,7 +33,7 @@ module EPUB
         doc.xpath('/container:container/container:rootfiles/container:rootfile', EPUB::NAMESPACES).each do |elem|
           rootfile = EPUB::OCF::Container::Rootfile.new
           %w[full-path media-type].each do |attr|
-            rootfile.send(attr.gsub(/-/, '_') + '=', elem[attr])
+            rootfile.__send__(attr.gsub(/-/, '_') + '=', elem[attr])
           end
           container.rootfiles << rootfile
         end

data/lib/epub/parser/publication.rb

@@ -19,7 +19,7 @@ module EPUB
       end
 
       def parse
-        # parse_metadata
+        parse_metadata
         parse_manifest
         parse_spine
         parse_guide
@@ -29,7 +29,52 @@ module EPUB
       end
 
       def parse_metadata
-        raise 'still not implemented'
+        metadata = @package.metadata = EPUB::Publication::Package::Metadata.new
+        elem = @doc.xpath('/opf:package/opf:metadata', EPUB::NAMESPACES).first
+
+        metadata.identifiers = elem.xpath('./dc:identifier', EPUB::NAMESPACES).collect do |e|
+          identifier = EPUB::Publication::Package::Metadata::Identifier.new
+          identifier.content = e.content
+          identifier.id = id = e['id']
+          metadata.unique_identifier = identifier if id == @package.unique_identifier
+          identifier
+        end
+
+        metadata.titles = elem.xpath('./dc:title', EPUB::NAMESPACES).collect do |e|
+          title = EPUB::Publication::Package::Metadata::Title.new
+          title.content = e.content
+          %w[ id lang dir ].each do |attr|
+            title.__send__("#{attr}=", e[attr])
+          end
+          title
+        end
+
+        metadata.languages = elem.xpath('./dc:language', EPUB::NAMESPACES).collect do |e|
+          e.content
+        end
+
+        %w[ contributor coverage creator date description format publisher relation source subject type ].each do |dcmes|
+          metadata.__send__ "#{dcmes}s=", elem.xpath("./dc:#{dcmes}", EPUB::NAMESPACES).collect { |e|
+            md = EPUB::Publication::Package::Metadata::DCMES.new
+            md.content = e.content
+            %w[ id lang dir ].each do |attr|
+              md.__send__("#{attr}=", e[attr])
+            end
+            md
+          }
+        end
+        metadata.rights = elem.xpath('./dc:rights', EPUB::NAMESPACES).collect do |e|
+          md = EPUB::Publication::Package::Manifest::DCMES.new
+          md.content = e.content
+          %w[ id lang dir ].each do |attr|
+            md.__send__("#{attr}=", e[attr])
+          end
+          md
+        end
+
+        # handle meta elements
+
+        metadata
       end
 
       def parse_manifest
@@ -38,15 +83,15 @@ module EPUB
         manifest.id = elem['id']
 
         fallback_map = {}
-        elem.xpath('./opf:item', EPUB::NAMESPACES).each do |elm|
+        elem.xpath('./opf:item', EPUB::NAMESPACES).each do |e|
           item = EPUB::Publication::Package::Manifest::Item.new
           %w[ id media-type media-overlay ].each do |attr|
-            item.send "#{attr.gsub(/-/, '_')}=", elm[attr]
+            item.__send__("#{attr.gsub(/-/, '_')}=", e[attr])
           end
-          item.href = elm['href']
-          item.iri = @rootfile.join Addressable::URI.parse(elm['href'])
-          fallback_map[elm['fallback']] = item if elm['fallback']
-          item.properties = elm['properties'] ? elm['properties'].split(' ') : []
+          item.href = e['href']
+          item.iri = @rootfile.join Addressable::URI.parse(e['href'])
+          fallback_map[e['fallback']] = item if e['fallback']
+          item.properties = e['properties'] ? e['properties'].split(' ') : []
           manifest << item
         end
         fallback_map.each_pair do |id, from|
@@ -60,16 +105,16 @@ module EPUB
         spine = @package.spine = EPUB::Publication::Package::Spine.new
         elem = @doc.xpath('/opf:package/opf:spine', EPUB::NAMESPACES).first
         %w[ id toc page-progression-direction ].each do |attr|
-          spine.send("#{attr.gsub(/-/, '_')}=", elem[attr])
+          spine.__send__("#{attr.gsub(/-/, '_')}=", elem[attr])
         end
 
-        elem.xpath('./opf:itemref', EPUB::NAMESPACES).each do |elm|
+        elem.xpath('./opf:itemref', EPUB::NAMESPACES).each do |e|
           itemref = EPUB::Publication::Package::Spine::Itemref.new
           %w[ idref id ].each do |attr|
-            itemref.send "#{attr}=", elm[attr]
+            itemref.__send__("#{attr}=", e[attr])
           end
-          itemref.linear = (elm['linear'] != 'no')
-          itemref.properties = elm['properties'] ? elm['properties'].split(' ') : []
+          itemref.linear = (e['linear'] != 'no')
+          itemref.properties = e['properties'] ? e['properties'].split(' ') : []
           spine << itemref
         end
 
@@ -81,7 +126,7 @@ module EPUB
         elem = @doc.xpath('/opf:package/opf:guide/opf:reference', EPUB::NAMESPACES).each do |ref|
           reference = EPUB::Publication::Package::Guide::Reference.new
           %w[ type title href ].each do |attr|
-            reference.send("#{attr}=", ref[attr])
+            reference.__send__("#{attr}=", ref[attr])
           end
           reference.iri = @rootfile.join Addressable::URI.parse(reference.href)
           guide << reference

data/lib/epub/parser/version.rb

@@ -1,5 +1,5 @@
 module EPUB
   class Parser
-    VERSION = "0.0.5"
+    VERSION = "0.0.6"
   end
 end

data/lib/epub/publication/package.rb

@@ -1,4 +1,4 @@
-%w[ metadata manifest spine guide ].each { |f| require "epub/publication/package/#{f}" }
+%w[ metadata manifest spine guide bindings ].each { |f| require "epub/publication/package/#{f}" }
 
 module EPUB
   module Publication
@@ -10,6 +10,8 @@ module EPUB
       alias lang= xml_lang=
 
       def metadata=(metadata)
+        metadata.package = self
+        @metadata = metadata
       end
 
       def manifest=(manifest)

data/lib/epub/publication/package/guide.rb

@@ -15,7 +15,6 @@ module EPUB
           references << reference
         end
 
-        # Should use epub:type list?
         %w[cover title-page toc index glossary acknowledgements bibliography colophon copyright-page dedication epigraph foreword loi lot notes preface text].each do |type|
           define_method type do
             var = instance_variable_get "@#{type}"

data/lib/epub/publication/package/manifest.rb

@@ -8,7 +8,6 @@ module EPUB
         attr_accessor :package,
                       :id
 
-        # syntax sugar for #items.<<
         def <<(item)
           @items ||= {}
           item.manifest = self
@@ -57,16 +56,16 @@ module EPUB
             if ad = options[:supported]
               supported = supported | (ad.respond_to?(:to_ary) ? ad : [ad])
             end
-            if del = options[:not_supported]
+            if del = options[:unsupported]
               supported = supported - (del.respond_to?(:to_ary) ? del : [del])
             end
 
             if supported.include? media_type
               yield self
             elsif fallback
-              fallback.use_fallback_chain(options) {|fallbacked| yield fallbacked}
+              fallback.use_fallback_chain(options) {|fb| yield fb}
             else
-              raise EPUB::MediaType::NotSupportedError
+              raise EPUB::MediaType::UnsupportedError
             end
           end
 

data/lib/epub/publication/package/metadata.rb

@@ -1,16 +1,51 @@
+require 'json'
+
 module EPUB
   module Publication
     class Package
       class Metadata
-        elems = [:identifiers, :titles, :languages] +
-                [:contributers, :coverages, :creators, :dates, :descriptions, :formats, :publishers,
-                 :relations, :rigths, :sources, :subjects, :types]
-        attr_accessor *(elems.collect {|elem| "dc_#{elem}"})
-        elems.each do |elem|
+        ELEMS = [:identifiers, :titles, :languages] +
+                [:contributors, :coverages, :creators, :dates, :descriptions, :formats, :publishers,
+                 :relations, :rights, :sources, :subjects, :types]
+        attr_accessor :package,
+                      *(ELEMS.collect {|elem| "dc_#{elem}"})
+        ELEMS.each do |elem|
           alias_method elem, "dc_#{elem}"
           alias_method "#{elem}=", "dc_#{elem}="
         end
-        attr_accessor :metas, :links
+        attr_accessor :links
+
+        def to_json(pretty = false)
+          obj = {}
+          ELEMS.each do |elem|
+            obj[elem] = __send__(elem)
+          end
+          obj.to_json
+        end
+
+        class Identifier
+          attr_accessor :content, :id
+
+          def to_s
+            content
+          end
+        end
+
+        class Title
+          attr_accessor :content, :id, :lang, :dir
+
+          def to_s
+            content
+          end
+        end
+
+        class DCMES
+          attr_accessor :content, :id, :lang, :dir
+
+          def to_s
+            content
+          end
+        end
       end
     end
   end

data/test/fixtures/book/OPS//343/203/253/343/203/274/343/203/210/343/203/225/343/202/241/343/202/244/343/203/253.opf

@@ -1,8 +1,44 @@
 <?xml version="1.0"?>
 <package version="3.0"
-	 uniqueue-identifier="pub-id"
+	 unique-identifier="pub-id"
 	 xmlns="http://www.idpf.org/2007/opf">
-  <metadata></metadata>
+  <metadata xmlns:dc="http://purl.org/dc/elements/1.1/">
+    <dc:identifier id="pub-id">da265185-8da8-462d-a146-17dd388f61fc</dc:identifier>
+
+    <dc:title id="t1" xml:lang="fr">Mon premier guide de cuisson, un Mémoire</dc:title>
+    <meta refines="#t1" property="title-type">main</meta>
+    <meta refines="#t1" property="display-seq">2</meta>
+    
+    <dc:title id="t2">The Great Cookbooks of the World</dc:title>
+    <meta refines="#t2" property="title-type">collection</meta>
+    <meta refines="#t2" property="display-seq">1</meta>
+    
+    <dc:title id="t3">The New French Cuisine Masters</dc:title>
+    <meta refines="#t3" property="title-type">collection</meta>
+    <meta refines="#t3" property="group-position">2</meta>
+    <meta refines="#t3" property="display-seq">3</meta>
+    
+    <dc:title id="t4">Special Anniversary Edition</dc:title>
+    <meta refines="#t4" property="title-type">edition</meta>
+    <meta refines="#t4" property="display-seq">4</meta>
+    
+    <dc:title id="t5">The Great Cookbooks of the World: 
+        Mon premier guide de cuisson, un Mémoire. 
+        The New French Cuisine Masters, Volume Two. 
+        Special Anniversary Edition</dc:title>
+    <meta refines="#t5" property="title-type">extended</meta>
+
+    <dc:language>ja</dc:language>
+    <meta property="dcterms:modified">2012-02-03T12:00:00Z</meta>
+
+    <dc:creator id="creator01">Lewis Carroll</dc:creator>
+    <meta refines="#creator01" property="role" scheme="marc:relators">aut</meta>
+    <meta refines="#creator01" property="display-seq">1</meta>
+    
+    <dc:creator id="creator02">John Tenniel</dc:creator>
+    <meta refines="#creator02" property="role" scheme="marc:relators">ill</meta>
+    <meta refines="#creator02" property="display-seq">2</meta>
+  </metadata>
   <manifest id="manifest-id">
     <item id="nav"
 	  href="nav.xhtml"

data/test/test_parser.rb

@@ -23,8 +23,8 @@ class TestParser < Test::Unit::TestCase
       @book = @parser.parse
     end
 
-    def test_each_page_by_spine_iterates_items_in_spines_order
-      @book.each_page_by_spine do |page|
+    def test_each_page_on_spine_iterates_items_in_spines_order
+      @book.each_page_on_spine do |page|
         assert_instance_of EPUB::Publication::Package::Manifest::Item, page
       end
     end
@@ -48,13 +48,5 @@ class TestParser < Test::Unit::TestCase
         assert_instance_of EPUB::Publication::Package::Manifest::Item, page
       end
     end
-
-    def test_h
-      Modify method name
-
-      @book.resources.each do |item|
-        assert_instance_of EPUB::Publication::Package::Manifest::Item, item
-      end
-    end
   end
 end

data/test/test_parser_publication.rb

@@ -7,6 +7,21 @@ class TestParserPublication < Test::Unit::TestCase
     @parser = EPUB::Parser::Publication.new 'test/fixtures/book/OPS/ルートファイル.opf'
   end
 
+  class TestParseMetaata < TestParserPublication
+    def setup
+      super
+      @metadata = @parser.parse_metadata
+    end
+
+    def test_has_identifier
+      assert_equal 'da265185-8da8-462d-a146-17dd388f61fc', @metadata.identifiers.first.content
+    end
+
+    def test_has_five_titles
+      assert_equal 5, @metadata.titles.length
+    end
+  end
+
   class TestParseManifest < TestParserPublication
     def setup
       super
@@ -70,7 +85,7 @@ class TestParserPublication < Test::Unit::TestCase
     def test_item_can_custome_not_supported_media_type_in_use_fallback_chain
       item = @manifest['manifest-item-2']
       fallback = item.fallback.fallback
-      result = item.use_fallback_chain not_supported: 'image/svg+xml' do |current|
+      result = item.use_fallback_chain unsupported: 'image/svg+xml' do |current|
         current
       end
 

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: epub-parser
 version: !ruby/object:Gem::Version
-  version: 0.0.5
+  version: 0.0.6
   prerelease: 
 platform: ruby
 authors:
@@ -9,11 +9,11 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-01-08 00:00:00.000000000Z
+date: 2012-02-14 00:00:00.000000000Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rubygems-test
-  requirement: &11713580 !ruby/object:Gem::Requirement
+  requirement: &7403740 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -21,10 +21,10 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *11713580
+  version_requirements: *7403740
 - !ruby/object:Gem::Dependency
   name: rake
-  requirement: &11712960 !ruby/object:Gem::Requirement
+  requirement: &7421660 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -32,10 +32,10 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *11712960
+  version_requirements: *7421660
 - !ruby/object:Gem::Dependency
   name: pry
-  requirement: &11712220 !ruby/object:Gem::Requirement
+  requirement: &7421060 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -43,10 +43,10 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *11712220
+  version_requirements: *7421060
 - !ruby/object:Gem::Dependency
   name: test-unit
-  requirement: &11730760 !ruby/object:Gem::Requirement
+  requirement: &7420480 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -54,10 +54,10 @@ dependencies:
         version: '2'
   type: :development
   prerelease: false
-  version_requirements: *11730760
+  version_requirements: *7420480
 - !ruby/object:Gem::Dependency
   name: simplecov
-  requirement: &11730060 !ruby/object:Gem::Requirement
+  requirement: &7419900 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -65,10 +65,10 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *11730060
+  version_requirements: *7419900
 - !ruby/object:Gem::Dependency
   name: thin
-  requirement: &11729480 !ruby/object:Gem::Requirement
+  requirement: &7419300 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -76,10 +76,10 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *11729480
+  version_requirements: *7419300
 - !ruby/object:Gem::Dependency
   name: yard
-  requirement: &11728780 !ruby/object:Gem::Requirement
+  requirement: &7418720 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -87,10 +87,10 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *11728780
+  version_requirements: *7418720
 - !ruby/object:Gem::Dependency
   name: epzip
-  requirement: &11728300 !ruby/object:Gem::Requirement
+  requirement: &7418100 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -98,10 +98,10 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *11728300
+  version_requirements: *7418100
 - !ruby/object:Gem::Dependency
   name: epubcheck
-  requirement: &11727620 !ruby/object:Gem::Requirement
+  requirement: &7416760 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -109,10 +109,10 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *11727620
+  version_requirements: *7416760
 - !ruby/object:Gem::Dependency
   name: epub_validator
-  requirement: &11727080 !ruby/object:Gem::Requirement
+  requirement: &7414340 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -120,10 +120,32 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *11727080
+  version_requirements: *7414340
+- !ruby/object:Gem::Dependency
+  name: zipruby
+  requirement: &7427980 !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: *7427980
+- !ruby/object:Gem::Dependency
+  name: rubyzip
+  requirement: &7442820 !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: *7442820
 - !ruby/object:Gem::Dependency
   name: enumerabler
-  requirement: &11726420 !ruby/object:Gem::Requirement
+  requirement: &7440660 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -131,10 +153,10 @@ dependencies:
         version: '0'
   type: :runtime
   prerelease: false
-  version_requirements: *11726420
+  version_requirements: *7440660
 - !ruby/object:Gem::Dependency
   name: nokogiri
-  requirement: &11725700 !ruby/object:Gem::Requirement
+  requirement: &7439140 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -142,10 +164,10 @@ dependencies:
         version: '0'
   type: :runtime
   prerelease: false
-  version_requirements: *11725700
+  version_requirements: *7439140
 - !ruby/object:Gem::Dependency
   name: addressable
-  requirement: &11725080 !ruby/object:Gem::Requirement
+  requirement: &7436940 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -153,11 +175,12 @@ dependencies:
         version: '0'
   type: :runtime
   prerelease: false
-  version_requirements: *11725080
+  version_requirements: *7436940
 description: Parse EPUB 3 book loosely
 email:
 - KitaitiMakoto@gmail.com
-executables: []
+executables:
+- epubinfo
 extensions: []
 extra_rdoc_files: []
 files:
@@ -165,8 +188,9 @@ files:
 - .gitignore
 - Gemfile
 - MIT-LICENSE
-- README
+- README.markdown
 - Rakefile
+- bin/epubinfo
 - epub-parser.gemspec
 - lib/epub.rb
 - lib/epub/book.rb
@@ -221,7 +245,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: -4205757875746414572
+      hash: 3679147164764535889
 required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
   requirements:
@@ -230,7 +254,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: -4205757875746414572
+      hash: 3679147164764535889
 requirements: []
 rubyforge_project: 
 rubygems_version: 1.8.8

data/README

@@ -1,14 +0,0 @@
-EPUB Parer
-==========
-
-USAGE
------
-    book = EPUB::Parser.parse 'book.epub'
-    book.each_page_in_spines_order do |page|
-      # do somethong...
-    end
-
-LICENSE
--------
-This library is distribuetd under the term of the MIT License.
-See MIT-LICENSE file for more info.