epub-parser 0.1.2 → 0.1.3

data/lib/epub/parser/content_document.rb

@@ -1,18 +1,48 @@
 require 'epub/content_document'
 require 'epub/constants'
+require 'epub/parser/utils'
 require 'nokogiri'
 
 module EPUB
   class Parser
     class ContentDocument
+      include Utils
+
       class << self
-        def parse
-          new.parse
+        # @param [EPUB::Publication::Package::Manifest::Item] item
+        def parse(item)
+          new(item).parse
         end
       end
 
+      # @param [EPUB::Publication::Package::Manifest::Item] item
+      def initialize(item)
+        @item = item
+      end
+
       def parse
-        raise 'Not implemented yet'
+        content_document = case @item.media_type
+                           when 'application/xhtml+xml'
+                             if @item.nav?
+                               EPUB::ContentDocument::Navigation.new
+                             else
+                               EPUB::ContentDocument::XHTML.new
+                             end
+                           when 'image/svg+xml'
+                             EPUB::ContentDocument::SVG.new
+                           else
+                             nil
+                           end
+        return content_document if content_document.nil?
+        content_document.item = @item
+        document = Nokogiri.XML(@item.read)
+        # parse_content_document(document)
+        if @item.nav?
+          content_document.navigations = parse_navigations(document)
+        else
+          raise NotImplementedError
+        end
+        content_document
       end
 
       # @param [Nokogiri::HTML::Document] document HTML document or element including nav
@@ -24,13 +54,51 @@ module EPUB
       # @param [Nokogiri::XML::Element] element nav element
       # @return [EPUB::ContentDocument::Navigation::Nav] nav Nav object
       def parse_navigation(element)
-        nav = EPUB::ContentDocument::Navigation::Nav.new
-        nav.heading = find_heading element
-        nav.type = element['type']
+        nav = EPUB::ContentDocument::Navigation::Navigation.new
+        nav.text = find_heading(element)
+        nav.type = extract_attribute(element, 'type', 'epub')
+        nav.items = element.xpath('./xhtml:ol/xhtml:li', EPUB::NAMESPACES).map {|elem| parse_navigation_item(elem)}
 
         nav
       end
 
+      # @param [Nokogiri::XML::Element] element li element
+      def parse_navigation_item(element)
+        item = EPUB::ContentDocument::Navigation::Item.new
+        a_or_span = element.xpath('./xhtml:a[1]|xhtml:span[1]', EPUB::NAMESPACES).first
+        return a_or_span if a_or_span.nil?
+
+        item.text = a_or_span.text
+        if a_or_span.name == 'a'
+          if item.text.empty?
+            embedded_content = a_or_span.xpath('./xhtml:audio[1]|xhtml:canvas[1]|xhtml:embed[1]|xhtml:iframe[1]|xhtml:img[1]|xhtml:math[1]|xhtml:object[1]|xhtml:svg[1]|xhtml:video[1]', EPUB::NAMESPACES).first
+            unless embedded_content.nil?
+              case embedded_content.name
+              when 'audio'
+              when 'canvas'
+              when 'embed'
+              when 'iframe'
+                item.text = (extract_attribute(embedded_content, 'name') || extract_attribute(embedded_content, 'srcdoc')).to_s
+              when 'img'
+                item.text = extract_attribute(embedded_content, 'alt').to_s
+              when 'math'
+              when 'object'
+                item.text = extract_attribute(embedded_content, 'name').to_s
+              when 'svg'
+              when 'video'
+              else
+              end
+            end
+            item.text = extract_attribute(a_or_span, 'title').to_s if item.text.nil? || item.text.empty?
+          end
+          item.href = Addressable::URI.parse(extract_attribute(a_or_span, 'href'))
+          item.item = @item.manifest.items.selector {|it| it.href.request_uri == item.href.request_uri}.first
+        end
+        item.items = element.xpath('./xhtml:ol[1]/xhtml:li', EPUB::NAMESPACES).map {|li| parse_navigation_item(li)}
+
+        item
+      end
+
       private
 
       # @param [Nokogiri::XML::Element] element nav element

data/lib/epub/parser/ocf.rb

@@ -6,6 +6,8 @@ require 'nokogiri'
 module EPUB
   class Parser
     class OCF
+      include Utils
+
       DIRECTORY = 'META-INF'
       EPUB::OCF::MODULES.each {|m| self.const_set "#{m.upcase}_FILE", "#{m}.xml"}
 
@@ -38,7 +40,8 @@ module EPUB
         doc.xpath('/ocf:container/ocf:rootfiles/ocf: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])
+            value = extract_attribute(elem, attr)
+            rootfile.__send__(attr.gsub(/-/, '_') + '=', value)
           end
           container.rootfiles << rootfile
         end

data/lib/epub/parser/publication.rb

@@ -8,9 +8,11 @@ require 'epub/constants'
 module EPUB
   class Parser
     class Publication
+      include Utils
+
       class << self
         def parse(zip_archive, file)
-          opf = zip_archive.fopen(file).read
+          opf = zip_archive.fopen(Addressable::URI.unencode(file)).read
           new(opf, file).parse
         end
       end
@@ -35,11 +37,10 @@ module EPUB
       def parse_package
         elem = @doc.root
         %w[version xml:lang dir id].each do |attr|
-          writer = attr.gsub(/\:/, '_') + '='
-          @package.__send__(writer, elem[attr])
+          @package.__send__ "#{attr.gsub(/\:/, '_')}=", extract_attribute(elem, attr)
         end
         @unique_identifier_id = elem['unique-identifier']
-        @package.prefix = parse_prefix(elem['prefix'])
+        @package.prefix = parse_prefix(extract_attribute(elem, 'prefix'))
 
         @package
       end
@@ -52,7 +53,7 @@ module EPUB
         metadata.identifiers = elem.xpath('./dc:identifier', EPUB::NAMESPACES).collect do |e|
           identifier = EPUB::Publication::Package::Metadata::DCMES.new
           identifier.content = e.content
-          identifier.id = id = e['id']
+          identifier.id = id = extract_attribute(e, 'id')
           metadata.unique_identifier = identifier if id == @unique_identifier_id
 
           identifier
@@ -62,7 +63,7 @@ module EPUB
         metadata.titles = elem.xpath('./dc:title', EPUB::NAMESPACES).collect do |e|
           title = EPUB::Publication::Package::Metadata::Title.new
           %w[ id lang dir ].each do |attr|
-            title.__send__("#{attr}=", e[attr])
+            title.__send__("#{attr}=", extract_attribute(e, attr))
           end
           title.content = e.content
 
@@ -85,9 +86,12 @@ module EPUB
 
         metadata.metas = elem.xpath('./opf:meta', EPUB::NAMESPACES).collect do |e|
           meta = EPUB::Publication::Package::Metadata::Meta.new
-          %w[ property id scheme ].each { |attr| meta.__send__("#{attr}=", e[attr]) }
+          %w[property id scheme].each do |attr|
+            meta.__send__ "#{attr}=", extract_attribute(e, attr)
+          end
           meta.content = e.content
-          if (refines = e['refines']) && refines[0] == '#'
+          refines = extract_attribute(e, 'refines')
+          if refines && refines[0] == '#'
             id = refines[1..-1]
             id_map[id] ||= {}
             id_map[id][:refiners] ||= []
@@ -101,11 +105,12 @@ module EPUB
         metadata.links = elem.xpath('./opf:link', EPUB::NAMESPACES).collect do |e|
           link = EPUB::Publication::Package::Metadata::Link.new
           %w[ id media-type ].each do |attr|
-            link.__send__(attr.gsub(/-/, '_') + '=', e[attr])
+            link.__send__ (attr.gsub(/-/, '_') + '='), extract_attribute(e, attr)
           end
-          link.href = Addressable::URI.parse(e['href'])
-          link.rel = e['rel'].strip.split
-          if (refines = e['refines']) && refines[0] == '#'
+          link.href = Addressable::URI.parse(extract_attribute(e, 'href'))
+          link.rel = extract_attribute(e, 'rel').strip.split
+          refines = extract_attribute(e, 'refines')
+          if refines && refines[0] == '#'
             id = refines[1..-1]
             id_map[id] ||= {}
             id_map[id][:refiners] ||= []
@@ -128,17 +133,19 @@ module EPUB
       def parse_manifest
         manifest = @package.manifest = EPUB::Publication::Package::Manifest.new
         elem = @doc.xpath('/opf:package/opf:manifest', EPUB::NAMESPACES).first
-        manifest.id = elem['id']
+        manifest.id = extract_attribute(elem, 'id')
 
         fallback_map = {}
         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(/-/, '_')}=", e[attr])
+            item.__send__ "#{attr.gsub(/-/, '_')}=", extract_attribute(e, attr)
           end
-          item.href = Addressable::URI.parse(e['href'])
-          fallback_map[e['fallback']] = item if e['fallback']
-          item.properties = e['properties'] ? e['properties'].split(' ') : []
+          item.href = Addressable::URI.parse(extract_attribute(e, 'href'))
+          fallback = extract_attribute(e, 'fallback')
+          fallback_map[fallback] = item if fallback
+          properties = extract_attribute(e, 'properties')
+          item.properties = properties ? properties.split(' ') : []
           manifest << item
         end
         fallback_map.each_pair do |id, from|
@@ -152,16 +159,17 @@ 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(/-/, '_')}=", extract_attribute(elem, attr)
         end
 
         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}=", e[attr])
+            itemref.__send__ "#{attr}=", extract_attribute(e, attr)
           end
-          itemref.linear = (e['linear'] != 'no')
-          itemref.properties = e['properties'] ? e['properties'].split(' ') : []
+          itemref.linear = (extract_attribute(e, 'linear') != 'no')
+          properties = extract_attribute(e, 'properties')
+          itemref.properties = properties ? properties.split(' ') : []
           spine << itemref
         end
 
@@ -173,9 +181,9 @@ module EPUB
         @doc.xpath('/opf:package/opf:guide/opf:reference', EPUB::NAMESPACES).each do |ref|
           reference = EPUB::Publication::Package::Guide::Reference.new
           %w[ type title ].each do |attr|
-            reference.__send__("#{attr}=", ref[attr])
+            reference.__send__ "#{attr}=", extract_attribute(ref, attr)
           end
-          reference.href = Addressable::URI.parse(ref['href'])
+          reference.href = Addressable::URI.parse(extract_attribute(ref, 'href'))
           guide << reference
         end
 
@@ -186,9 +194,9 @@ module EPUB
         bindings = @package.bindings = EPUB::Publication::Package::Bindings.new
         @doc.xpath('/opf:package/opf:bindings/opf:mediaType', EPUB::NAMESPACES).each do |elem|
           media_type = EPUB::Publication::Package::Bindings::MediaType.new
-          media_type.media_type = elem['media-type']
+          media_type.media_type = extract_attribute(elem, 'media-type')
           items = @package.manifest.items
-          media_type.handler = items.detect {|item| item.id == elem['handler']}
+          media_type.handler = items.detect {|item| item.id == extract_attribute(elem, 'handler')}
           bindings << media_type
         end
 
@@ -218,7 +226,7 @@ module EPUB
           md = EPUB::Publication::Package::Metadata::DCMES.new
           md.content = e.content
           %w[ id lang dir ].each do |attr|
-            md.__send__("#{attr}=", e[attr])
+            md.__send__ "#{attr}=", extract_attribute(e, attr)
           end
           yield(md, e) if block_given?
           md

data/lib/epub/parser/utils.rb

@@ -0,0 +1,20 @@
+module EPUB
+  class Parser
+    module Utils
+      # Extract the value of attribute of element
+      # 
+      # @todo Refinement Nokogiri::XML::Node instead of use this method after Ruby 2.0 becomes popular
+      # 
+      # @param [Nokogiri::XML::Element] element
+      # @param [String] name name of attribute excluding namespace prefix
+      # @param [String, nil] prefix XML namespace prefix in {EPUB::Constants::NAMESPACES} keys
+      # @return [String] value of attribute when the attribute exists
+      # @return nil when the attribute doesn't exist
+      def extract_attribute(element, name, prefix=nil)
+        attr = element.attribute_with_ns(name, EPUB::NAMESPACES[prefix])
+        attr.nil? ? nil : attr.value
+      end
+      module_function :extract_attribute
+    end
+  end
+end

data/lib/epub/parser/version.rb

@@ -1,5 +1,5 @@
 module EPUB
   class Parser
-    VERSION = "0.1.2"
+    VERSION = "0.1.3"
   end
 end

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

@@ -1,5 +1,6 @@
 require 'enumerabler'
 require 'epub/constants'
+require 'epub/parser/content_document'
 
 module EPUB
   module Publication
@@ -14,9 +15,8 @@ module EPUB
           @items[item.id] = item
         end
 
-        # syntax sugar
         def navs
-          items.selector {|i| i.properties.include? 'nav'}
+          items.selector(&:nav?)
         end
 
         def nav
@@ -62,10 +62,14 @@ module EPUB
             rootfile = Addressable::URI.parse(manifest.package.book.ocf.container.rootfile.full_path)
             Zip::Archive.open(manifest.package.book.epub_file) {|zip|
               path = Addressable::URI.unescape(rootfile + href.normalize.request_uri)
-              zip.fopen(path.to_s).read
+              zip.fopen(path).read
             }
           end
 
+          def nav?
+            properties.include? 'nav'
+          end
+
           # @todo Handle circular fallback chain
           def use_fallback_chain(options = {})
             supported = EPUB::MediaType::CORE
@@ -84,6 +88,11 @@ module EPUB
             raise EPUB::MediaType::UnsupportedError
           end
 
+          def content_document
+            return nil unless %w[application/xhtml+xml image/svg+xml].include? media_type
+            @content_document ||= Parser::ContentDocument.parse(self)
+          end
+
           protected
 
           def traverse_fallback_chain(chain)

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

@@ -31,17 +31,31 @@ module EPUB
             titles.select {|title| title.title_type.to_s == type}.sort.join(' ')
           end
         end
+
         def subtitle
           titles.select {|title| title.title_type.to_s == 'subtitle'}.sort.join(' ')
         end
 
-        def to_hash
+        def description
+          descriptions.join ' '
+        end
+
+        def date
+          dates.first
+        end
+
+        def to_h
           DC_ELEMS.inject({}) do |hsh, elem|
             hsh[elem] = __send__(elem)
             hsh
           end
         end
 
+        def to_hash
+          warn "#{self.class}##{__method__} is obsolete"
+          to_h
+        end
+
         def primary_metas
           metas.select {|meta| meta.primary_expression?}
         end
@@ -69,6 +83,13 @@ module EPUB
 
           attr_accessor :content, :id, :lang, :dir
 
+          def inspect
+            ivs = instance_variables.map {|iv|
+              [iv, instance_variable_get(iv).inspect].join('=')
+            }.join(' ')
+            '<#%s:%#0x %s>' % [self.class, __id__, ivs]
+          end
+
           def to_s
             content
           end
@@ -104,6 +125,13 @@ module EPUB
             ! subexpression?
           end
 
+          def inspect
+            ivs = instance_variables.map {|iv|
+              [iv, instance_variable_get(iv).inspect].join('=')
+            }.join(' ')
+            '<#%s:%#0x %s>' % [self.class, __id__, ivs]
+          end
+
           def to_s
             content
           end

data/test/helper.rb

@@ -4,3 +4,4 @@ SimpleCov.start do
 end
 
 require 'test/unit/full'
+require 'epub/parser'

data/test/test_parser.rb

@@ -1,5 +1,4 @@
 require File.expand_path 'helper', File.dirname(__FILE__)
-require 'epub/parser'
 
 class MyBook
   include EPUB

data/test/test_parser_content_document.rb

@@ -1,18 +1,31 @@
 require_relative 'helper'
-require 'epub/parser/content_document'
 
 class TestParserContentDocument < Test::Unit::TestCase
   def setup
+    @manifest = EPUB::Publication::Package::Manifest.new
+    %w[item-1.xhtml item-2.xhtml nav.xhtml].each.with_index do |href, index|
+      item = EPUB::Publication::Package::Manifest::Item.new
+      item.id = index
+      item.href = Addressable::URI.parse(href)
+      @manifest << item
+    end
+
     @dir = 'test/fixtures/book'
-    @parser = EPUB::Parser::ContentDocument.new
+    @parser = EPUB::Parser::ContentDocument.new(@manifest.items.last)
   end
 
   def test_parse_navigations
     doc = Nokogiri.XML open("#{@dir}/OPS/nav.xhtml")
     navs = @parser.parse_navigations doc
+    nav = navs.first
 
     assert_equal 1, navs.length
-    assert_equal 'Table of Contents', navs.first.heading
-    assert_equal 'toc', navs.first.type
+    assert_equal 'Table of Contents', nav.heading
+    assert_equal 'toc', nav.type
+
+    assert_equal 2, nav.items.length
+    assert_equal @manifest.items.first, nav.items.first.item
+    assert_equal @manifest.items[1], nav.items[1].items[0].item
+    assert_equal @manifest.items[1], nav.items[1].items[1].item
   end
 end