epub-parser 0.0.1

data/.gitignore

@@ -0,0 +1,6 @@
+*.gem
+.bundle
+Gemfile.lock
+pkg/*
+vendor/*
+*~

data/Gemfile

@@ -0,0 +1,4 @@
+source "http://rubygems.org"
+
+# Specify your gem's dependencies in epub-parser.gemspec
+gemspec

data/Rakefile

@@ -0,0 +1,12 @@
+require 'bundler/gem_tasks'
+require 'rake/testtask'
+require 'yard'
+
+task :default => :test
+
+Rake::TestTask.new do |task|
+  task.test_files = FileList['test/**/test_*.rb']
+  ENV['TESTOPTS'] = '--no-show-detail-immediately --verbose'
+end
+
+YARD::Rake::YardocTask.new

data/epub-parser.gemspec

@@ -0,0 +1,33 @@
+# -*- encoding: utf-8 -*-
+$:.push File.expand_path("../lib", __FILE__)
+require "epub/parser/version"
+
+Gem::Specification.new do |s|
+  s.name        = "epub-parser"
+  s.version     = EPUB::Parser::VERSION
+  s.authors     = ["KITAITI Makoto"]
+  s.email       = ["KitaitiMakoto@gmail.com"]
+  s.homepage    = "https://gitorious.org/epub/parser"
+  s.summary     = %q{EPUB 3 Parser}
+  s.description = %q{Parse EPUB 3 book loosely}
+
+  # s.rubyforge_project = "epub-parser"
+
+  s.files         = `git ls-files`.split("\n").push('test/fixtures/book/OPS/ルートファイル.opf')
+  s.files.delete('"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"')
+  s.test_files    = `git ls-files -- {test,spec,features}/**/*.rb`.split("\n") 
+  s.executables   = `git ls-files -- bin/*`.split("\n").map{ |f| File.basename(f) }
+  s.require_paths = ["lib"]
+
+  s.add_development_dependency 'rubygems-test'
+  s.add_development_dependency 'rake'
+  s.add_development_dependency 'pry'
+  s.add_development_dependency 'test-unit', '~> 2'
+  s.add_development_dependency 'simplecov'
+  s.add_development_dependency 'thin'
+  s.add_development_dependency 'yard'
+
+  s.add_runtime_dependency 'enumerabler'
+  s.add_runtime_dependency 'nokogiri'
+  s.add_runtime_dependency 'addressable'
+end

data/lib/epub/book.rb

@@ -0,0 +1,37 @@
+require 'epub/ocf'
+require 'epub/publication'
+require 'epub/content_document'
+
+module EPUB
+  class Book
+    attr_accessor :ocf, :package, :content_document
+
+    def each_page_by_spine(&blk)
+      enum = @package.spine.items
+      if block_given?
+        enum.each &blk
+      else
+        enum
+      end
+    end
+    def each_page_by_toc(&blk)
+    end
+
+    def each_content(&blk)
+      enum = @package.manifest.items
+      if block_given?
+        enum.each &blk
+      else
+        enum.to_enum
+      end
+    end
+
+    def other_navigation
+    end
+
+    # Syntax suger
+    def rootfile_path
+      ocf.container.rootfile.full_path
+    end
+  end
+end

data/lib/epub/content_document.rb

@@ -0,0 +1,6 @@
+require 'epub/content_document/navigation'
+
+module EPUB
+  module ContentDocument
+  end
+end

data/lib/epub/content_document/navigation.rb

@@ -0,0 +1,80 @@
+module EPUB
+  module ContentDocument
+    class Navigation
+      attr_accessor :navs
+      alias navigations navs
+      alias navigations= navs=
+
+      def toc
+        navs.selector {|nav| nav.type == Type::TOC}.first
+      end
+
+      def page_list
+        navs.selector {|nav| nav.type == Type::PAGE_LIST}.first
+      end
+
+      def landmarks
+        navs.selector {|nav| nav.type == Type::LANDMARKS}.first
+      end
+
+      # Enumerator version of toc
+      #  Usage: nagivation.enum_for(:contents)
+      def contents
+      end
+
+      # Enumerator version of page_list
+      #  Usage: navigation.enum_for(:pages)
+      def pages
+      end
+
+      # iterator for #toc
+      def each_content
+      end
+
+      # iterator for #page_list
+      def each_page
+      end
+
+      # iterator for #landmark
+      def each_landmark
+      end
+
+      class Nav
+        attr_accessor :heading, :ol,
+                      :items, # children of ol, thus li
+                      :type, # toc, page-list, landmarks or other
+                      :hidden
+
+        # #show method and #hide are unneccessary
+        # because this is for parser, not for builder nor manipulator
+        def hidden?
+        end
+
+        class Ol
+          # list-style :none
+          attr_accessor :hidden
+
+          def hidden?
+          end
+
+          # may be followed by ol or be a leaf node
+          class A
+            attr_accessor :ol, # optional
+                          :hidden
+
+            def hidden?
+            end
+          end
+
+          # must be followed by ol, or must not be a leaf node
+          class Span
+            attr_accessor :ol, # required
+                          :hidden
+            def hidden?
+            end
+          end
+        end
+      end
+    end
+  end
+end

data/lib/epub/ocf.rb

@@ -0,0 +1,8 @@
+module EPUB
+  class OCF
+    MODULES = %w[container encryption manifest metadata rights signatures]
+    MODULES.each {|m| require "epub/ocf/#{m}"}
+
+    attr_accessor *MODULES
+  end
+end

data/lib/epub/ocf/container.rb

@@ -0,0 +1,22 @@
+module EPUB
+  class OCF
+    class Container
+      FILE = 'container.xml'
+
+      attr_accessor :rootfiles
+
+      def initialize
+        @rootfiles = []
+      end
+
+      # syntax sugar
+      def rootfile
+        rootfiles.first
+      end
+
+      class Rootfile
+        attr_accessor :full_path, :media_type
+      end
+    end
+  end
+end

data/lib/epub/ocf/encryption.rb

@@ -0,0 +1,6 @@
+module EPUB
+  class OCF
+    class Encryption
+    end
+  end
+end

data/lib/epub/ocf/manifest.rb

@@ -0,0 +1,6 @@
+module EPUB
+  class OCF
+    class Manifest
+    end
+  end
+end

data/lib/epub/ocf/metadata.rb

@@ -0,0 +1,6 @@
+module EPUB
+  class OCF
+    class Metadata
+    end
+  end
+end

data/lib/epub/ocf/rights.rb

@@ -0,0 +1,6 @@
+module EPUB
+  class OCF
+    class Rights
+    end
+  end
+end

data/lib/epub/ocf/signatures.rb

@@ -0,0 +1,6 @@
+module EPUB
+  class OCF
+    class Signatures
+    end
+  end
+end

data/lib/epub/parser.rb

@@ -0,0 +1,46 @@
+require 'epub/book'
+require 'epub/parser/version'
+require 'epub/parser/ocf'
+require 'epub/parser/publication'
+require 'epub/parser/content_document'
+require 'nokogiri'
+
+module EPUB
+  class Parser
+    def initialize(filepath, root_directory, options = {})
+      raise 'File #{filepath} not readable' unless File.readable_real? filepath
+      raise 'File #{root_directory} already exists' if File.file? root_directory
+        
+      @filepath = File.realpath filepath
+      Dir.mkdir(root_directory) unless File.directory? root_directory
+      @dir = File.realpath root_directory
+
+      @book = Book.new
+
+      unzip_cmd = options['unzip-command'] || 'unzip'
+      unzip_cmd << " #{@filepath} -d #{@dir}"
+      system unzip_cmd
+    end
+
+    def parse
+      @book.ocf = parse_ocf
+      @book.package = parse_publication
+      @book.content_document = parse_content_document
+      # ...
+
+      @book
+    end
+
+    def parse_ocf
+      OCF.parse @dir
+    end
+
+    def parse_publication
+      Publication.parse File.join(@dir, @book.rootfile_path)
+    end
+
+    def parse_content_document
+      # ContentDocument.parse @dir
+    end
+  end
+end

data/lib/epub/parser/content_document.rb

@@ -0,0 +1,58 @@
+require 'epub/content_document'
+require 'epub/parser'
+require 'nokogiri'
+
+module EPUB
+  class Parser
+    class ContentDocument
+      class << self
+        def parse(root_directory)
+          new(root_directory).parse
+        end
+      end
+
+      def initialize(root_directory)
+        @dir = root_directory
+      end
+
+      def parse
+        raise 'Not implemented yet'
+      end
+
+      # @param [Nokogiri::HTML::Document] document HTML document or element including nav
+      # @return [Array<EPUB::ContentDocument::Navigation::Nav>] navs array of Nav object
+      def parse_navigations(document)
+        navs = document.search('/xhtml:html/xhtml:body//xhtml:nav', Parser::NAMESPACES).collect {|elem| parse_navigation elem}
+      end
+
+      # @param [Nokogiri::XML::Element] nav 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
+
+        # to find type, need to use strict xpath for handling namespaces?
+        # And if so, where should the namespaces be defined?
+        # nav.type = element['epub:type']
+        element.namespaces['epub'] = "http://www.idpf.org/2007/ops"
+        p element.namespaces
+        nav.type = element['epub:type']
+        p nav.type
+
+        nav
+      end
+
+      private
+
+      # @param [Nokogiri::XML::Element] nav nav element
+      # @return [String] heading heading text
+      def find_heading(element)
+        heading = element.xpath('./h1|h2|h3|h4|h5|h6|hgroup').first
+
+        return heading.text unless heading.name == 'hgroup'
+
+        (heading/'h1' || heading/'h2' || heading/'h3' || heading/'h4' || heading/'h5' || heading/'h6').first.text
+      end
+    end
+  end
+end

data/lib/epub/parser/ocf.rb

@@ -0,0 +1,59 @@
+require 'epub/ocf'
+require 'nokogiri'
+
+module EPUB
+  class Parser
+    class OCF
+      DIRECTORY = 'META-INF'
+      EPUB::OCF::MODULES.each {|m| self.const_set "#{m.upcase}_FILE", "#{m}.xml"}
+
+      class << self
+        def parse(root_directory)
+          new(root_directory).parse
+        end
+      end
+
+      def initialize(root_directory)
+        @dir = root_directory
+        @ocf = EPUB::OCF.new
+      end
+
+      def parse
+        EPUB::OCF::MODULES.each do |m|
+          @ocf.send "#{m}=", send("parse_#{m}")
+        end
+        @ocf
+      end
+
+      def parse_container
+        container = EPUB::OCF::Container.new
+        doc = Nokogiri.XML open(File.join @dir, DIRECTORY, CONTAINER_FILE)
+
+        doc.xpath('/xmlns:container/xmlns:rootfiles/xmlns:rootfile', doc.namespaces).each do |elem|
+          rootfile = EPUB::OCF::Container::Rootfile.new
+          %w[full-path media-type].each do |attr|
+            rootfile.send(attr.gsub(/-/, '_') + '=', elem[attr])
+            container.rootfiles << rootfile
+          end
+        end
+
+        container
+      end
+
+      def parse_encryption
+      end
+
+      def parse_manifest
+      end
+
+      def parse_metadata
+      end
+
+      def parse_rights
+      end
+
+      def parse_signatures
+      end
+    end
+  end
+end

data/lib/epub/parser/publication.rb

@@ -0,0 +1,87 @@
+require 'nokogiri'
+require 'addressable/uri'
+require 'epub/publication'
+
+module EPUB
+  class Parser
+    class Publication
+      class << self
+        def parse(file)
+          new(file).parse
+        end
+      end
+
+      def initialize(file)
+        @package = EPUB::Publication::Package.new
+        @rootfile = Addressable::URI.parse File.realpath(file)
+        @doc = Nokogiri.XML open(file)
+      end
+
+      def parse
+        # parse_metadata
+        parse_manifest
+        parse_spine
+        # parse_guide
+        # parse_bindings
+
+        @package
+      end
+
+      def parse_metadata
+        raise 'still not implemented'
+      end
+
+      def parse_manifest
+        manifest = @package.manifest = EPUB::Publication::Package::Manifest.new
+        elem = @doc.xpath('/xmlns:package/xmlns:manifest', @doc.namespaces).first
+        manifest.id = elem['id']
+
+        fallback_map = {}
+        elem.xpath('./xmlns:item').each do |elm|
+          item = EPUB::Publication::Package::Manifest::Item.new
+          %w[ id media-type media-overlay ].each do |attr|
+            item.send "#{attr.gsub(/-/, '_')}=", elm[attr]
+          end
+          # item.href = Addressable::URI.parse elm['href']
+          item.href = @rootfile.join Addressable::URI.parse(elm['href'])
+          fallback_map[elm['fallback']] = item if elm['fallback']
+          item.properties = elm['properties'] ? elm['properties'].split(' ') : []
+          manifest << item
+        end
+        fallback_map.each_pair do |id, from|
+          from.fallback = manifest[id]
+        end
+
+        manifest
+      end
+
+      def parse_spine
+        spine = @package.spine = EPUB::Publication::Package::Spine.new
+        elem = @doc.xpath('/xmlns:package/xmlns:spine', @doc.namespaces).first
+        %w[ id toc page-progression-direction ].each do |attr|
+          spine.send("#{attr.gsub(/-/, '_')}=", elem[attr])
+        end
+
+        elem.xpath('./xmlns:itemref', @doc.namespaces).each do |elm|
+          itemref = EPUB::Publication::Package::Spine::Itemref.new
+          %w[ idref id ].each do |attr|
+            itemref.send "#{attr}=", elm[attr]
+          end
+          itemref.linear = !(elm['linear'] == 'no')
+          itemref.properties = elm['properties'] ? elm['properties'].split(' ') : []
+          spine << itemref
+        end
+
+        spine
+      end
+
+      def parse_guide
+        raise 'still not implemented'
+      end
+
+      def parse_bindings
+        raise 'still not implemented'
+      end
+    end
+  end
+end