epub-parser 0.1.9 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 1b57da74df66cba76e58cbb0098d7d5618ec1188
-  data.tar.gz: 66d1bf92e61f15da35215d0d46638ec7801e5993
+  metadata.gz: 0da9ce97158d8bd76d740d45a8df755f7016c32c
+  data.tar.gz: 79c264c87d61cf10c2cf3f3bd9fd6dd316756b98
 SHA512:
-  metadata.gz: 2a499a7de09c4b906b84d63e10e104f3dd00d028f8e4700a979343cbcade7edd06f63d3651422070a55139de64461b4fcf5b3edb46de477d8fe1d5e675509a01
-  data.tar.gz: 2b223250e08e3e9061042bbcd7b9e36662ac86542a7144e84bac71081f18b7b591fcf0acb03c6cfec14ac24b1b3fa256ede7ff2fedb6c0d0807ec14a7267387a
+  metadata.gz: ba7491533f29d1cbf2350b23e24c5a786eda2db6d51f0f07b002b94103a52dc4322e0c265bc83cdc2b85054f582f1e977d94f7aa3c2b7fa43138781821f493a2
+  data.tar.gz: a3d50715ac0c54fbd0093507651fbd22da5448d4bb28c62bb9adc15c6c9c80a4a40c793c42adaba8f2d53e81561ca739cd5ebf2f2c6d7989e2d9a5c44d045536

data/CHANGELOG.markdown

@@ -1,6 +1,13 @@
 CHANGELOG
 =========
 
+0.2.0
+-----
+
+* Introduce abstraction layer for OCF physical container
+* Add `EPUB::OCF::PhysicalContainer::File` and make it possible to parse file system directory an EPUB file. See {file:docs/UnpackedArchive.markdown} for details.
+* Remove `EPUB::Parser::OCF::CONTAINER_FILE` and other constants
+
 0.1.9
 -----
 

data/README.markdown

@@ -110,6 +110,10 @@ If you find other gems, please tell me or request a pull request.
 RECENT CHANGES
 --------------
 
+### 0.2.0
+
+* Make it possible to parse file system directory an EPUB file. See {file:docs/UnpackedArchive.markdown} for details.
+
 ### 0.1.9
 
 * Introduce [Nokogumbo][] for XHTML Content Documents
@@ -160,6 +164,7 @@ DONE
 * Content Document(only for Navigation Documents)
 * Fixed Layout
 * Vocabulary Association Mechanisms(only for itemref)
+* Archive library abstraction
 
 LICENSE
 -------

data/bin/epub-open

@@ -20,6 +20,8 @@ EOB
 
 $0 = File.basename($PROGRAM_NAME)
 include EPUB::Book::Features
-EPUB::Parser.parse(ARGV.shift, :book => self)
+file = ARGV.shift
+EPUB::OCF::PhysicalContainer.adapter = :File if File.directory? file
+EPUB::Parser.parse(file, :book => self)
 $stderr.puts "Enter \"exit\" to exit #{shell}"
 shell.start

data/bin/epubinfo

@@ -30,6 +30,7 @@ unless file
   abort
 end
 
+EPUB::OCF::PhysicalContainer.adapter = :File if File.directory? file
 book = EPUB::Parser.parse(file)
 data = {'Title' => [book.title]}
 data.merge!(book.metadata.to_h)

data/docs/FixedLayout.markdown

@@ -1,4 +1,4 @@
-{file:docs/Home.markdown} > **{file:docs/FixedLayout.markdow}**
+{file:docs/Home.markdown} > **{file:docs/FixedLayout.markdown}**
 
 Fixed-Layout Documents
 ======================

data/docs/Home.markdown

@@ -95,6 +95,9 @@ More documentations are avaiable in:
 * {file:docs/Publication.markdown}
 * {file:docs/Item.markdown}
 * {file:docs/FixedLayout.markdown}
+* {file:docs/Navigation.markdown}
+* {file:docs/Searcher.markdown}
+* {file:docs/UnpackedArchive.markdown}
 
 Requirements
 ------------

data/docs/UnpackedArchive.markdown

@@ -0,0 +1,113 @@
+{file:docs/Home.markdown} > **{file:docs/UnpackedArchive.markdown}**
+
+Unpacked Archive
+================
+
+From version 0.2.0, EPUB Parser can parse EPUB books from unpacked archive, or file system directory.
+
+Let's parse pretty comic Page Blanche:
+
+    % tree page-blanche
+    page-blanche
+    ├── EPUB
+    │   ├── Content
+    │   │   ├── PageBlanche_Page_000.xhtml
+    │   │   ├── PageBlanche_Page_001.xhtml
+    │   │   ├── PageBlanche_Page_002.xhtml
+    │   │   ├── PageBlanche_Page_003.xhtml
+    │   │   ├── PageBlanche_Page_004.xhtml
+    │   │   ├── PageBlanche_Page_005.xhtml
+    │   │   ├── PageBlanche_Page_006.xhtml
+    │   │   ├── PageBlanche_Page_007.xhtml
+    │   │   ├── PageBlanche_Page_008.xhtml
+    │   │   └── cover.xhtml
+    │   ├── Image
+    │   │   ├── PageBlanche_Page_001.jpg
+    │   │   ├── PageBlanche_Page_002.jpg
+    │   │   ├── PageBlanche_Page_003.jpg
+    │   │   ├── PageBlanche_Page_004.jpg
+    │   │   ├── PageBlanche_Page_005.jpg
+    │   │   ├── PageBlanche_Page_006.jpg
+    │   │   ├── PageBlanche_Page_007.jpg
+    │   │   ├── PageBlanche_Page_008.jpg
+    │   │   └── cover.jpg
+    │   ├── Navigation
+    │   │   ├── nav.xhtml
+    │   │   └── toc.ncx
+    │   ├── Style
+    │   │   └── style.css
+    │   └── package.opf
+    ├── META-INF
+    │   └── container.xml
+    └── mimetype
+
+To load EPUB books from directory, you need specify file adapter via {EPUB::OCF::PhysicalContainer} at first:
+
+    require 'epub/parser'
+    
+    EPUB::OCF::PhysicalContainer.adapter = :File
+
+And then, directory path as EPUB path:
+
+    epub = EPUB::Parser.parse('./page-blanche')
+
+Now you can handle the EPUB book as always.
+
+    epub.title # => "Page Blache"
+    epub.each_page_on_spine.to_a.length # => 10
+    puts epub.nav.content_document.contents.map {|content| "#{File.basename(content.href.to_s)} ... #{content.text}"}
+    # PageBlanche_Page_002.xhtml ... Dédicace
+    # PageBlanche_Page_005.xhtml ... Commencer la lecture
+    # => nil
+
+If set {EPUB::OCF::PhysicalContainer.adapter}, it is used every time EPUB Parser parses books even when it's packaged EPUB file. Instead of setting adapter globally, you can also specify adapter for parsing individually by passing keyword argument `container_adapter` to `.parse` method:
+
+    # From packaged file
+    File.ftype './page-blanche.epub' # => "file"
+    archived_book = EPUB::Parser.parse('./page-blanche.epub') # => EPUB::Book
+    # From directory
+    File.ftype './page-blanche' # => "directory"
+    unpacked_book = EPUB::Parser.parse('./page-blanche', container_adapter: :File) # => EPUB::Book
+
+Command-line tools
+------------------
+
+Command-line tools `epubinfo` and `epub-open` may also handle with directory as EPUB books.
+
+Executing `epubinfo`:
+
+    $ epubinfo page-blanche
+    Title:              Page Blanche
+    Identifiers:        code.google.com.epub-samples.page-blanche
+    Titles:             Page Blanche
+    Languages:          fr
+    Contributors:       Vincent Gros
+    Coverages:          
+    Creators:           Boulet, Bagieu Pénélope
+    Dates:              2012-01-18
+    Descriptions:       
+    Formats:            
+    Publishers:         éditions Delcourt
+    Relations:          
+    Rights:             This work is shared with the public using the Attribution-ShareAlike 3.0 Unported (CC BY-SA 3.0) license.
+    Sources:            
+    Subjects:           
+    Types:              
+    Unique identifier:  code.google.com.epub-samples.page-blanche
+    Epub version:       3.0
+
+Executing `epub-open`:
+
+    $ epub-open page-blanche
+    Enter "exit" to exit IRB
+    irb: warn: can't alias bindings from irb_workspaces.
+    irb(main):001:0> title
+    => "Page Blanche"
+    irb(main):002:0> exit
+
+Note
+----
+
+Actually loading EPUB books from unpacked directory is not recommended. The reason why is it's too complex to handle with files properly because of character encoding of file names such as Unicode normalization matters like UTF-8 NFD, NFC, NFKD, NFKC and OS X-specific custom NFD, IRI normalization like percent-encoding, case sensitivity or so on. And, you know, this is not standardized way to load EPUB books. So, at least in the near future, there's not plan to support various environment.
+
+Of course, always pathces are welcome.

data/epub-parser.gemspec

@@ -32,7 +32,6 @@ Gem::Specification.new do |s|
   s.add_development_dependency 'test-unit-rr'
   s.add_development_dependency 'test-unit-notify'
   s.add_development_dependency 'simplecov'
-  s.add_development_dependency 'thin'
   s.add_development_dependency 'yard'
   s.add_development_dependency 'gem-man'
   s.add_development_dependency 'ronn'

data/lib/epub/book/features.rb

@@ -29,6 +29,14 @@ module EPUB
         end
       end
 
+      def container_adapter
+        @adapter || OCF::PhysicalContainer.adapter
+      end
+
+      def container_adapter=(adapter)
+        @adapter = adapter.instance_of?(Class) ? adapter : OCF::PhysicalContainer.const_get(adapter)
+      end
+
       # @overload each_page_on_spine(&blk)
       #   iterate over items in order of spine when block given
       #   @yieldparam item [Publication::Package::Manifest::Item]

data/lib/epub/content_document/navigation.rb

@@ -28,6 +28,7 @@ module EPUB
       # Enumerator version of page_list
       #  Usage: navigation.enum_for(:pages)
       def pages
+        raise NotImplementedError
       end
 
       # @todo Enumerator version of landmarks
@@ -41,10 +42,12 @@ module EPUB
 
       # iterator for #page_list
       def each_page
+        raise NotImplementedError
       end
 
       # iterator for #landmark
       def each_landmark
+        raise NotImplementedError
       end
 
       def navigation

data/lib/epub/ocf/physical_container.rb

@@ -0,0 +1,50 @@
+require 'epub/ocf/physical_container/zipruby'
+require 'epub/ocf/physical_container/file'
+
+module EPUB
+  class OCF
+    class PhysicalContainer
+      @adapter = Zipruby
+
+      class << self
+        def adapter
+          if self == PhysicalContainer
+            @adapter
+          else
+            raise NoMethodError.new("undefined method `#{__method__}' for #{self}")
+          end
+        end
+
+        def adapter=(adapter)
+          if self == PhysicalContainer
+            @adapter = adapter.instance_of?(Class) ? adapter : const_get(adapter)
+          else
+            raise NoMethodError.new("undefined method `#{__method__}' for #{self}")
+          end
+        end
+
+        def open(container_path)
+          _adapter.new(container_path).open do |container|
+            yield container
+          end
+        end
+
+        def read(container_path, path_name)
+          open(container_path) {|container|
+            container.read(path_name)
+          }
+        end
+
+        private
+
+        def _adapter
+          (self == PhysicalContainer) ? @adapter : self
+        end
+      end
+
+      def initialize(container_path)
+        @container_path = container_path
+      end
+    end
+  end
+end

data/lib/epub/ocf/physical_container/file.rb

@@ -0,0 +1,15 @@
+module EPUB
+  class OCF
+    class PhysicalContainer
+      class File < self
+        def open
+          yield self
+        end
+
+        def read(path_name)
+          ::File.read(::File.join(@container_path, path_name))
+        end
+      end
+    end
+  end
+end

data/lib/epub/ocf/physical_container/zipruby.rb

@@ -0,0 +1,28 @@
+require 'zipruby'
+
+module EPUB
+  class OCF
+    class PhysicalContainer
+      class Zipruby < self
+        def open
+          Zip::Archive.open @container_path do |archive|
+            @archive = archive
+            result = yield self
+            @archive = nil
+            result
+          end
+        end
+
+        def read(path_name)
+          if @archive
+            @archive.fopen(path_name) {|entry| entry.read}
+          else
+            Zip::Archive.open(@container_path) {|archive|
+              archive.fopen(path_name) {|entry| entry.read}
+            }
+          end
+        end
+      end
+    end
+  end
+end

data/lib/epub/parser.rb

@@ -1,6 +1,6 @@
 require 'epub'
 require 'epub/constants'
-require 'zipruby'
+require 'epub/book'
 require 'nokogiri'
 
 module EPUB
@@ -29,27 +29,34 @@ module EPUB
       #   For details of options, see below.
       # @option options [EPUB] :book instance of class which includes {EPUB} module
       # @option options [Class] :class class which includes {EPUB} module
+      # @option options [EPUB::OCF::PhysicalContainer, Symbol] :container_adapter OCF physical container adapter to use when parsing EPUB container
+      #   When class passed, it is used. When symbol passed, it is considered as subclass name of {EPUB::OCF::PhysicalContainer}.
+      #   If omitted, {EPUB::OCF::PhysicalContainer.adapter} is used.
       # @return [EPUB] object which is an instance of class including {EPUB} module.
       #   When option :book passed, returns the same object whose attributes about EPUB are set.
       #   When option :class passed, returns the instance of the class.
       #   Otherwise returns {EPUB::Book} object.
-      def parse(filepath, options = {})
+      def parse(filepath, **options)
         new(filepath, options).parse
       end
     end
 
-    def initialize(filepath, options = {})
+    def initialize(filepath, **options)
       raise "File #{filepath} not readable" unless File.readable_real? filepath
 
       @filepath = File.realpath filepath
       @book = create_book options
       @book.epub_file = @filepath
+      if options[:container_adapter]
+        adapter = options[:container_adapter]
+        @book.container_adapter = adapter
+      end
     end
 
     def parse
-      Zip::Archive.open @filepath do |zip|
-        @book.ocf = OCF.parse(zip)
-        @book.package = Publication.parse(zip, @book.ocf.container.rootfile.full_path.to_s)
+      @book.container_adapter.open @filepath do |container|
+        @book.ocf = OCF.parse(container)
+        @book.package = Publication.parse(container, @book.rootfile_path)
       end
 
       @book
@@ -64,7 +71,6 @@ module EPUB
       when params[:class]
         params[:class].new
       else
-        require 'epub/book'
         Book.new
       end
     end

data/lib/epub/parser/ocf.rb

@@ -1,5 +1,6 @@
 require 'epub/constants'
 require 'epub/ocf'
+require 'epub/ocf/physical_container'
 require 'zipruby'
 require 'nokogiri'
 
@@ -9,25 +10,24 @@ module EPUB
       include Utils
 
       DIRECTORY = 'META-INF'
-      EPUB::OCF::MODULES.each {|m| self.const_set "#{m.upcase}_FILE", "#{m}.xml"} # Deprecated
 
       class << self
-        def parse(zip_archive)
-          new(zip_archive).parse
+        def parse(container)
+          new(container).parse
         end
       end
 
-      def initialize(zip_archive)
-        @zip = zip_archive
+      def initialize(container)
+        @container = container
         @ocf = EPUB::OCF.new
       end
 
       def parse
         EPUB::OCF::MODULES.each do |m|
           begin
-            data = @zip.fopen(File.join(DIRECTORY, "#{m}.xml")) {|file| file.read}
+            data = @container.read(File.join(DIRECTORY, "#{m}.xml"))
             @ocf.__send__ "#{m}=", __send__("parse_#{m}", data)
-          rescue Zip::Error
+          rescue ::Zip::Error, ::Errno::ENOENT
           end
         end
 

data/lib/epub/parser/publication.rb

@@ -11,8 +11,8 @@ module EPUB
       include Utils
 
       class << self
-        def parse(zip_archive, file)
-          opf = zip_archive.fopen(Addressable::URI.unencode(file)) {|member| member.read}
+        def parse(container, file)
+          opf = container.read(Addressable::URI.unencode(file))
 
           new(opf, file).parse
         end

data/lib/epub/parser/version.rb

@@ -1,5 +1,5 @@
 module EPUB
   class Parser
-    VERSION = "0.1.9"
+    VERSION = "0.2.0"
   end
 end

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

@@ -105,9 +105,7 @@ module EPUB
           end
 
           def read
-            raw_content = Zip::Archive.open(manifest.package.book.epub_file) {|zip|
-              zip.fopen(entry_name) {|member| member.read}
-            }
+            raw_content = manifest.package.book.container_adapter.read(manifest.package.book.epub_file, entry_name)
 
             unless media_type.start_with?('text/') or
                 media_type.end_with?('xml') or

data/lib/epub/searcher/xhtml.rb

@@ -57,6 +57,11 @@ module EPUB
       ALGORITHMS[:restricted] = Restricted
 
       class Seamless < self
+        def initialize(element)
+          super
+          @indices = nil
+        end
+
         def search(word)
           unless @indices
             @indices, @content = build_indices(@element)

data/test/test_ocf_physical_container.rb

@@ -0,0 +1,73 @@
+# coding: utf-8
+require_relative 'helper'
+require 'epub/ocf/physical_container'
+
+class TestOCFPhysicalContainer < Test::Unit::TestCase
+  def setup
+    @container_path = 'test/fixtures/book.epub'
+    @path = 'OPS/nav.xhtml'
+    @content = File.read(File.join('test/fixtures/book', @path))
+  end
+
+  def test_read
+    assert_equal @content, EPUB::OCF::PhysicalContainer.read(@container_path, @path).force_encoding('UTF-8')
+  end
+
+  module ConcreteContainer
+    def test_class_method_open
+      @class.open @container_path do |container|
+        assert_instance_of @class, container
+        assert_equal @content, container.read(@path).force_encoding('UTF-8')
+        assert_equal File.read('test/fixtures/book/OPS/日本語.xhtml'), container.read('OPS/日本語.xhtml').force_encoding('UTF-8')
+      end
+    end
+
+    def test_class_method_read
+      assert_equal @content, @class.read(@container_path, @path).force_encoding('UTF-8')
+    end
+
+    def test_open_yields_over_container_with_opened_archive
+      @container.open do |container|
+        assert_instance_of @class, container
+      end
+    end
+
+    def test_container_in_open_block_can_readable
+      @container.open do |container|
+        assert_equal @content, container.read(@path).force_encoding('UTF-8')
+      end
+    end
+
+    def test_read
+      assert_equal @content, @container.read(@path).force_encoding('UTF-8')
+    end
+  end
+
+  class TestZipruby < self
+    include ConcreteContainer
+
+    def setup
+      super
+      @class = EPUB::OCF::PhysicalContainer::Zipruby
+      @container = @class.new(@container_path)
+    end
+  end
+
+  class TestFile < self
+    include ConcreteContainer
+
+    def setup
+      super
+      @container_path = @container_path[0..-'.epub'.length-1]
+      @class = EPUB::OCF::PhysicalContainer::File
+      @container = @class.new(@container_path)
+    end
+
+    def test_adapter_can_changable
+      adapter = EPUB::OCF::PhysicalContainer.adapter
+      EPUB::OCF::PhysicalContainer.adapter = @class
+      assert_equal @content, EPUB::OCF::PhysicalContainer.read(@container_path, @path).force_encoding('UTF-8')
+      EPUB::OCF::PhysicalContainer.adapter = adapter
+    end
+  end
+end

data/test/test_parser.rb

@@ -1,3 +1,4 @@
+# coding: utf-8
 require File.expand_path 'helper', File.dirname(__FILE__)
 
 class MyBook
@@ -25,6 +26,27 @@ class TestParser < Test::Unit::TestCase
     assert_kind_of EPUB::Book::Features, EPUB::Parser.parse('test/fixtures/book.epub', class: MyBook)
   end
 
+  def test_parse_from_file_system
+    adapter = EPUB::OCF::PhysicalContainer.adapter
+    begin
+      EPUB::OCF::PhysicalContainer.adapter = :File
+      epub = EPUB::Parser.parse('test/fixtures/book')
+      assert_instance_of EPUB::Book, epub
+      assert_equal 'Mon premier guide de cuisson, un Mémoire', epub.main_title
+    ensure
+      EPUB::OCF::PhysicalContainer.adapter = adapter
+    end
+  end
+
+  def test_can_specify_container_adapter_when_parsing_individually
+    epub = EPUB::Parser.parse('test/fixtures/book', container_adapter: :File)
+
+    assert_equal 'Mon premier guide de cuisson, un Mémoire', epub.main_title
+    assert_equal File.read('test/fixtures/book/OPS/nav.xhtml'), epub.nav.read
+    assert_equal EPUB::OCF::PhysicalContainer::File, epub.container_adapter
+    assert_equal EPUB::OCF::PhysicalContainer::Zipruby, EPUB::OCF::PhysicalContainer.adapter
+  end
+
   class TestBook < TestParser
     def setup
       super

data/test/test_parser_ocf.rb

@@ -4,13 +4,12 @@ require File.expand_path 'helper', File.dirname(__FILE__)
 class TestParserOCF < Test::Unit::TestCase
   def setup
     file = 'test/fixtures/book.epub'
-    @zip = Zip::Archive.open(file)
-    @parser = EPUB::Parser::OCF.new(@zip)
-    @container_xml = @zip.fopen('META-INF/container.xml').read
-  end
-
-  def teardown
-    @zip.close
+    EPUB::OCF::PhysicalContainer.open(file) {|container|
+      @parser = EPUB::Parser::OCF.new(container)
+    }
+    @container_xml = Zip::Archive.open(file) {|archive|
+      archive.fopen('META-INF/container.xml').read
+    }
   end
 
   def test_parsed_container_has_one_rootfile

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: epub-parser
 version: !ruby/object:Gem::Version
-  version: 0.1.9
+  version: 0.2.0
 platform: ruby
 authors:
 - KITAITI Makoto
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-06-09 00:00:00.000000000 Z
+date: 2015-06-13 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
@@ -108,20 +108,6 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
-- !ruby/object:Gem::Dependency
-  name: thin
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
 - !ruby/object:Gem::Dependency
   name: yard
   requirement: !ruby/object:Gem::Requirement
@@ -291,6 +277,7 @@ files:
 - docs/Navigation.markdown
 - docs/Publication.markdown
 - docs/Searcher.markdown
+- docs/UnpackedArchive.markdown
 - epub-parser.gemspec
 - features/epubinfo.feature
 - features/step_definitions/epubinfo_steps.rb
@@ -308,6 +295,9 @@ files:
 - lib/epub/ocf/encryption.rb
 - lib/epub/ocf/manifest.rb
 - lib/epub/ocf/metadata.rb
+- lib/epub/ocf/physical_container.rb
+- lib/epub/ocf/physical_container/file.rb
+- lib/epub/ocf/physical_container/zipruby.rb
 - lib/epub/ocf/rights.rb
 - lib/epub/ocf/signatures.rb
 - lib/epub/parser.rb
@@ -350,6 +340,7 @@ files:
 - test/test_epub.rb
 - test/test_fixed_layout.rb
 - test/test_inspect.rb
+- test/test_ocf_physical_container.rb
 - test/test_parser.rb
 - test/test_parser_content_document.rb
 - test/test_parser_fixed_layout.rb
@@ -390,6 +381,7 @@ test_files:
 - test/test_epub.rb
 - test/test_fixed_layout.rb
 - test/test_inspect.rb
+- test/test_ocf_physical_container.rb
 - test/test_parser.rb
 - test/test_parser_content_document.rb
 - test/test_parser_fixed_layout.rb