epub-parser 0.4.0 → 0.4.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: b8eeea7d5a67b4f9aaa7c71088837f4bb7156090c48e5411628e960de7b24233
-  data.tar.gz: b4d0b181946da274c66927caea49cf59c62d5507cd9543dde59bf1d3deae43d8
+  metadata.gz: db9d1db7323d581efa9dd08265ed17adc0aff3c176ef87c783968c99dc8b14b2
+  data.tar.gz: c8eed1c594124b4bf9ba8bed18562840a9199c4e8dcac234e0cbc615601c16a5
 SHA512:
-  metadata.gz: 3f9441b1dd0b6c4ac0d8acf04ff35fcdd0df4d3e880bb8a496e4a2e5127d99dc2ffc98927dd9d9907327cef08b7e8b76280f324cc1543ca38b7c80fe824fd8f1
-  data.tar.gz: 158ac997006c5ee7967132c294d6f4183dbb4dc29927653dca49e05399c63793929910beb9cd2673d8a9ea5b5175adff5f45bc712097e8b5055cb1e69df9565e
+  metadata.gz: cb742185c933cfdbee595d9d777cef56a928933838e66d68383a1cdddb330b28ce514808ef292d1d90d5885957ff0cd2e8bd7905d0319b9e8f4548bb4c23fe17
+  data.tar.gz: 4fc71e9d5a20a1a8e1c610c8c0dd389639f4fe9ad01212944f5a11ab1e29be980bb296049d3f99228db9e595183911e13cb22c6a30e6437d5aca8f6ad71621cd

data/.gitlab-ci.yml

@@ -7,7 +7,7 @@ before_script:
 
 test:2.3:
   image: ruby:2.3
-  script: bundle exec rake test
+  script: bundle exec rake test XML_BACKEND=REXML 2>/dev/null && bundle exec rake test XML_BACKEND=Oga 2>/dev/null && bundle exec rake test XML_BACKEND=Nokogiri
   except:
     - tags
   cache:
@@ -17,7 +17,7 @@ test:2.3:
 
 test:2.4:
   image: ruby:2.4
-  script: bundle exec rake test
+  script: bundle exec rake test XML_BACKEND=REXML 2>/dev/null && bundle exec rake test XML_BACKEND=Oga 2>/dev/null && bundle exec rake test XML_BACKEND=Nokogiri
   except:
     - tags
   cache:
@@ -27,7 +27,7 @@ test:2.4:
 
 test:2.5:
   image: ruby:2.5
-  script: bundle exec rake test
+  script: bundle exec rake test XML_BACKEND=REXML 2>/dev/null && bundle exec rake test XML_BACKEND=Oga 2>/dev/null && bundle exec rake test XML_BACKEND=Nokogiri
   except:
     - tags
   artifacts:
@@ -40,7 +40,7 @@ test:2.5:
 
 test:2.6:
   image: ruby:2.6
-  script: bundle exec rake test
+  script: bundle exec rake test XML_BACKEND=REXML 2>/dev/null && bundle exec rake test XML_BACKEND=Oga 2>/dev/null && bundle exec rake test XML_BACKEND=Nokogiri
   except:
     - tags
   cache:
@@ -50,6 +50,7 @@ test:2.6:
 
 pages:
   stage: deploy
+  image: ruby:2.6
   dependencies:
     - test:2.6
   script:

data/CHANGELOG.adoc

@@ -1,5 +1,9 @@
 = CHANGELOG
 
+== 0.4.1
+
+* Add Oga backend for XML document
+
 == 0.4.0
 
 * [REFACTORING]Make submodules for `XMLDocument::Refinements`

data/Gemfile

@@ -1,6 +1,12 @@
 source 'https://rubygems.org'
 gemspec
 
+if ENV['EDGE_MAKER'] == '1'
+  group :development do
+    gem 'epub-maker', path: '../epub-maker'
+  end
+end
+
 if RUBY_PLATFORM.match /darwin/
   gem 'terminal-notifier'
 end

data/README.adoc

@@ -174,6 +174,10 @@ If you find other gems, please tell me or request a pull request.
 
 == RECENT CHANGES
 
+=== 0.4.1
+
+* Add Oga backend for XML document
+
 === 0.4.0
 
 * [BUG FIX]Make epub:type a Set

data/Rakefile

@@ -5,8 +5,8 @@ require 'yard'
 require 'rdoc/task'
 require 'epub/parser/version'
 require 'archive/zip'
-require 'stringio'
 require 'epub/maker'
+require "tmpdir"
 
 task :default => :test
 task :test => 'test:default'
@@ -14,30 +14,19 @@ task :test => 'test:default'
 namespace :test do
   task :default => [:build, :test]
 
-  desc 'Run all tests'
-  task :all => [:build, :test]
-
   desc 'Build test fixture EPUB file'
-  task :build => :clean do
-    input_dir  = 'test/fixtures/book'
-    EPUB::Maker.archive input_dir
-    small_file = File.read("#{input_dir}/OPS/case-sensitive.xhtml")
-    File.open "#{input_dir}.epub" do |archive_in|
-      File.open "#{input_dir}.epub.tmp", "w" do |archive_out|
-        Archive::Zip.open archive_in, :r do |z_in|
-          Archive::Zip.open archive_out, :w do |z_out|
-            z_in.each do |entry|
-              z_out << entry
-            end
-            entry = Archive::Zip::Entry::File.new("OPS/CASE-SENSITIVE.xhtml")
-            entry.file_data = StringIO.new(small_file.sub('small file name', 'LARGE FILE NAME'))
-            z_out << entry
-          end
-        end
-      end
+  task :build => [:clean, "test/fixtures/book.epub"]
+
+  file "test/fixtures/book.epub" => "test/fixtures/book" do |task|
+    EPUB::Maker.archive task.source
+    small_file = File.read("#{task.source}/OPS/case-sensitive.xhtml")
+    Dir.mktmpdir do |dir|
+      upcase_file_path = File.join(dir, "CASE-SENSITIVE.xhtml")
+      File.write upcase_file_path, small_file.sub('small file name', 'LARGE FILE NAME')
+      Archive::Zip.archive task.name, upcase_file_path, path_prefix: "OPS"
     end
-    File.rename "#{input_dir}.epub.tmp", "#{input_dir}.epub"
   end
+  CLEAN.include "test/fixtures/book.epub"
 
   Rake::TestTask.new do |task|
     task.test_files = FileList['test/**/test_*.rb']
@@ -53,12 +42,14 @@ namespace :doc do
 
   YARD::Rake::YardocTask.new
   Rake::RDocTask.new do |rdoc|
-    rdoc.rdoc_files = FileList['lib/**/*.rb']
-    rdoc.rdoc_files.include 'README.adoc'
-    rdoc.rdoc_files.include 'CHANGELOG.adoc'
-    rdoc.rdoc_files.include 'MIT-LICENSE'
-    rdoc.rdoc_files.include 'docs/**/*.adoc'
-    rdoc.rdoc_files.include 'docs/**/*.md'
+    rdoc.rdoc_files.include %w[
+      lib/**/*.rb
+      README.adoc
+      CHANGELOG.adoc
+      MIT-LICENSE
+      docs/**/*.adoc
+      docs/**/*.md
+    ]
   end
 end
 

data/docs/Home.adoc

@@ -118,7 +118,7 @@ ret == book # => true; this API is not good I feel... Welcome suggestion!
 
 ==== Switching XML Library
 
-EPUB Parser uses https://www.nokogiri.org/[Nokogiri], a Ruby bindings for http://xmlsoft.org/[Libxml2] and http://xmlsoft.org/XSLT/[Libxslt] and more if you have already installed Nokogiri gem by RubyGems or Bundler. If Nokogiri is not available, it fallbacks to https://ruby-doc.org/stdlib-2.5.3/libdoc/rexml/rdoc/index.html[REXML], a standard-bundled library. You can also specify REXML explicitly:
+EPUB Parser tries to load https://gitlab.com/yorickpeterse/oga[Oga] a fast XML parser at first. If Oga is not available, then it tries https://www.nokogiri.org/[Nokogiri], a Ruby bindings for http://xmlsoft.org/[Libxml2] and http://xmlsoft.org/XSLT/[Libxslt] and more. If both are not available, it fallbacks to https://ruby-doc.org/stdlib-2.5.3/libdoc/rexml/rdoc/index.html[REXML], a standard-bundled library. You can also specify REXML explicitly:
 
 ----
 EPUB::Parser::XMLDocument.backend = :REXML

data/epub-parser.gemspec

@@ -40,7 +40,8 @@ Gem::Specification.new do |s|
   s.add_development_dependency 'pretty_backtrace'
   s.add_development_dependency 'epub-maker'
   s.add_development_dependency 'asciidoctor'
-  s.add_development_dependency 'nokogiri', '>= 1.6.0', '< 1.9'
+  s.add_development_dependency 'nokogiri', '>= 1.6.0', '< 1.11'
+  s.add_development_dependency 'oga', '>= 2.16'
 
   s.add_runtime_dependency 'archive-zip'
   s.add_runtime_dependency 'addressable', '>= 2.3.5'

data/lib/epub/content_document/xhtml.rb

@@ -32,6 +32,12 @@ module EPUB
         @rexml ||= REXML::Document.new(raw_document)
       end
 
+      # @return [Oga::XML::Document] content as Oga::XML::Document object
+      def oga
+        require "oga"
+        @oga ||= Oga.parse_xml(raw_document)
+      end
+
       # @return [Nokogiri::XML::Document] content as Nokogiri::XML::Document object
       def nokogiri
         require 'nokogiri'

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

@@ -35,7 +35,7 @@ module EPUB
                 end
               end
               next if index < @last_iterated_entry_index
-              # We can force encoding UTF-8 becase EPUB spec allows only UTF-8 filenames
+              # We can force encoding UTF-8 because EPUB spec allows only UTF-8 filenames
               entry_path = entry.zip_path.force_encoding('UTF-8')
               @entries[entry_path] = index
               @last_iterated_entry_index = index

data/lib/epub/parser/content_document.rb

@@ -35,13 +35,13 @@ module EPUB
         content_document
       end
 
-      # @param [XMLDocument, REXML::Document, Nokogiri::HTML::Document] document HTML document or element including nav
+      # @param [XMLDocument, REXML::Document, Oga::XML::Document, 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)
         document.each_element_by_xpath('/xhtml:html/xhtml:body//xhtml:nav', EPUB::NAMESPACES).collect {|elem| parse_navigation elem}
       end
 
-      # @param [REXML::Element, Nokogiri::XML::Element] element nav element
+      # @param [REXML::Element, Oga::XML::Element, Nokogiri::XML::Element] element nav element
       # @return [EPUB::ContentDocument::Navigation::Nav] nav Nav object
       def parse_navigation(element)
         nav = EPUB::ContentDocument::Navigation::Navigation.new
@@ -56,7 +56,7 @@ module EPUB
         nav
       end
 
-      # @param [REXML::Element, Nokogiri::XML::Element] element li element
+      # @param [REXML::Element, Oga::XML::Element, Nokogiri::XML::Element] element li element
       def parse_navigation_item(element)
         item = EPUB::ContentDocument::Navigation::Item.new
         a_or_span = element.each_element_by_xpath('./xhtml:a[1]|xhtml:span[1]', EPUB::NAMESPACES).first
@@ -90,7 +90,7 @@ module EPUB
 
       private
 
-      # @param [REXML::Element, Nokogiri::XML::Element] element nav element
+      # @param [REXML::Element, Oga::XML::Element, Nokogiri::XML::Element] element nav element
       # @return [String] heading heading text
       def find_heading(element)
         heading = element.each_element_by_xpath('./xhtml:h1|xhtml:h2|xhtml:h3|xhtml:h4|xhtml:h5|xhtml:h6|xhtml:hgroup', EPUB::NAMESPACES).first

data/lib/epub/parser/version.rb

@@ -1,5 +1,5 @@
 module EPUB
   class Parser
-    VERSION = "0.4.0"
+    VERSION = "0.4.1"
   end
 end

data/lib/epub/parser/xml_document.rb

@@ -5,7 +5,10 @@ module EPUB
         attr_accessor :backend
 
         def new(xml)
-          if backend == :Nokogiri
+          case backend
+          when :Oga
+            Oga.parse_xml(xml)
+          when :Nokogiri
             Nokogiri.XML(xml)
           else
             REXML::Document.new(xml)
@@ -16,10 +19,10 @@ module EPUB
   end
 end
 
-require "epub/parser/xml_document/refinements/rexml"
-begin
-  require "epub/parser/xml_document/refinements/nokogiri"
-  EPUB::Parser::XMLDocument.backend = :Nokogiri
-rescue LoadError
-  EPUB::Parser::XMLDocument.backend = :REXML
+%i[Nokogiri Oga REXML].each do |backend|
+  begin
+    require "epub/parser/xml_document/refinements/#{backend.downcase}"
+    EPUB::Parser::XMLDocument.backend ||= backend
+  rescue LoadError
+  end
 end

data/lib/epub/parser/xml_document/refinements/oga.rb

@@ -0,0 +1,77 @@
+require "oga"
+
+module EPUB
+  class Parser
+    class XMLDocument
+      module Refinements
+        module Oga
+          [::Oga::XML::Document, ::Oga::XML::Node].each do |klass|
+            refine klass do
+              [
+                [:document, ::Oga::XML::Document],
+                [:element, ::Oga::XML::Element],
+                [:text, ::Oga::XML::Text]
+              ].each do |(type, klass)|
+                define_method "#{type}?" do
+                  kind_of? klass
+                end
+              end
+
+              def root
+                # Couldn't use find(&:element?) for Rubies under 2.6
+                root_node.children.find {|child| child.element?}
+              end
+
+              def elements
+                # Couldn't use find(&:element?) for Rubies under 2.6
+                children.select {|child| child.element?}
+              end
+
+              # Need for Rubies under 2.6
+              def respond_to?(name, include_all = false)
+                [:root, :elements].include?(name) || super
+              end
+
+              def each_element_by_xpath(xpath, namespaces = nil, &block)
+                xpath(xpath, namespaces: namespaces).each &block
+              end
+            end
+          end
+
+          refine ::Oga::XML::Element do
+            def attribute_with_prefix(name, prefix = nil)
+              name = prefix ? "#{prefix}:#{name}" : name
+              get(name)
+            end
+
+            def each_element(xpath = nil, &block)
+              each_node do |node|
+                throw :skip_children unless node.kind_of?(::Oga::XML::Element)
+                block.call node
+              end
+            end
+
+            def namespace_uri
+              namespace&.uri
+            end
+
+            alias original_namespaces namespaces
+            def namespaces
+              original_namespaces.each_with_object({}) {|(name, namespace), nss|
+                nss[name] = namespace.uri
+              }
+            end
+
+            alias content text
+          end
+
+          refine ::Oga::XML::Text do
+            alias content text
+          end
+        end
+
+        include Oga
+      end
+    end
+  end
+end

data/lib/epub/searcher/publication.rb

@@ -42,7 +42,7 @@ module EPUB
 
       # @todo: Refactoring
       # @return [Array<Hash>] An array of rearch results. Each result is composed of:
-      #   * +:element+: [REXML::Element, Nokogiri::XML::ELement] Found element
+      #   * +:element+: [REXML::Element, Oga::XML::Element, Nokogiri::XML::ELement] Found element
       #   * +:itemref+: [EPUB::Publication::Package::Spine::Itemref] Itemref that element's document belongs to
       #   * +:location+: [EPUB::CFI::Location] CFI that indicates the element
       #   * +:package+: [EPUB::Publication::Package] Package that the element belongs to

data/lib/epub/searcher/xhtml.rb

@@ -9,7 +9,7 @@ module EPUB
       ALGORITHMS = {}
 
       class << self
-        # @param element [REXML::Element, REXML::Document, Nokogiri::XML::Element, Nokogiri::XML::Document]
+        # @param element [REXML::Element, REXML::Document, Oga::XML::ELement, Oga::XML::Document, Nokogiri::XML::Element, Nokogiri::XML::Document]
         # @param word [String]
         # @return [Array<Result>]
         def search_text(element, word)
@@ -23,7 +23,7 @@ module EPUB
       end
 
       class Restricted < self
-        # @param element [REXML::Element, Nokogiri::XML::Element]
+        # @param element [REXML::Element, Oga::XML::Element, Nokogiri::XML::Element]
         # @return [Array<Result>]
         def search_text(word, element=nil)
           results = []

data/test/helper.rb

@@ -15,3 +15,6 @@ if ENV["PRETTY_BACKTRACE"]
 end
 
 require 'epub/parser'
+if ENV["XML_BACKEND"]
+  EPUB::Parser::XMLDocument.backend = ENV["XML_BACKEND"].to_sym
+end

data/test/test_parser_publication.rb

@@ -4,7 +4,7 @@ require File.expand_path 'helper', File.dirname(__FILE__)
 class TestParserPublication < Test::Unit::TestCase
   def setup
     opf = File.read("test/fixtures/book/OPS/ルートファイル.opf")
-    @opf = Nokogiri.XML(opf)
+    @opf = EPUB::Parser::XMLDocument.new(opf)
     @parser = EPUB::Parser::Publication.new(opf)
     @package = @parser.parse_package(@opf)
   end

data/test/test_parser_xmldocument.rb

@@ -6,77 +6,81 @@ class TestParserXMLDocument < Test::Unit::TestCase
     @container = File.read("test/fixtures/book/META-INF/container.xml")
     @opf = File.read("test/fixtures/book/OPS/ルートファイル.opf")
     @nav = File.read("test/fixtures/book/OPS/nav.xhtml")
+    @backends = [:REXML, :Nokogiri, :Oga]
   end
 
   def test_parse_container
-    EPUB::Parser::XMLDocument.backend = :REXML
-    rexml_container = EPUB::Parser::OCF.new(nil).parse_container(@container)
-    EPUB::Parser::XMLDocument.backend = :Nokogiri
-    nokogiri_container = EPUB::Parser::OCF.new(nil).parse_container(@container)
-
-    assert_equal rexml_container.rootfiles.length, nokogiri_container.rootfiles.length
-    rexml_container.rootfiles.each do |rootfile|
-      rf = nokogiri_container.rootfiles.find {|rf| rf.full_path == rootfile.full_path}
-      assert_equal rootfile.full_path, rf.full_path
-      assert_equal rootfile.media_type, rf.media_type
+    assert_equal_results @backends do
+      container = EPUB::Parser::OCF.new(nil).parse_container(@container)
+      [
+        container.rootfiles.length,
+        container.rootfiles.collect {|rootfile|
+          [rootfile.full_path, rootfile.media_type]
+        }
+      ]
     end
   end
 
   def test_parse_package_document
-    EPUB::Parser::XMLDocument.backend = :REXML
-    rexml_package = EPUB::Parser::Publication.new(@opf).parse
-    EPUB::Parser::XMLDocument.backend = :Nokogiri
-    nokogiri_package = EPUB::Parser::Publication.new(@opf).parse
+    assert_equal_results @backends do
+      result = []
 
-    %i[version prefix xml_lang dir id].each do |attr|
-      assert_equal rexml_package.send(attr), nokogiri_package.send(attr)
-    end
+      package = EPUB::Parser::Publication.new(@opf).parse
 
-    EPUB::Metadata::DC_ELEMS.each do |attr|
-      rexml_attr = rexml_package.metadata.send(attr)
-      nokogiri_attr = nokogiri_package.metadata.send(attr)
+      result << %i[version prefix xml_lang dir id].collect {|attr| package.send(attr)}
 
-      assert_equal rexml_attr.length, nokogiri_attr.length
+      result << EPUB::Metadata::DC_ELEMS.collect {|attr|
+        attr = package.metadata.send(attr)
+        [
+          attr.length,
+          attr.collect {|model|
+            %i[content id lang dir].collect {|a| model.send(a)}
+          }
+        ]
+      }
 
-      rexml_attr.each_with_index do |model, index|
-        %i[content id lang dir].each do |a|
-          assert_equal model.send(a), nokogiri_attr[index].send(a)
-        end
-      end
-    end
+      result << package.metadata.metas.collect {|meta|
+        %i[property id scheme content name meta_content].collect {|attr| meta.send(attr)}
+      }
 
-    rexml_package.metadata.metas.each_with_index do |meta, index|
-      %i[property id scheme content name meta_content].each do |attr|
-        assert_equal meta.send(attr), nokogiri_package.metadata.metas[index].send(attr)
-      end
-    end
+      result << package.metadata.links.collect {|link|
+        %i[href rel id media_type].collect {|attr| link.send(attr)}
+      }
 
-    rexml_package.metadata.links.each_with_index do |link, index|
-      %i[href rel id media_type].each do |attr|
-        assert_equal link.send(attr), nokogiri_package.metadata.links[index].send(attr)
-      end
+      result
     end
   end
 
   def test_parse_navigation
-    EPUB::Parser::XMLDocument.backend = :REXML
-    rexml_nav = EPUB::Parser.parse("test/fixtures/book.epub").nav.content_document
-    EPUB::Parser::XMLDocument.backend = :Nokogiri
-    nokogiri_nav = EPUB::Parser.parse("test/fixtures/book.epub").nav.content_document
-
-    assert_equal rexml_nav.navigations.length, nokogiri_nav.navigations.length
-
-    rexml_nav.navigations.each_with_index do |item, index|
-      nokogiri_item = nokogiri_nav.navigations[index]
-      %i[text content_document type].each do |attr|
-        assert_equal item.send(attr), nokogiri_item.send(attr)
-      end
-
-      item.items.each_with_index do |i, index2|
-        %i[text content_document].each do |attr|
-          assert_equal i.send(attr), nokogiri_item.items[index2].send(attr)
-        end
-      end
+    assert_equal_results @backends do
+      result = []
+
+      nav = EPUB::Parser.parse("test/fixtures/book.epub").nav.content_document
+
+      result << nav.navigations.length
+
+      result << nav.navigations.collect {|item|
+        %i[text content_document type].collect {|attr| item.send(attr)}
+
+        item.items.collect {|i|
+          %i[text content_document].collect {|attr| i.send(attr)}
+        }
+      }
+
+      result
+    end
+  end
+
+  private
+
+  def assert_equal_results(backends)
+    results = backends.collect {|backend|
+      EPUB::Parser::XMLDocument.backend = backend
+      yield
+    }
+    expected = results.shift
+    results.each do |actual|
+      assert_equal expected, actual
     end
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: epub-parser
 version: !ruby/object:Gem::Version
-  version: 0.4.0
+  version: 0.4.1
 platform: ruby
 authors:
 - KITAITI Makoto
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2019-05-13 00:00:00.000000000 Z
+date: 2019-12-06 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
@@ -229,7 +229,7 @@ dependencies:
         version: 1.6.0
     - - "<"
       - !ruby/object:Gem::Version
-        version: '1.9'
+        version: '1.11'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
@@ -239,7 +239,21 @@ dependencies:
         version: 1.6.0
     - - "<"
       - !ruby/object:Gem::Version
-        version: '1.9'
+        version: '1.11'
+- !ruby/object:Gem::Dependency
+  name: oga
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '2.16'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '2.16'
 - !ruby/object:Gem::Dependency
   name: archive-zip
   requirement: !ruby/object:Gem::Requirement
@@ -365,6 +379,7 @@ files:
 - lib/epub/parser/version.rb
 - lib/epub/parser/xml_document.rb
 - lib/epub/parser/xml_document/refinements/nokogiri.rb
+- lib/epub/parser/xml_document/refinements/oga.rb
 - lib/epub/parser/xml_document/refinements/rexml.rb
 - lib/epub/publication.rb
 - lib/epub/publication/fixed_layout.rb