docbook_xsl_wrapper 0.1.2 → 0.5.2

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA256:
+  metadata.gz: b742b5b8c9a3e4e949bb1326bd7636849b772f56e9dc056e8598b5bcd16b5973
+  data.tar.gz: 91de30769f58801d6248f363c4c1264b50fb6e50f0faba800d12238c972ec202
+SHA512:
+  metadata.gz: e5e3629a4080d708ebab618ca1c560000f463a704655433b0f4ef1522525f49c79af08df546cf7935d1ffada325ab49cb05ffb8610ada4cae576ed9c6fa59894
+  data.tar.gz: 5a9aae5725da000feb11cd98666b7bba4bab67f2bb4a5ccff10d9973ed3a0dc6d4002227c0aaaeb4059977c7a97df6779928d87f02d4e98e52cc406852d5a76c

data/.gitignore

@@ -3,6 +3,8 @@
 .bundle
 .config
 .yardoc
+.ruby-gemset
+.ruby-version
 Gemfile.lock
 InstalledFiles
 _yardoc

data/CHANGELOG

@@ -0,0 +1,41 @@
+= Release History
+
+== 0.1.2 / 2013-04-02 Michael Cook
+
+* Now using Namespaced XSL: should produce less errors with some XML
+
+
+== 0.1.0 / 2013-03-27 Michael Cook
+
+* EPUB3 output supported
+* Some other small refactoring
+
+
+== 0.0.7 / 2013-03-05 Michael Cook
+
+* Refactored the way images are copied
+
+
+== 0.0.6 / 2013-03-05 Michael Cook
+
+* Fix bg with the OPF path in container.xml
+
+
+== 0.0.4 / 2013-03-02 Michael Cook
+
+* Bg refactor of the Epub class
+
+
+== 0.0.3 / 2013-02-27 Michael Cook
+
+* Removed development dependency on 'autotest'
+
+
+== 0.0.2 / 2013-02-23 Michael Cook
+
+* Better command-line options
+
+
+== 0.0.1 / 2013-02-22 Michael Cook
+
+* Initial gemification of the original Ruby script

data/README.md

@@ -1,5 +1,13 @@
 # DocBook XSL Wrapper
 
+**IMPORTANT NOTICE - THIS GEM IS NO LONGER UNDER DEVELOPMENT**
+
+I've been struggling to re-work this GEM to be usable as a library, rather than just as a command-line tool. On top of that I think perhaps the GEM naming is not the best.
+
+For these reasons, I'm unlikely to develop this GEM any further. I may release an alternate _Docbook to EPUB_ libary at a later date, so keep an eye on my Github account for that.
+
+====
+
 _Please consider this as an Alpha release._
 
 The DocBook XSL stylesheets are very powerful and provide an easy way to output
@@ -16,24 +24,8 @@ The original Ruby script can be found at: http://docbook.sourceforge.net/release
 ## Requirements
 
 * DocBook
-* DocBook XSL (~> v1.78.1)
-* xsltproc
-
-### NOTE
-
-The Docbook XSL Wrapper uses xsltproc, which allows all stylesheets to be
-pulled from the http://docbook.sourceforge.net... URI, but if you have the
-stylesheets installed locally, xsltproc will rewrite the URI to use local files.
-
-On my OSX Lion system, "docbook" was installed via Homebrew, but two issues
-needed fixing before everything worked correctly.
-
-The catalog file ($XML_CATALOG_FILES) needed updating;
-
-1. Add an entry to your 1.78.1 path.
-2. Remove older XSL entries (e.g. 1.76.0 & 1.77.0).
+* DocBook XSL (included in this repo)
 
-_Please make sure that xsltproc uses the *1.78.1* stylesheets as default_
 
 
 ## Installation

data/bin/docbook_xsl_wrapper

@@ -2,24 +2,19 @@
 
 $LOAD_PATH << File.expand_path(File.dirname(__FILE__) + '/../lib')
 
-require 'tmpdir'
 require 'docbook_xsl_wrapper'
 
 
 options = DocbookXslWrapper::Options.parse(ARGV)
+docbook = options.docbook
 
-unless File.exist?(options.docbook)
-  raise ArgumentError.new("File #{options.docbook} does not exist")
-end
-
-tmp_dir = Dir.mktmpdir
-options.destination = tmp_dir
+raise ArgumentError.new("File #{docbook} does not exist") unless File.exist?(docbook)
 
-begin
-  puts "Rendering DocBook file #{options.docbook} to #{options.output}\n\n" if options.verbose
-
-  epub = DocbookXslWrapper::Epub.new(options)
-  epub.create
-ensure
-  FileUtils.remove_entry_secure tmp_dir
+doc = DocbookXslWrapper::Validate.new(File.open(docbook, 'rb'))
+unless doc.valid?
+  puts "YOUR DOCBOOK XML DOES NOT SEEM TO BE VALID - PLEASE FIX."
+  puts
+  puts doc.errors
 end
+
+DocbookXslWrapper.build_epub(options)

data/docbook_xsl_wrapper.gemspec

@@ -1,13 +1,14 @@
-# -*- encoding: utf-8 -*-
 require File.expand_path('../lib/docbook_xsl_wrapper/version', __FILE__)
 
 Gem::Specification.new do |gem|
   gem.name          = "docbook_xsl_wrapper"
   gem.version       = DocbookXslWrapper::VERSION
-  gem.authors       = ["Mike Cook"]
-  gem.email         = ["m@mikecook.co.uk"]
-  gem.summary       = %q{Wrapper for the DocBook XSL stylesheets for easy XML to EPUB}
-  gem.description   = %q{DocBook XSL Wrapper let's you easily convert DocBook XML to EPUB using the official DocBook XSL stylesheets.}
+  gem.authors       = ["Michael R. Cook"]
+  gem.email         = ["git@mrcook.uk"]
+  gem.summary       = %q{DocBook XSL Wrapper: validate/transform DocBook 5.x XML documents}
+  gem.description   = %q{DocBook XSL Wrapper let's you easily validate then convert DocBook 5.x XML documents to EPUB using the official DocBook XSL stylesheets.}
+  gem.license       = 'MIT'
+  gem.homepage      = 'https://mrcook.uk'
 
   gem.platform      = Gem::Platform::RUBY
   gem.files         = `git ls-files`.split($\)
@@ -15,7 +16,7 @@ Gem::Specification.new do |gem|
   gem.executables   = gem.files.grep(%r{^bin/}).map{ |f| File.basename(f) }
   gem.test_files    = gem.files.grep(%r{^(test|spec|features)/})
 
-  gem.required_ruby_version = '~> 1.9.3'
+  gem.add_dependency 'nokogiri', '~> 1.6', '>= 1.6.1'
 
-  gem.add_development_dependency('rspec', '~> 2.13.0')
+  gem.add_development_dependency 'rspec', '~> 2.14', '>= 2.14.1'
 end

data/lib/docbook_xsl_wrapper.rb

@@ -1,4 +1,27 @@
-require "docbook_xsl_wrapper/version"
+require 'tmpdir'
+require 'nokogiri'
+
 require 'docbook_xsl_wrapper/epub'
 require 'docbook_xsl_wrapper/options'
+require 'docbook_xsl_wrapper/validate'
+
+require "docbook_xsl_wrapper/version"
+
+module DocbookXslWrapper
+  GEM_PATH = File.expand_path(File.join('..', '..'), __FILE__)
+
+  def self.build_epub(options)
+    tmp_dir = Dir.mktmpdir
+    options.destination = tmp_dir
+
+    begin
+      puts "Rendering DocBook file #{options.docbook} to #{options.output}\n\n" if options.verbose
+
+      epub = Epub.new(options)
+      epub.create
+    ensure
+      FileUtils.remove_entry_secure tmp_dir
+    end
+  end
 
+end

data/lib/docbook_xsl_wrapper/epub.rb

@@ -4,184 +4,154 @@ require 'rexml/parsers/pullparser'
 module DocbookXslWrapper
 
   class Epub
-    attr_reader :options
+    attr_reader :options, :xml
 
     def initialize(options)
       @options = options
-
-      case options.format
-      when 'epub3'
-        xsl = File.join('epub3', 'chunk.xsl')
-      else
-        xsl = File.join('epub', 'docbook.xsl')
-      end
-      official_docbook_xsl = File.join('http://docbook.sourceforge.net/release/xsl-ns/current', xsl)
-
-      @options.stylesheet = official_docbook_xsl unless options.stylesheet
+      @xml     = Nokogiri::XML(File.open(@options.docbook, 'rb'))
     end
 
     def create
-      begin
-        render_to_epub
-        bundle_epub
-      ensure
-        FileUtils.remove_entry_secure @collapsed_docbook_file
-      end
+      # Nokogiri doesn't create directories, so we do it manually
+      Dir.mkdir(File.join(options.destination, meta_inf_directory))
+      Dir.mkdir(File.join(options.destination, oebps_directory))
+
+      render_to_epub
+      bundle_epub
     end
 
   private
 
     def render_to_epub
-      @collapsed_docbook_file = collapse_docbook
+      errors = xslt_transform_and_rescue_because_it_currently_throws_unknown_runtime_error
+      raise "Could not render as .epub to #{options.output}\n\n(#{errors})" unless errors.empty?
+    end
 
-      # Double-quote stylesheet & file to help Windows cmd.exe
-      db2epub_cmd = %Q(cd "#{options.destination}" && xsltproc #{xsl_parser_options} "#{options.stylesheet}" "#{@collapsed_docbook_file}")
-      STDERR.puts db2epub_cmd if options.debug
-      success = system(db2epub_cmd)
-      raise "Could not render as .epub to #{options.output} (#{db2epub_cmd})" unless success
-    end
-
-    def xsl_parser_options
-      chunk_quietly = "--stringparam chunk.quietly 1" if options.verbose == false
-      css           = "--stringparam html.stylesheet #{File.basename(options.css)}/" if options.css
-      base          = "--stringparam base.dir #{oebps_path}/"
-      unless options.fonts.empty?
-        fonts = options.fonts.map {|f| File.basename(f)}.join(',')
-        font  = "--stringparam epub.embedded.fonts \"#{fonts}\""
+    def xslt_transform_and_rescue_because_it_currently_throws_unknown_runtime_error
+      begin
+        errors = stylesheet.transform(xml, params)
+      rescue
+        errors = ''
       end
-      meta  = "--stringparam epub.metainf.dir #{meta_inf_directory}/"
-      oebps = "--stringparam epub.oebps.dir #{oebps_directory}/"
+      errors
+    end
+
+    def stylesheet
+      xsl = docbook_xsl_path
+      xsl = options.stylesheet unless options.stylesheet.empty?
+
+      Nokogiri::XSLT(File.open(xsl, 'rb'))
+    end
 
-      [
-        chunk_quietly,
-        base,
-        font,
-        meta,
-        oebps,
-        css,
-      ].join(" ")
+    def docbook_xsl_path
+      case options.format
+      when 'epub3'
+        File.join(GEM_PATH, 'xsl', 'epub3', 'chunk.xsl')
+      else
+        File.join(GEM_PATH, 'xsl', 'epub', 'docbook.xsl')
+      end
     end
 
-    def collapse_docbook
-      # Input must be collapsed because REXML couldn't find figures in files that
-      # were XIncluded or added by ENTITY
-      #   http://sourceforge.net/tracker/?func=detail&aid=2750442&group_id=21935&atid=373747
+    def params
+      params_list = [
+        'chunk.quietly', "#{verbosity}",
+        'chunk.first.sections', 1,
+        'othercredit.like.author.enabled', 1,
+        'chapter.autolabel', 0,
+        'section.autolabel', 0,
+        'part.autolabel', 0,
+        'base.dir', File.join(options.destination, '/'),
+      ]
+      params_list.concat(css) if options.css
+      params_list.concat(fonts) unless options.fonts.empty?
 
-      # Double-quote stylesheet & file to help Windows cmd.exe
-      collapsed_file = File.join(File.expand_path(File.dirname(options.docbook)),
-                                 '.collapsed.' + File.basename(options.docbook))
-      entity_collapse_command = %Q(xmllint --loaddtd --noent -o "#{collapsed_file}" "#{options.docbook}")
-      entity_success = system(entity_collapse_command)
-      raise "Could not collapse named entites in #{options.docbook}" unless entity_success
+      Nokogiri::XSLT.quote_params(params_list)
+    end
 
-      xinclude_collapse_command = %Q(xmllint --xinclude -o "#{collapsed_file}" "#{collapsed_file}")
-      xinclude_success = system(xinclude_collapse_command)
-      raise "Could not collapse XIncludes in #{options.docbook}" unless xinclude_success
+    def fonts
+      ['epub.embedded.fonts', options.fonts.map {|f| File.basename(f)}.join(',')]
+    end
 
-      collapsed_file
+    def css
+      ['html.stylesheet', File.join(File.basename(options.css), '/')]
     end
 
     def bundle_epub
-      quiet = options.verbose ? "" : "-q"
-      mimetype_filename = write_mimetype
-      images = copy_images
-      csses  = copy_css
-      fonts  = copy_fonts
-      callouts = copy_callouts
-      # zip -X -r ../book.epub mimetype META-INF OEBPS
+      copy_media_files_to_epub_dir
+      create_mimetype_file if options.format == 'epub' # EPUB3 stylesheet creates this automatically
+
+      quiet = options.verbose ? '' : '-q'
       # Double-quote stylesheet & file to help Windows cmd.exe
-      zip_cmd = %Q(cd "#{options.destination}" && zip #{quiet} -X -r  "#{File.expand_path(options.output)}" "#{mimetype_filename}" "#{meta_inf_directory}" "#{oebps_directory}")
+      zip_cmd = %Q(cd "#{options.destination}" && zip #{quiet} -X -r  "#{File.expand_path(options.output)}" "mimetype" "#{meta_inf_directory}" "#{oebps_directory}")
+
       puts zip_cmd if options.debug
       success = system(zip_cmd)
       raise "Could not bundle into .epub file to #{options.output}" unless success
     end
 
-    def write_mimetype
+    def create_mimetype_file
       filename = File.join(options.destination, "mimetype")
       File.open(filename, "w") {|f| f.print "application/epub+zip"}
       File.basename(filename)
     end
 
-    def copy_callouts
-      return unless has_callouts?
-
-      images = Array.new
-      calloutglob = "#{options.callout_full_path}/*#{options.callout_ext}"
-      Dir.glob(calloutglob).each do |image|
-        new_filename = File.join(oebps_path, options.callout_path, File.basename(image))
-
-        # TODO: What to rescue for these two?
-        FileUtils.mkdir_p(File.dirname(new_filename))
-        FileUtils.cp(image, new_filename)
-        images << image
-      end
-      images
-    end
+    def copy_media_files_to_epub_dir
+      copy_fonts
+      copy_css
+      copy_images
 
-    # Returns true if the document has code callouts
-    def has_callouts?
-      parser = REXML::Parsers::PullParser.new(File.new(@collapsed_docbook_file))
-      while parser.has_next?
-        element = parser.pull
-        return true if element.start_element? and (element[0] == "calloutlist" or element[0] == "co")
-      end
-      false
+      # Callouts disabled in this release until more testing can be done
+      #copy_callouts
     end
 
     def copy_fonts
-      fonts = Array.new
-      options.fonts.each {|font|
-        new_filename = File.join(oebps_path, File.basename(font))
-        FileUtils.cp(font, new_filename)
-        fonts << font
-      }
-      fonts
+      return if options.fonts.empty?
+
+      font_directory = File.join(options.destination, oebps_directory, 'fonts')
+      Dir.mkdir(font_directory)
+
+      options.fonts.each do |font|
+        FileUtils.cp(font, File.join(font_directory, File.basename(font)))
+      end
     end
 
     def copy_css
       return unless options.css
 
-      filename = File.join(oebps_path, File.basename(options.css))
-      FileUtils.cp(options.css, filename)
+      FileUtils.cp(options.css, File.join(options.destination, oebps_directory, File.basename(options.css)))
     end
 
     def copy_images
-      images = Array.new
       xml_image_references.each do |image|
-        images << copy_image(image)
+        copy_image(image)
       end
-      images
-    end
-
-    def copy_image(image)
-      source_file      = File.join(File.expand_path(File.dirname(options.docbook)), image)
-      destination_file = File.join(oebps_path, image)
-
-      FileUtils.mkdir_p(File.dirname(destination_file))
-
-      puts "Copying image: #{source_file} to #{destination_file}" if options.debug
-      FileUtils.cp(source_file, destination_file)
-
-      destination_file
     end
 
     def xml_image_references
-      parser = REXML::Parsers::PullParser.new(File.new(@collapsed_docbook_file))
-      references = Array.new
-      while parser.has_next?
-        element = parser.pull
-        references << element[1]['fileref'] if is_valid_image_reference?(element)
+      refs = Array.new
+      xml.xpath('//xmlns:imagedata', '//xmlns:graphic', 'xmlns' => 'http://docbook.org/ns/docbook').each do |node|
+        img = node.attribute('fileref').value
+        refs << img if is_valid_image?(img)
       end
-      references.uniq
+
+      refs.uniq
     end
 
-    def is_valid_image_reference?(element)
-      return false unless element.start_element?
-      return false unless element[0] == 'graphic' or element[0] == 'imagedata'
-      return true if element[1]['fileref'].match(/\.(jpe?g|png|gif|svg|xml)\Z/i)
+    def is_valid_image?(image)
+      return true if File.extname(image).match(/\.(jpe?g|png|gif|svg|xml)\z/i)
       false
     end
 
+    def copy_image(image)
+      source      = File.join(File.dirname(options.docbook), image)
+      destination = File.join(options.destination, oebps_directory, image)
+
+      FileUtils.mkdir_p(File.dirname(destination))
+
+      puts "Copying image: #{source} to #{destination}" if options.debug
+      FileUtils.cp(source, destination)
+    end
+
     def oebps_directory
       'OEBPS'
     end
@@ -194,5 +164,37 @@ module DocbookXslWrapper
       'META-INF'
     end
 
+    def verbosity
+      return 0 if options.verbose == true
+      return 1
+    end
+
+
+    # TODO: This method is not being called for the moment....needs to be tested
+    def copy_callouts
+      return unless has_callouts?
+
+      images = Array.new
+      calloutglob = "#{options.callout_full_path}/*#{options.callout_ext}"
+      Dir.glob(calloutglob).each do |image|
+        new_filename = File.join(oebps_path, options.callout_path, File.basename(image))
+
+        # TODO: What to rescue for these two?
+        FileUtils.mkdir_p(File.dirname(new_filename))
+        FileUtils.cp(image, new_filename)
+        images << image
+      end
+      images
+    end
+
+    def has_callouts?
+      parser = REXML::Parsers::PullParser.new(File.new(@collapsed_docbook_file))
+      while parser.has_next?
+        element = parser.pull
+        return true if element.start_element? and (element[0] == "calloutlist" or element[0] == "co")
+      end
+      false
+    end
+
   end
 end