# frozen_string_literal: true require 'mime/types' require 'open3' require 'sass' require_relative 'font_icon_map' module Asciidoctor module Epub3 # Public: The main converter for the epub3 backend that handles packaging the EPUB3 publication file. class Converter include ::Asciidoctor::Converter include ::Asciidoctor::Logging include ::Asciidoctor::Writer register_for 'epub3' def write(output, target) output.generate_epub target logger.debug %(Wrote to #{target}) if @extract extract_dir = target.sub EPUB_EXTENSION_RX, '' ::FileUtils.remove_dir extract_dir if ::File.directory? extract_dir ::Dir.mkdir extract_dir ::Dir.chdir extract_dir do ::Zip::File.open target do |entries| entries.each do |entry| next unless entry.file? unless (entry_dir = ::File.dirname entry.name) == '.' || (::File.directory? entry_dir) ::FileUtils.mkdir_p entry_dir end entry.extract entry.name end end end logger.debug %(Extracted to #{extract_dir}) end return unless @validate validate_epub target end CSV_DELIMITED_RX = /\s*,\s*/.freeze DATA_DIR = ::File.expand_path ::File.join(__dir__, '..', '..', 'data') IMAGE_MACRO_RX = /^image::?(.*?)\[(.*?)\]$/.freeze IMAGE_SRC_SCAN_RX = /}.freeze TRAILING_PUNCT_RX = /[[:punct:]]$/.freeze FROM_HTML_SPECIAL_CHARS_MAP = { '<' => '<', '>' => '>', '&' => '&' }.freeze FROM_HTML_SPECIAL_CHARS_RX = /(?:#{FROM_HTML_SPECIAL_CHARS_MAP.keys * '|'})/.freeze TO_HTML_SPECIAL_CHARS_MAP = { '&' => '&', '<' => '<', '>' => '>' }.freeze TO_HTML_SPECIAL_CHARS_RX = /[#{TO_HTML_SPECIAL_CHARS_MAP.keys.join}]/.freeze EPUB_EXTENSION_RX = /\.epub$/i.freeze QUOTE_TAGS = begin tags = { monospaced: ['', '', true], emphasis: ['', '', true], strong: ['', '', true], double: ['“', '”'], single: ['‘', '’'], mark: ['', '', true], superscript: ['', '', true], subscript: ['', '', true], asciimath: ['', '', true], latexmath: ['', '', true] } tags.default = ['', ''] tags.freeze end def initialize(backend, opts = {}) super basebackend 'html' outfilesuffix '.epub' htmlsyntax 'xml' end def convert(node, name = nil, _opts = {}) method_name = %(convert_#{name ||= node.node_name}) if respond_to? method_name send method_name, node else logger.warn %(conversion missing in backend #{@backend} for #{name}) nil end end # @param node [Asciidoctor::AbstractNode] # @return [String, nil] def get_chapter_filename(node) node.id if node.chapter? end def get_numbered_title(node) doc_attrs = node.document.attributes level = node.level if node.caption node.captioned_title elsif node.respond_to?(:numbered) && node.numbered && level <= (doc_attrs['sectnumlevels'] || 3).to_i if level < 2 && node.document.doctype == 'book' case node.sectname when 'chapter' %(#{(signifier = doc_attrs['chapter-signifier']) ? "#{signifier} " : ''}#{node.sectnum} #{node.title}) when 'part' %(#{(signifier = doc_attrs['part-signifier']) ? "#{signifier} " : ''}#{node.sectnum nil, ':'} #{node.title}) else %(#{node.sectnum} #{node.title}) end else %(#{node.sectnum} #{node.title}) end else node.title end end def icon_names @icon_names ||= [] end def convert_document(node) @validate = node.attr? 'ebook-validate' @extract = node.attr? 'ebook-extract' @compress = node.attr 'ebook-compress' @epubcheck_path = node.attr 'ebook-epubcheck-path' @xrefs_seen = ::Set.new @media_files = {} @footnotes = [] @book = GEPUB::Book.new 'EPUB/package.opf' @book.epub_backward_compat = true @book.language node.attr('lang', 'en'), id: 'pub-language' if node.attr? 'uuid' @book.primary_identifier node.attr('uuid'), 'pub-identifier', 'uuid' else @book.primary_identifier node.id, 'pub-identifier', 'uuid' end # replace with next line once the attributes argument is supported # unique_identifier doc.id, 'pub-id', 'uuid', 'scheme' => 'xsd:string' # NOTE: we must use :plain_text here since gepub reencodes @book.add_title sanitize_doctitle_xml(node, :plain_text), id: 'pub-title' # see https://www.w3.org/publishing/epub3/epub-packages.html#sec-opf-dccreator (1..(node.attr 'authorcount', 1).to_i).map do |idx| author = node.attr(idx == 1 ? 'author' : %(author_#{idx})) @book.add_creator author, role: 'aut' unless author.nil_or_empty? end publisher = node.attr 'publisher' # NOTE: Use producer as both publisher and producer if publisher isn't specified publisher = node.attr 'producer' if publisher.nil_or_empty? @book.publisher = publisher unless publisher.nil_or_empty? if node.attr? 'reproducible' # We need to set lastmodified to some fixed value. Otherwise, gepub will set it to current date. @book.lastmodified = (::Time.at 0).utc # Is it correct that we do not populate dc:date when 'reproducible' is set? else if node.attr? 'revdate' begin @book.date = node.attr 'revdate' rescue ArgumentError => e logger.error %(#{::File.basename node.attr('docfile')}: failed to parse revdate: #{e}) @book.date = node.attr 'docdatetime' end else @book.date = node.attr 'docdatetime' end @book.lastmodified = node.attr 'localdatetime' end @book.description = node.attr 'description' if node.attr? 'description' @book.source = node.attr 'source' if node.attr? 'source' @book.rights = node.attr 'copyright' if node.attr? 'copyright' (node.attr 'keywords', '').split(CSV_DELIMITED_RX).each do |s| @book.metadata.add_metadata 'subject', s end if node.attr? 'series-name' series_name = node.attr 'series-name' series_volume = node.attr 'series-volume', 1 series_id = node.attr 'series-id' series_meta = @book.metadata.add_metadata 'meta', series_name, id: 'pub-collection', group_position: series_volume series_meta['property'] = 'belongs-to-collection' series_meta.refine 'dcterms:identifier', series_id unless series_id.nil? # Calibre only understands 'series' series_meta.refine 'collection-type', 'series' end # For list of supported landmark types see # https://idpf.github.io/epub-vocabs/structure/ landmarks = [] front_cover = add_cover_page node, 'front-cover' if front_cover.nil? && node.doctype == 'book' # TODO(#352): add textual front cover similar to PDF end landmarks << { type: 'cover', href: front_cover.href, title: 'Front Cover' } unless front_cover.nil? front_matter_page = add_front_matter_page node unless front_matter_page.nil? landmarks << { type: 'frontmatter', href: front_matter_page.href, title: 'Front Matter' } end nav_item = @book.add_item('nav.xhtml', id: 'nav').nav toclevels = [(node.attr 'toclevels', 1).to_i, 0].max outlinelevels = [(node.attr 'outlinelevels', toclevels).to_i, 0].max if node.attr? 'toc' toc_item = @book.add_ordered_item 'toc.xhtml', id: 'toc' landmarks << { type: 'toc', href: toc_item.href, title: node.attr('toc-title') } else toc_item = nil end if node.doctype == 'book' toc_items = node.sections node.content else toc_items = [node] add_chapter node end _back_cover = add_cover_page node, 'back-cover' # TODO: add landmark for back cover? But what epub:type? unless toc_items.empty? landmarks << { type: 'bodymatter', href: %(#{get_chapter_filename toc_items[0]}.xhtml), title: 'Start of Content' } end toc_items.each do |item| next unless %w[appendix bibliography glossary index preface].include? item.style landmarks << { type: item.style, href: %(#{get_chapter_filename item}.xhtml), title: item.title } end nav_item.add_content nav_doc(node, toc_items, landmarks, outlinelevels).to_ios # User is not supposed to see landmarks, so pass empty array here toc_item&.add_content nav_doc(node, toc_items, [], toclevels).to_ios # NOTE: gepub doesn't support building a ncx TOC with depth > 1, so do it ourselves toc_ncx = ncx_doc node, toc_items, outlinelevels @book.add_item 'toc.ncx', content: toc_ncx.to_ios, id: 'ncx' docimagesdir = (node.attr 'imagesdir', '.').chomp '/' docimagesdir = (docimagesdir == '.' ? nil : %(#{docimagesdir}/)) @media_files.each do |name, file| if name.start_with? %(#{docimagesdir}jacket/cover.) logger.warn %(path is reserved for cover artwork: #{name}; skipping file found in content) elsif file[:path].nil? || File.readable?(file[:path]) mime_types = MIME::Types.type_for name mime_types.delete_if { |x| x.media_type != file[:media_type] } preferred_mime_type = mime_types.empty? ? nil : mime_types[0].content_type @book.add_item name, content: file[:path], media_type: preferred_mime_type else logger.error %(#{File.basename node.attr('docfile')}: media file not found or not readable: #{file[:path]}) end end # add_metadata 'ibooks:specified-fonts', true add_theme_assets node if node.doctype != 'book' usernames = [node].map { |item| item.attr 'username' }.compact.uniq add_profile_images node, usernames end @book end # FIXME: move to Asciidoctor::Helpers def sanitize_doctitle_xml(doc, content_spec) doctitle = doc.doctitle use_fallback: true sanitize_xml doctitle, content_spec end # FIXME: move to Asciidoctor::Helpers def sanitize_xml(content, content_spec) if content_spec != :pcdata && (content.include? '<') && ((content = (content.gsub XML_ELEMENT_RX, '').strip).include? ' ') content = content.tr_s ' ', ' ' end case content_spec when :attribute_cdata content = content.gsub '"', '"' if content.include? '"' when :cdata, :pcdata # noop when :plain_text if content.include? ';' content = content.gsub(CHAR_ENTITY_RX) { [::Regexp.last_match(1).to_i].pack 'U*' } if content.include? '&#' content = content.gsub FROM_HTML_SPECIAL_CHARS_RX, FROM_HTML_SPECIAL_CHARS_MAP end else raise ::ArgumentError, %(Unknown content spec: #{content_spec}) end content end # @param node [Asciidoctor::AbstractBlock] def add_chapter(node) filename = get_chapter_filename node return nil if filename.nil? chapter_item = @book.add_ordered_item %(#{filename}.xhtml) doctitle = node.document.doctitle partition: true, use_fallback: true chapter_title = doctitle.combined if node.context == :document && doctitle.subtitle? title = %(#{doctitle.main} ) subtitle = doctitle.subtitle elsif node.title # HACK: until we get proper handling of title-only in CSS title = '' subtitle = get_numbered_title node chapter_title = subtitle else title = nil subtitle = nil end if node.document.doctype == 'book' byline = '' else author = node.attr 'author' username = node.attr 'username', 'default' imagesdir = (node.document.attr 'imagesdir', '.').chomp '/' imagesdir = imagesdir == '.' ? '' : %(#{imagesdir}/) byline = %(

#{author}

#{LF}) end mark_last_paragraph node unless node.document.doctype == 'book' @xrefs_seen.clear content = node.content # NOTE: must run after content is resolved # TODO perhaps create dynamic CSS file? if icon_names.empty? icon_css_head = '' else icon_defs = icon_names.map { |name| %(.i-#{name}::before { content: "#{FontIconMap.unicode name}"; }) } * LF icon_css_head = %( ) end header = if title || subtitle %(
#{byline}

#{title}#{subtitle ? %(#{subtitle}) : ''}

) else '' end # We want highlighter CSS to be stored in a separate file # in order to avoid style duplication across chapter files linkcss = true lines = [%( #{chapter_title} #{icon_css_head})] syntax_hl = node.document.syntax_highlighter epub_type_attr = node.respond_to?(:section) && node.sectname != 'section' ? %( epub:type="#{node.sectname}") : '' if syntax_hl&.docinfo? :head lines << (syntax_hl.docinfo :head, node, linkcss: linkcss, self_closing_tag_slash: '/') end lines << %(
#{header} #{content}) unless (fns = node.document.footnotes - @footnotes).empty? @footnotes += fns lines << '' end lines << '
' if syntax_hl&.docinfo? :footer lines << (syntax_hl.docinfo :footer, node.document, linkcss: linkcss, self_closing_tag_slash: '/') end lines << ' ' chapter_item.add_content((lines * LF).to_ios) epub_properties = node.attr 'epub-properties' chapter_item.add_property 'svg' if epub_properties&.include? 'svg' # # QUESTION reenable? # #linear 'yes' if i == 0 chapter_item end # @param node [Asciidoctor::Section] def convert_section(node) return unless add_chapter(node).nil? hlevel = node.level.clamp 1, 6 epub_type_attr = node.sectname == 'section' ? '' : %( epub:type="#{node.sectname}") div_classes = [%(sect#{node.level}), node.role].compact title = get_numbered_title node %(
#{title}#{if (content = node.content).empty? '' else %( #{content}) end}
) end # NOTE: embedded is used for AsciiDoc table cell content def convert_embedded(node) node.content end # TODO: support use of quote block as abstract def convert_preamble(node) return unless add_chapter(node).nil? if ((first_block = node.blocks[0]) && first_block.style == 'abstract') || # REVIEW: should we treat the preamble as an abstract in general? (first_block && node.blocks.size == 1) convert_abstract first_block else node.content end end def convert_open(node) id_attr = node.id ? %( id="#{node.id}") : nil class_attr = node.role ? %( class="#{node.role}") : nil if id_attr || class_attr %( #{output_content node} ) else output_content node end end def convert_abstract(node) %(
#{output_content node}
) end def convert_paragraph(node) id_attr = node.id ? %( id="#{node.id}") : '' role = node.role # stack-head is the alternative to the default, inline-head (where inline means "run-in") head_stop = node.attr 'head-stop', (role && (node.has_role? 'stack-head') ? nil : '.') head = if node.title? %(#{title = node.title}#{head_stop && title !~ TRAILING_PUNCT_RX ? head_stop : ''} ) else '' end if role node.set_option 'hardbreaks' if node.has_role? 'signature' %(#{head}#{node.content}

) else %(#{head}#{node.content}

) end end def convert_pass(node) content = node.content if content == '' '
' else content end end def convert_admonition(node) id_attr = node.id ? %( id="#{node.id}") : '' if node.title? title = node.title title_sanitized = xml_sanitize title title_attr = %( title="#{node.caption}: #{title_sanitized}") title_el = %(

#{title}

) else title_attr = %( title="#{node.caption}") title_el = '' end type = node.attr 'name' epub_type = case type when 'tip' 'tip' when 'important', 'warning', 'caution', 'note' 'notice' else logger.warn %(unknown admonition type: #{type}) 'notice' end %( #{title_el}
#{output_content node}
) end def convert_example(node) id_attr = node.id ? %( id="#{node.id}") : '' title_div = if node.title? %(
#{node.title}
) else '' end %( #{title_div}
#{output_content node}
) end def convert_floating_title(node) tag_name = %(h#{node.level + 1}) id_attribute = node.id ? %( id="#{node.id}") : '' %(<#{tag_name}#{id_attribute} class="#{['discrete', node.role].compact * ' '}">#{node.title}) end # @param node [Asciidoctor::Block] def convert_listing(node) id_attribute = node.id ? %( id="#{node.id}") : '' nowrap = (node.option? 'nowrap') || !(node.document.attr? 'prewrap') if node.style == 'source' lang = node.attr 'language' syntax_hl = node.document.syntax_highlighter if syntax_hl opts = if syntax_hl.highlight? { css_mode: ((doc_attrs = node.document.attributes)[%(#{syntax_hl.name}-css)] || :class).to_sym, style: doc_attrs[%(#{syntax_hl.name}-style)] } else {} end opts[:nowrap] = nowrap else pre_open = %(
)
            pre_close = '
' end else pre_open = %() pre_close = '' syntax_hl = nil end figure_classes = ['listing'] figure_classes << 'coalesce' if node.option? 'unbreakable' title_div = node.title? ? %(
#{node.captioned_title}
) : '' %(#{title_div} #{syntax_hl ? (syntax_hl.format node, lang, opts) : pre_open + (node.content || '') + pre_close} ) end def convert_stem(node) return convert_listing node if node.style != 'asciimath' || !asciimath_available? id_attr = node.id ? %( id="#{node.id}") : '' title_element = node.title? ? %(
#{node.captioned_title}
) : '' equation_data = AsciiMath.parse(node.content).to_mathml 'mml:' %( #{title_element}
#{equation_data}
) end def asciimath_available? (@asciimath_status ||= load_asciimath) == :loaded end def load_asciimath Helpers.require_library('asciimath', true, :warn).nil? ? :unavailable : :loaded end def convert_literal(node) id_attribute = node.id ? %( id="#{node.id}") : '' title_element = node.title? ? %(
#{node.captioned_title}
) : '' %( #{title_element}
#{node.content}
) end def convert_page_break(_node) '
' end def convert_thematic_break(_node) '
' end def convert_quote(node) id_attr = node.id ? %( id="#{node.id}") : '' class_attr = (role = node.role) ? %( class="blockquote #{role}") : ' class="blockquote"' footer_content = [] if (attribution = node.attr 'attribution') footer_content << attribution end if (citetitle = node.attr 'citetitle') citetitle_sanitized = xml_sanitize citetitle footer_content << %(#{citetitle}) end footer_content << %(#{node.title}) if node.title? footer_tag = if footer_content.empty? '' else %(
~ #{footer_content * ' '}
) end content = (output_content node).strip %(
#{content}#{footer_tag}
) end def convert_verse(node) id_attr = node.id ? %( id="#{node.id}") : '' class_attr = (role = node.role) ? %( class="verse #{role}") : ' class="verse"' footer_content = [] if (attribution = node.attr 'attribution') footer_content << attribution end if (citetitle = node.attr 'citetitle') citetitle_sanitized = xml_sanitize citetitle footer_content << %(#{citetitle}) end footer_tag = if footer_content.empty? '' else %( ~ #{footer_content * ', '}) end %(
#{node.content}#{footer_tag}
) end def convert_sidebar(node) classes = ['sidebar'] if node.title? classes << 'titled' title = node.title title_sanitized = xml_sanitize title title_attr = %( title="#{title_sanitized}") title_el = %(

#{title}

) else title_attr = title_el = '' end %() end def convert_table(node) lines = [%(
)] lines << %(
) table_id_attr = node.id ? %( id="#{node.id}") : '' table_classes = [ 'table', %(table-framed-#{node.attr 'frame', 'rows', 'table-frame'}), %(table-grid-#{node.attr 'grid', 'rows', 'table-grid'}) ] if (role = node.role) table_classes << role end if (float = node.attr 'float') table_classes << float end table_styles = [] if (autowidth = node.option? 'autowidth') && !(node.attr? 'width') table_classes << 'fit-content' else table_styles << %(width: #{node.attr 'tablepcwidth'}%;) end table_class_attr = %( class="#{table_classes * ' '}") table_style_attr = table_styles.empty? ? '' : %( style="#{table_styles * '; '}") lines << %() lines << %(#{node.captioned_title}) if node.title? if (node.attr 'rowcount').positive? lines << '' if autowidth lines += (Array.new node.columns.size, %()) else node.columns.each do |col| lines << (col.option?('autowidth') ? %() : %()) end end lines << '' %i[head body foot].reject { |tsec| node.rows[tsec].empty? }.each do |tsec| lines << %() node.rows[tsec].each do |row| lines << '' row.each do |cell| if tsec == :head cell_content = cell.text else case cell.style when :asciidoc cell_content = %(
#{cell.content}
) when :verse cell_content = %(
#{cell.text}
) when :literal cell_content = %(
#{cell.text}
) else cell_content = '' cell.content.each do |text| cell_content = %(#{cell_content}

#{text}

) end end end cell_tag_name = tsec == :head || cell.style == :header ? 'th' : 'td' cell_classes = [ "halign-#{cell.attr 'halign'}", "valign-#{cell.attr 'valign'}" ] cell_class_attr = cell_classes.empty? ? '' : %( class="#{cell_classes * ' '}") cell_colspan_attr = cell.colspan ? %( colspan="#{cell.colspan}") : '' cell_rowspan_attr = cell.rowspan ? %( rowspan="#{cell.rowspan}") : '' cell_style_attr = node.document.attr?('cellbgcolor') ? %( style="background-color: #{node.document.attr 'cellbgcolor'}") : '' lines << %(<#{cell_tag_name}#{cell_class_attr}#{cell_colspan_attr}#{cell_rowspan_attr}#{cell_style_attr}>#{cell_content}) end lines << '' end lines << %(
) end end lines << '
' lines * LF end def convert_colist(node) lines = ['
    '] num = CALLOUT_START_NUM node.items.each_with_index do |item, i| lines << %(
  1. #{num} #{item.text}#{item.content if item.blocks?}
  2. ) num = num.next end lines << '
' end # TODO: add complex class if list has nested blocks def convert_dlist(node) lines = [] id_attribute = node.id ? %( id="#{node.id}") : '' classes = case node.style when 'horizontal' ['hdlist', node.role] when 'itemized', 'ordered' # QUESTION: should we just use itemized-list and ordered-list as the class here? or just list? ['dlist', %(#{node.style}-list), node.role] else ['description-list'] end.compact class_attribute = %( class="#{classes.join ' '}") lines << %() lines << %(
#{node.title}
) if node.title? case (style = node.style) when 'itemized', 'ordered' list_tag_name = style == 'itemized' ? 'ul' : 'ol' role = node.role subject_stop = node.attr 'subject-stop', (role && (node.has_role? 'stack') ? nil : ':') list_class_attr = node.option?('brief') ? ' class="brief"' : '' lines << %(<#{list_tag_name}#{list_class_attr}#{list_tag_name == 'ol' && (node.option? 'reversed') ? ' reversed="reversed"' : ''}>) node.items.each do |subjects, dd| # consists of one term (a subject) and supporting content subject = Array(subjects).first.text subject_plain = xml_sanitize subject, :plain subject_element = %(#{subject}#{subject_stop && subject_plain !~ TRAILING_PUNCT_RX ? subject_stop : ''}) lines << '
  • ' if dd # NOTE: must wrap remaining text in a span to help webkit justify the text properly lines << %(#{subject_element}#{dd.text? ? %( #{dd.text}) : ''}) lines << dd.content if dd.blocks? else lines << %(#{subject_element}) end lines << '
  • ' end lines << %() when 'horizontal' lines << '' if (node.attr? 'labelwidth') || (node.attr? 'itemwidth') lines << '' col_style_attribute = node.attr?('labelwidth') ? %( style="width: #{(node.attr 'labelwidth').chomp '%'}%;") : '' lines << %() col_style_attribute = node.attr?('itemwidth') ? %( style="width: #{(node.attr 'itemwidth').chomp '%'}%;") : '' lines << %() lines << '' end node.items.each do |terms, dd| lines << '' lines << %(' lines << '' lines << '' end lines << '
    ) first_term = true terms.each do |dt| lines << %(
    ) unless first_term lines << '

    ' lines << dt.text lines << '

    ' first_term = nil end lines << '
    ' if dd lines << %(

    #{dd.text}

    ) if dd.text? lines << dd.content if dd.blocks? end lines << '
    ' else lines << '
    ' node.items.each do |terms, dd| Array(terms).each do |dt| lines << %(
    #{dt.text}
    ) end next unless dd lines << '
    ' if dd.blocks? lines << %(#{dd.text}) if dd.text? lines << dd.content else lines << %(#{dd.text}) end lines << '
    ' end lines << '
    ' end lines << '' lines * LF end def convert_olist(node) complex = false div_classes = ['ordered-list', node.style, node.role].compact ol_classes = [node.style, (node.option?('brief') ? 'brief' : nil)].compact ol_class_attr = ol_classes.empty? ? '' : %( class="#{ol_classes * ' '}") ol_start_attr = node.attr?('start') ? %( start="#{node.attr 'start'}") : '' id_attribute = node.id ? %( id="#{node.id}") : '' lines = [%()] lines << %(

    #{node.title}

    ) if node.title? lines << %() node.items.each do |item| lines << %(
  • #{item.text}) if item.blocks? lines << item.content complex = true unless item.blocks.size == 1 && item.blocks[0].is_a?(::Asciidoctor::List) end lines << '
  • ' end if complex div_classes << 'complex' lines[0] = %(
    ) end lines << '
    ' lines * LF end def convert_ulist(node) complex = false div_classes = ['itemized-list', node.style, node.role].compact ul_classes = [node.style, (node.option?('brief') ? 'brief' : nil)].compact ul_class_attr = ul_classes.empty? ? '' : %( class="#{ul_classes * ' '}") id_attribute = node.id ? %( id="#{node.id}") : '' lines = [%()] lines << %(

    #{node.title}

    ) if node.title? lines << %() node.items.each do |item| lines << %(
  • #{item.text}) if item.blocks? lines << item.content complex = true unless item.blocks.size == 1 && item.blocks[0].is_a?(::Asciidoctor::List) end lines << '
  • ' end if complex div_classes << 'complex' lines[0] = %(
    ) end lines << '
    ' lines * LF end def doc_option(document, key) loop do value = document.options[key] return value unless value.nil? document = document.parent_document break if document.nil? end nil end def root_document(document) document = document.parent_document until document.parent_document.nil? document end def register_media_file(node, target, media_type) if target.end_with?('.svg') || target.start_with?('data:image/svg+xml') chapter = get_enclosing_chapter node chapter.set_attr 'epub-properties', [] unless chapter.attr? 'epub-properties' epub_properties = chapter.attr 'epub-properties' epub_properties << 'svg' unless epub_properties.include? 'svg' end return if target.start_with? 'data:' if Asciidoctor::Helpers.uriish? target # We need to add both local and remote media files to manifest fs_path = nil else out_dir = node.attr('outdir', nil, true) || doc_option(node.document, :to_dir) fs_path = (::File.join out_dir, target) unless ::File.exist? fs_path base_dir = root_document(node.document).base_dir fs_path = ::File.join base_dir, target end end # We need *both* virtual and physical image paths. Unfortunately, references[:images] only has one of them. @media_files[target] ||= { path: fs_path, media_type: media_type } end # @param node [Asciidoctor::Block] # @return [Array] def resolve_image_attrs(node) img_attrs = [] unless (alt = encode_attribute_value(node.alt)).empty? img_attrs << %(alt="#{alt}") end # Unlike browsers, Calibre/Kindle *do* scale image if only height is specified # So, in order to match browser behavior, we just always omit height if (scaledwidth = node.attr 'scaledwidth') img_attrs << %(style="width: #{scaledwidth}") elsif (width = node.attr 'width') # HTML5 spec (and EPUBCheck) only allows pixels in width, but browsers also accept percents # and there are multiple AsciiDoc files in the wild that have width=percents% # So, for compatibility reasons, output percentage width as a CSS style img_attrs << if width[/^\d+%$/] %(style="width: #{width}") else %(width="#{width}") end end img_attrs end def convert_audio(node) id_attr = node.id ? %( id="#{node.id}") : '' target = node.media_uri node.attr 'target' register_media_file node, target, 'audio' title_element = node.title? ? %(\n
    #{node.captioned_title}
    ) : '' autoplay_attr = node.option?('autoplay') ? ' autoplay="autoplay"' : '' controls_attr = node.option?('nocontrols') ? '' : ' controls="controls"' loop_attr = node.option?('loop') ? ' loop="loop"' : '' start_t = node.attr 'start' end_t = node.attr 'end' time_anchor = if start_t || end_t %(#t=#{start_t || ''}#{end_t ? ",#{end_t}" : ''}) else '' end %(#{title_element}
    ) end # TODO: Support multiple video files in different formats for a single video def convert_video(node) id_attr = node.id ? %( id="#{node.id}") : '' target = node.media_uri node.attr 'target' register_media_file node, target, 'video' title_element = node.title? ? %(\n
    #{node.captioned_title}
    ) : '' width_attr = node.attr?('width') ? %( width="#{node.attr 'width'}") : '' height_attr = node.attr?('height') ? %( height="#{node.attr 'height'}") : '' autoplay_attr = node.option?('autoplay') ? ' autoplay="autoplay"' : '' controls_attr = node.option?('nocontrols') ? '' : ' controls="controls"' loop_attr = node.option?('loop') ? ' loop="loop"' : '' start_t = node.attr 'start' end_t = node.attr 'end' time_anchor = if start_t || end_t %(#t=#{start_t || ''}#{end_t ? ",#{end_t}" : ''}) else '' end if (poster = node.attr 'poster').nil_or_empty? poster_attr = '' else poster = node.media_uri poster register_media_file node, poster, 'image' poster_attr = %( poster="#{poster}") end %(#{title_element}
    ) end # @param node [Asciidoctor::Block] # @return [String] def convert_image(node) target = node.image_uri node.attr 'target' register_media_file node, target, 'image' id_attr = node.id ? %( id="#{node.id}") : '' title_element = node.title? ? %(\n
    #{node.captioned_title}
    ) : '' img_attrs = resolve_image_attrs node %(
    #{title_element} ) end def get_enclosing_chapter(node) loop do return nil if node.nil? return node unless get_chapter_filename(node).nil? node = node.parent end end def convert_inline_anchor(node) case node.type when :xref doc = node.document refid = node.attr('refid') target = node.target text = node.text id_attr = '' if (path = node.attributes['path']) # NOTE: non-nil path indicates this is an inter-document xref that's not included in current document text = node.text || path elsif refid == '#' logger.warn %(#{::File.basename doc.attr('docfile')}: <> xref syntax isn't supported anymore. Use either <> or <>) elsif refid ref = doc.references[:refs][refid] our_chapter = get_enclosing_chapter node ref_chapter = get_enclosing_chapter ref if ref_chapter ref_docname = get_chapter_filename ref_chapter if ref_chapter == our_chapter # ref within same chapter file id_attr = %( id="xref-#{refid}") target = %(##{refid}) elsif refid == ref_docname # ref to top section of other chapter file id_attr = %( id="xref--#{refid}") target = %(#{refid}.xhtml) else # ref to section within other chapter file id_attr = %( id="xref--#{ref_docname}--#{refid}") target = %(#{ref_docname}.xhtml##{refid}) end id_attr = '' unless @xrefs_seen.add? refid text ||= (ref.xreftext node.attr('xrefstyle', nil, true)) else logger.warn %(#{::File.basename doc.attr('docfile')}: invalid reference to unknown anchor: #{refid}) end end %(#{text || "[#{refid}]"}) when :ref # NOTE: id is used instead of target starting in Asciidoctor 2.0.0 %() when :link %(#{node.text}) when :bibref # NOTE: reftext is no longer enclosed in [] starting in Asciidoctor 2.0.0 # NOTE id is used instead of target starting in Asciidoctor 2.0.0 if (reftext = node.reftext) reftext = %([#{reftext}]) unless reftext.start_with? '[' else reftext = %([#{node.target || node.id}]) end %(#{reftext}) else logger.warn %(unknown anchor type: #{node.type.inspect}) nil end end def convert_inline_break(node) %(#{node.text}
    ) end # @param node [Asciidoctor::Inline] # @return [String] def convert_inline_button(node) %(#{node.text}) end def convert_inline_callout(node) num = CALLOUT_START_NUM int_num = node.text.to_i (int_num - 1).times { num = num.next } %(#{num}) end # @param node [Asciidoctor::Inline] # @return [String] def convert_inline_footnote(node) if (index = node.attr 'index') attrs = [] attrs << %(id="#{node.id}") if node.id %([#{index}]) elsif node.type == :xref %(#{node.text}) end end def convert_inline_image(node) if node.type == 'icon' icon_names << (icon_name = node.target) i_classes = ['icon', %(i-#{icon_name})] i_classes << %(icon-#{node.attr 'size'}) if node.attr? 'size' i_classes << %(icon-flip-#{(node.attr 'flip')[0]}) if node.attr? 'flip' i_classes << %(icon-rotate-#{node.attr 'rotate'}) if node.attr? 'rotate' i_classes << node.role if node.role? i_classes << node.attr('float') if node.attr 'float' %() else target = node.image_uri node.target register_media_file node, target, 'image' img_attrs = resolve_image_attrs node img_attrs << %(class="inline#{prepend_space node.role}#{prepend_space node.attr('float')}") %() end end def convert_inline_indexterm(node) node.type == :visible ? node.text : '' end def convert_inline_kbd(node) if (keys = node.attr 'keys').size == 1 %(#{keys[0]}) else key_combo = keys.map { |key| %(#{key}) }.join '+' %(#{key_combo}) end end def convert_inline_menu(node) menu = node.attr 'menu' # NOTE: we swap right angle quote with chevron right from FontAwesome using CSS caret = %(#{NO_BREAK_SPACE}#{RIGHT_ANGLE_QUOTE} ) if !(submenus = node.attr 'submenus').empty? submenu_path = submenus.map { |submenu| %(#{submenu}#{caret}) }.join.chop %(#{menu}#{caret}#{submenu_path} #{node.attr 'menuitem'}) elsif (menuitem = node.attr 'menuitem') %(#{menu}#{caret}#{menuitem}) else %(#{menu}) end end def convert_inline_quoted(node) open, close, tag = QUOTE_TAGS[node.type] content = if node.type == :asciimath && asciimath_available? AsciiMath.parse(node.text).to_mathml 'mml:' else node.text end node.add_role 'literal' if %i[monospaced asciimath latexmath].include? node.type if node.id class_attr = class_string node if tag %(#{open.chop} id="#{node.id}"#{class_attr}>#{content}#{close}) else %(#{open}#{content}#{close}) end elsif role_valid_class? node.role class_attr = class_string node if tag %(#{open.chop}#{class_attr}>#{content}#{close}) else %(#{open}#{content}#{close}) end else %(#{open}#{content}#{close}) end end def output_content(node) node.content_model == :simple ? %(

    #{node.content}

    ) : node.content end def encode_attribute_value(val) val.gsub '"', '"' end # FIXME: merge into with xml_sanitize helper def xml_sanitize(value, target = :attribute) sanitized = value.include?('<') ? value.gsub(XML_ELEMENT_RX, '').strip.tr_s(' ', ' ') : value if target == :plain && (sanitized.include? ';') if sanitized.include? '&#' sanitized = sanitized.gsub(CHAR_ENTITY_RX) do [::Regexp.last_match(1).to_i].pack 'U*' end end sanitized = sanitized.gsub FROM_HTML_SPECIAL_CHARS_RX, FROM_HTML_SPECIAL_CHARS_MAP elsif target == :attribute sanitized = sanitized.gsub '"', '"' if sanitized.include? '"' end sanitized end # TODO: make check for last content paragraph a feature of Asciidoctor def mark_last_paragraph(root) return unless (last_block = root.blocks[-1]) last_block = last_block.blocks[-1] while last_block.context == :section && last_block.blocks? if last_block.context == :paragraph last_block.attributes['role'] = last_block.role? ? %(#{last_block.role} last) : 'last' end nil end # Prepend a space to the value if it's non-nil, otherwise return empty string. def prepend_space(value) value ? %( #{value}) : '' end def add_theme_assets(doc) workdir = if doc.attr? 'epub3-stylesdir' stylesdir = doc.attr 'epub3-stylesdir' # FIXME: make this work for Windows paths!! if stylesdir.start_with? '/' stylesdir else docdir = doc.attr 'docdir', '.' docdir = '.' if docdir.empty? ::File.join docdir, stylesdir end else ::File.join DATA_DIR, 'styles' end # TODO: improve design/UX of custom theme functionality, including custom fonts %w[epub3 epub3-css3-only].each do |f| css = load_css_file File.join(workdir, %(#{f}.scss)) @book.add_item %(styles/#{f}.css), content: css.to_ios end syntax_hl = doc.syntax_highlighter if syntax_hl&.write_stylesheet? doc Dir.mktmpdir do |dir| syntax_hl.write_stylesheet doc, dir Pathname.glob("#{dir}/**/*").map do |filename| # Workaround for https://github.com/skoji/gepub/pull/117 next unless filename.file? filename.open do |f| @book.add_item filename.basename.to_s, content: f end end end end font_files, font_css = select_fonts load_css_file(File.join(DATA_DIR, 'styles/epub3-fonts.scss')), (doc.attr 'scripts', 'latin') @book.add_item 'styles/epub3-fonts.css', content: font_css.to_ios unless font_files.empty? # NOTE: metadata property in oepbs package manifest doesn't work; must use proprietary iBooks file instead @book.add_optional_file 'META-INF/com.apple.ibooks.display-options.xml', ' '.to_ios font_files.each do |font_file| @book.add_item font_file, content: File.join(DATA_DIR, font_file) end end nil end # @param doc [Asciidoctor::Document] # @param name [String] # @return [GEPUB::Item, nil] def add_cover_page(doc, name) image_attr_name = %(#{name}-image) return nil if (image_path = doc.attr image_attr_name).nil? imagesdir = (doc.attr 'imagesdir', '.').chomp '/' imagesdir = (imagesdir == '.' ? '' : %(#{imagesdir}/)) image_attrs = {} if (image_path.include? ':') && image_path =~ IMAGE_MACRO_RX logger.warn %(deprecated block macro syntax detected in :#{image_attr_name}: attribute) if image_path.start_with? 'image::' image_path = %(#{imagesdir}#{::Regexp.last_match(1)}) unless ::Regexp.last_match(2).empty? (::Asciidoctor::AttributeList.new ::Regexp.last_match(2)).parse_into image_attrs, %w[alt width height] end end image_href = %(#{imagesdir}jacket/#{name}#{::File.extname image_path}) workdir = doc.attr 'docdir' workdir = '.' if workdir.nil_or_empty? image_path = File.join workdir, image_path unless File.absolute_path? image_path begin @book.add_item(image_href, content: image_path).cover_image rescue StandardError => e logger.error %(#{::File.basename doc.attr('docfile')}: error adding cover image. Make sure that :#{image_attr_name}: attribute points to a valid image file. #{e}) return nil end unless !image_attrs.empty? && (width = image_attrs['width']) && (height = image_attrs['height']) width = 1050 height = 1600 end # NOTE: SVG wrapper maintains aspect ratio and confines image to view box content = %( #{sanitize_doctitle_xml doc, :cdata} ) @book.add_ordered_item %(#{name}.xhtml), content: content.to_ios, id: name end def get_frontmatter_files(doc, workdir) if doc.attr? 'epub3-frontmatterdir' fmdir = doc.attr 'epub3-frontmatterdir' fmglob = 'front-matter.*\.html' fm_path = File.join workdir, fmdir unless Dir.exist? fm_path logger.warn %(#{File.basename doc.attr('docfile')}: directory specified by 'epub3-frontmattderdir' doesn't exist! Ignoring ...) return [] end fms = Dir.entries(fm_path).delete_if { |x| !x.match fmglob }.sort.map { |y| File.join fm_path, y } if fms && !fms.empty? fms else logger.warn %(#{File.basename doc.attr('docfile')}: directory specified by 'epub3-frontmattderdir' contains no suitable files! Ignoring ...) [] end elsif File.exist? File.join workdir, 'front-matter.html' [File.join(workdir, 'front-matter.html')] else [] end end def add_front_matter_page(doc) workdir = doc.attr 'docdir' workdir = '.' if workdir.nil_or_empty? result = nil get_frontmatter_files(doc, workdir).each do |front_matter| front_matter_content = ::File.read front_matter front_matter_file = File.basename front_matter, '.html' item = @book.add_ordered_item "#{front_matter_file}.xhtml", content: front_matter_content.to_ios item.add_property 'svg' if SVG_IMG_SNIFF_RX =~ front_matter_content # Store link to first frontmatter page result = item if result.nil? front_matter_content.scan IMAGE_SRC_SCAN_RX do @book.add_item ::Regexp.last_match(1), content: File.join(File.dirname(front_matter), ::Regexp.last_match(1)) end end result end def add_profile_images(doc, usernames) imagesdir = (doc.attr 'imagesdir', '.').chomp '/' imagesdir = (imagesdir == '.' ? nil : %(#{imagesdir}/)) @book.add_item %(#{imagesdir}avatars/default.jpg), content: ::File.join(DATA_DIR, 'images/default-avatar.jpg') @book.add_item %(#{imagesdir}headshots/default.jpg), content: ::File.join(DATA_DIR, 'images/default-headshot.jpg') workdir = '.' if (workdir = doc.attr 'docdir').nil_or_empty? usernames.each do |username| avatar = %(#{imagesdir}avatars/#{username}.jpg) if ::File.readable?(resolved_avatar = (::File.join workdir, avatar)) @book.add_item avatar, content: resolved_avatar else logger.error %(avatar for #{username} not found or readable: #{avatar}; falling back to default avatar) @book.add_item avatar, content: ::File.join(DATA_DIR, 'images/default-avatar.jpg') end headshot = %(#{imagesdir}headshots/#{username}.jpg) if ::File.readable?(resolved_headshot = (::File.join workdir, headshot)) @book.add_item headshot, content: resolved_headshot elsif doc.attr? 'builder', 'editions' logger.error %(headshot for #{username} not found or readable: #{headshot}; falling back to default headshot) @book.add_item headshot, content: ::File.join(DATA_DIR, 'images/default-headshot.jpg') end end nil end def nav_doc(doc, items, landmarks, depth) lines = [%( #{sanitize_doctitle_xml doc, :cdata}

    #{doc.attr 'toc-title'}

    ' unless landmarks.empty? lines << ' ' end lines << '
    ' lines * LF end def nav_level(items, depth, state = {}) lines = [] lines << '
      ' items.each do |item| # index = (state[:index] = (state.fetch :index, 0) + 1) if (chapter_filename = get_chapter_filename item).nil? item_label = sanitize_xml get_numbered_title(item), :pcdata item_href = %(#{state[:content_doc_href]}##{item.id}) else # NOTE: we sanitize the chapter titles because we use formatting to control layout item_label = if item.context == :document sanitize_doctitle_xml item, :cdata else sanitize_xml get_numbered_title(item), :cdata end item_href = (state[:content_doc_href] = %(#{chapter_filename}.xhtml)) end lines << %(
    1. #{item_label}) if depth.zero? || (child_sections = item.sections).empty? lines[-1] = %(#{lines[-1]}
    2. ) else lines << (nav_level child_sections, depth - 1, state) lines << '' end state.delete :content_doc_href unless chapter_filename.nil? end lines << '
    ' lines * LF end def ncx_doc(doc, items, depth) # TODO: populate docAuthor element based on unique authors in work lines = [%( %s #{sanitize_doctitle_xml doc, :cdata} )] lines << (ncx_level items, depth, state = {}) lines[0] = lines[0].sub '%s', %() lines << %( ) lines * LF end def ncx_level(items, depth, state = {}) lines = [] state[:max_depth] = (state.fetch :max_depth, 0) + 1 items.each do |item| index = (state[:index] = (state.fetch :index, 0) + 1) item_id = %(nav_#{index}) if (chapter_filename = get_chapter_filename item).nil? item_label = sanitize_xml get_numbered_title(item), :cdata item_href = %(#{state[:content_doc_href]}##{item.id}) else item_label = if item.context == :document sanitize_doctitle_xml item, :cdata else sanitize_xml get_numbered_title(item), :cdata end item_href = (state[:content_doc_href] = %(#{chapter_filename}.xhtml)) end lines << %() lines << %(#{item_label}) lines << %() unless depth.zero? || (child_sections = item.sections).empty? lines << (ncx_level child_sections, depth - 1, state) end lines << %() state.delete :content_doc_href unless chapter_filename.nil? end lines * LF end # Swap fonts in CSS based on the value of the document attribute 'scripts', # then return the list of fonts as well as the font CSS. def select_fonts(font_css, scripts = 'latin') font_css = font_css.gsub(/(?<=-)latin(?=\.ttf\))/, scripts) unless scripts == 'latin' # match CSS font urls in the forms of: # src: url(../fonts/notoserif-regular-latin.ttf); # src: url(../fonts/notoserif-regular-latin.ttf) format("truetype"); font_list = font_css.scan(%r{url\(\.\./([^)]+?\.ttf)\)}).flatten [font_list, font_css] end def load_css_file(filename) template = File.read filename load_paths = [File.dirname(filename)] sass_engine = Sass::Engine.new template, syntax: :scss, cache: false, load_paths: load_paths, style: :compressed sass_engine.render end def build_epubcheck_command unless @epubcheck_path.nil? logger.debug %(Using ebook-epubcheck-path attribute: #{@epubcheck_path}) return [@epubcheck_path] end unless (result = ENV.fetch('EPUBCHECK', nil)).nil? logger.debug %(Using EPUBCHECK env variable: #{result}) return [result] end begin result = ::Gem.bin_path 'epubcheck-ruby', 'epubcheck' logger.debug %(Using EPUBCheck from gem: #{result}) [::Gem.ruby, result] rescue ::Gem::Exception => e logger.debug %(#{e}; Using EPUBCheck from PATH) ['epubcheck'] end end def validate_epub(epub_file) argv = build_epubcheck_command + ['-w', epub_file] begin out, err, res = Open3.capture3(*argv) rescue Errno::ENOENT => e raise 'Unable to run EPUBCheck. Either install epubcheck-ruby gem or place `epubcheck` executable on PATH or set EPUBCHECK environment variable with path to it', cause: e end out.each_line do |line| logger.info line end err.each_line do |line| log_line line end logger.error %(EPUB validation failed: #{epub_file}) unless res.success? end def log_line(line) line = line.strip case line when /^fatal/i logger.fatal line when /^error/i logger.error line when /^warning/i logger.warn line else logger.info line end end private def class_string(node) role = node.role return '' unless role_valid_class? role %( class="#{role}") end # Handles asciidoctor 1.5.6 quirk when role can be parent def role_valid_class?(role) role.is_a? String end end Extensions.register do if (document = @document).backend == 'epub3' document.set_attribute 'listing-caption', 'Listing' # TODO: bw theme for CodeRay document.set_attribute 'pygments-style', 'bw' unless document.attr? 'pygments-style' document.set_attribute 'rouge-style', 'bw' unless document.attr? 'rouge-style' # Backward compatibility for documents that were created before we dropped MOBI support document.set_attribute 'ebook-format', 'epub3' document.set_attribute 'ebook-format-epub3', '' # Enable generation of section ids because we use them for chapter filenames document.set_attribute 'sectids' treeprocessor do process do |doc| # :sectids: doesn't generate id for top-level section (why?), do it manually doc.id = Section.generate_id(doc.first_section&.title || doc.attr('docname') || 'document', doc) if doc.id.nil_or_empty? if (preamble = doc.blocks[0]) && preamble.context == :preamble && preamble.id.nil_or_empty? # :sectids: doesn't generate id for preamble (because it is not a section), do it manually preamble.id = Section.generate_id(preamble.title || 'preamble', doc) end nil end end end end end end