Pandocで生成したepubファイルの目次情報と章番号を修正するRubyスクリプト

modify_epub_toc_and_headings.rb

#! /usr/bin/env ruby

require 'nokogiri'
require 'pathname'

# na.xhtml

# nav.xhtml専用
def modify_appendix_header(node, ns, prefix_letter='A') 
    # 付録を特別扱いすためのメソッド
    # puts node.name
    chapter_node = node.xpath('./xmlns:a')

    prev_section_number = chapter_node.children[0].text  # span tag

    chapter_node.children[0].content = "付録" + prefix_letter

    title = chapter_node.children[1].text.gsub(/付録.?\s+(.+)/, '\1')
    chapter_node.children[0]['class'] = 'appendix-header-number'
    chapter_node.children[1].content = title

    for n in node.xpath(".//xmlns:span[@class='section-header-number']") do
        new_section_header = n.text.gsub(/#{prev_section_number}\.(\d+?)/, "#{prefix_letter}." + '\1' )
        n.content = new_section_header
    end
end

def modify_nav_chapter_header(node)
    chapter_node = node.xpath('./xmlns:a')
    if chapter_node.children[0].name != 'span' then
        return
    else
        chapter_number = chapter_node.children[0].text
        chapter_node.children[0].content = "第#{chapter_number}章"
    end
    # mn = node.text.strip.match(/^(\d)\s+?(.+)/)

end

def modify_nav_xhtml()

    nav_xml = Nokogiri::XML(open('_temp/EPUB/nav.xhtml').read)

    ns = nav_xml.namespaces

    letter = 'A'
    for node in nav_xml.xpath('//xmlns:nav/xmlns:ol/xmlns:li') do
        if node.text =~ /付録/ then
            modify_appendix_header(node, ns, prefix_letter=letter)
            letter = letter.succ
        else
            modify_nav_chapter_header(node)
        end
    end

    File.write('_temp/EPUB/nav.xhtml', nav_xml)
end

################################################################
# toc.ncx
def modify_toc_ncx()
        
    ncx_xml = Nokogiri::XML(open('_temp/EPUB/toc.ncx').read)

    replace_section_numbers = []
    prefix_letters = ['A', 'B', 'C', 'D', 'F', 'G' ]

    for text_node in ncx_xml.xpath('//xmlns:navLabel/xmlns:text') do
        if not text_node.text =~ /付録/
            mc = text_node.text.strip.match(/^(\d)\s+?(.+)/)
            if mc then
                text_node.content = "第#{mc[1]}章 #{mc[2]}"
            end
        else
            # 付録の章見出しには番号は不要
            mo = text_node.text.strip.match(/(\d)\s+?(.+)/)
            if mo then
                replace_section_numbers.push(mo[1])
                text_node.content = mo[2].to_s
            end
        end
    end

    for sub_text_node in ncx_xml.xpath('//xmlns:navLabel/xmlns:text') do
        for number, letter in replace_section_numbers.zip(prefix_letters) do

            if sub_text_node .text.strip =~ /^#{number}\.(\d+?)/ then
                new_label_text = sub_text_node.text.strip.gsub(/^#{number}\.(\d+?)/, "#{letter}." + '\1' )
                sub_text_node.content = new_label_text
            end
        end
    end

    File.write('_temp/EPUB/toc.ncx', ncx_xml.to_xml)
end

################################################################
# for each ch_xxx.xhtml
def modify_chapter(file, prefix_letter='A')
    xhtml = Nokogiri::XML(open(file))

    md = xhtml.xpath('//xmlns:h1').children[1].text.strip.match(/(付録.*?)\s+?(.+)/)
    prev_chapter_number = xhtml.xpath('//xmlns:h1//xmlns:span[@class="header-section-number"]').first.text

    if md then
        #xhtml.xpath('//xmlns:h1//xmlns:span[@class="header-section-number"]').children[0].content = md[1]
        xhtml.xpath('//xmlns:h1').children[0].content = md[1]
        xhtml.xpath('//xmlns:h1').children[1].content = " #{md[2]}"

        xhtml.xpath('//xmlns:h1').children[0]['class'] = 'appendix-header-number'

        for section_node in xhtml.xpath('//xmlns:span[@class="header-section-number"]') do
            section_node.content = section_node.text.strip.gsub( /^#{prev_chapter_number}(.+)/ , prefix_letter + '\1')
        end

        File.write(file, xhtml)
    end
end

def modify_appendix_chapter_files()

    appendix_files = []

    chapter_files = Pathname.glob('_temp/EPUB/text/ch???.xhtml').sort()

    for file in chapter_files do
        xhtml = Nokogiri::XML(file.open())

        if xhtml.xpath('//xmlns:title').text =~ /付録/ then
            appendix_files.push(file)
        end
    end

    prefix_letters = ['A', 'B', 'C', 'D', 'F', 'G' ]
    for apx, letter in appendix_files.zip(prefix_letters) do
        modify_chapter(apx.to_s,prefix_letter=letter)
    end

end



# main
modify_nav_xhtml()
modify_toc_ncx()
modify_appendix_chapter_files()