cbeta 0.4.3 → 0.5.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 1c29d0e70255be20d828cecb8f3feff8f0d4b448
-  data.tar.gz: f90911d3955acb1f491e1e7c61fd1797617aa8f2
+  metadata.gz: 416026a7921c0c2bb6c602be134095f4079951fe
+  data.tar.gz: d72055681fb1c8ff22a4317ed0310fc368381e1a
 SHA512:
-  metadata.gz: 2a545c0cb3fdc26d8b79a79172112b967f225aadaf03afe41b4117acfc44a67f7626cb1f07cb3262d0b6b4041978af9f87f15f9b6d4b574472f5da6c81153d21
-  data.tar.gz: fcad67ccefa4b4250a6d236671373cc4586f48d4da49b60bfddb659ae1fc0246e613f31e580ad6958b0cd0a1a358b68f9b37f34764a95efbe43f94e8711e8b7e
+  metadata.gz: 0d71cdacc97c1d6111cc2dc1dc9b5190d2398f05b9a45280744b7475579dd8f612efa382c0cfd16f07429391f99e330a8fe1c3db1c48d770076f90abe20d15c4
+  data.tar.gz: bd62c4cdfe0def5b62d7440f0ec029c2535e1384d0ac9865e2ba91a643db46b6a1fed02ed8a4c4435d9e26b878d8acfc7e3c91994bac50899b7e8ec4413b8717

data/lib/cbeta.rb

@@ -35,7 +35,7 @@ class CBETA
 
 	# 載入藏經資料
   def initialize()
-    fn = File.join(File.dirname(__FILE__), 'canons.csv')
+    fn = File.join(File.dirname(__FILE__), 'data/canons.csv')
     text = File.read(fn)
     @canon_abbr = {}
     CSV.parse(text, :headers => true) do |row|
@@ -61,6 +61,7 @@ end
 
 require 'cbeta/gaiji'
 require 'cbeta/bm_to_text'
+require 'cbeta/p5a_to_epub'
 require 'cbeta/p5a_to_html'
 require 'cbeta/p5a_to_simple_html'
 require 'cbeta/p5a_to_text'

data/lib/cbeta/gaiji.rb

@@ -4,7 +4,7 @@ require 'json'
 class CBETA::Gaiji
 	# 載入 CBETA 缺字資料庫
   def initialize()
-    fn = File.join(File.dirname(__FILE__), 'gaiji.json')
+    fn = File.join(File.dirname(__FILE__), '../data/gaiji.json')
     @gaijis = JSON.parse(File.read(fn))
   end
 

data/lib/cbeta/p5a_to_epub.rb

@@ -0,0 +1,780 @@
+require 'cgi'
+require 'date'
+require 'fileutils'
+require 'json'
+require 'nokogiri'
+require 'set'
+require 'gepub'
+require 'pp'
+
+# Convert CBETA XML P5a to EPUB
+#
+# CBETA XML P5a 可由此取得: https://github.com/cbeta-git/xml-p5a
+class CBETA::P5aToEPUB
+  # 內容不輸出的元素
+  PASS=['back', 'teiHeader']
+
+  # 某版用字缺的符號
+  MISSING = '－'
+  
+  NAV_TEMPLATE = File.read(File.join(File.dirname(__FILE__), '../data/epub-nav.xhtml'))
+  MAIN = 'main.xhtml'  
+
+  # @param temp_folder [String] 供 EPUB 暫存工作檔案的路徑
+  # @param graphic_base [String] 存放圖片的路徑
+  def initialize(temp_folder, graphic_base)
+    @temp_folder = temp_folder
+    @graphic_base = graphic_base
+    @cbeta = CBETA.new
+    @gaijis = CBETA::Gaiji.new
+  end
+
+  # 將某個 xml 轉為一個 EPUB
+  def convert_file(input_path, output_path)
+    return false unless input_path.end_with? '.xml'
+    
+    FileUtils.remove_dir(@temp_folder, force=true)
+    FileUtils::mkdir_p @temp_folder
+    
+    @book_id = File.basename(input_path, ".xml")
+    
+    sutra_init
+    
+    handle_file(input_path)
+    create_epub(output_path)
+  end
+
+  # 將某個資料夾下的每個 xml 檔都轉為一個對應的 EPUB。
+  # 資料夾可以是巢狀，全部都會遞迴處理。
+  #
+  # @example
+  #   require 'cbeta'
+  #   
+  #   TEMP = '/temp/epub-work'
+  #   IMG = '/Users/ray/Documents/Projects/D道安/figures'
+  #   
+  #   c = CBETA::P5aToEPUB.new(TEMP, IMG)
+  #   c.convert_folder('/Users/ray/Documents/Projects/D道安/xml-p5a/DA', '/temp/cbeta-epub/DA')
+  def convert_folder(input_folder, output_folder)
+    FileUtils.remove_dir(output_folder, force=true)
+    FileUtils::mkdir_p output_folder
+    Dir.foreach(input_folder) do |f|
+      next if f.start_with? '.'
+      p1 = File.join(input_folder, f)
+      if File.file?(p1)
+        f.sub!(/.xml$/, '.epub')
+        p2 = File.join(output_folder, f)
+        convert_file(p1, p2)
+      else
+        p2 = File.join(output_folder, f)
+        convert_folder(p1, p2)
+      end
+    end
+  end
+  
+  # 將多個 xml 檔案合成一個 EPUB
+  #
+  # @example 大般若經 跨三冊 合成一個 EPUB
+  #   require 'cbeta'
+  #   
+  #   TEMP = '/temp/epub-work'
+  #   
+  #   xml_files = [
+  #     '/Users/ray/git-repos/cbeta-xml-p5a/T/05/T05n0220a.xml',
+  #     '/Users/ray/git-repos/cbeta-xml-p5a/T/06/T06n0220b.xml',
+  #     '/Users/ray/git-repos/cbeta-xml-p5a/T/07/T07n0220c.xml',
+  #     '/Users/ray/git-repos/cbeta-xml-p5a/T/07/T07n0220d.xml',
+  #     '/Users/ray/git-repos/cbeta-xml-p5a/T/07/T07n0220e.xml',
+  #     '/Users/ray/git-repos/cbeta-xml-p5a/T/07/T07n0220f.xml',
+  #     '/Users/ray/git-repos/cbeta-xml-p5a/T/07/T07n0220g.xml',
+  #     '/Users/ray/git-repos/cbeta-xml-p5a/T/07/T07n0220h.xml',
+  #     '/Users/ray/git-repos/cbeta-xml-p5a/T/07/T07n0220i.xml',
+  #     '/Users/ray/git-repos/cbeta-xml-p5a/T/07/T07n0220j.xml',
+  #     '/Users/ray/git-repos/cbeta-xml-p5a/T/07/T07n0220k.xml',
+  #     '/Users/ray/git-repos/cbeta-xml-p5a/T/07/T07n0220l.xml',
+  #     '/Users/ray/git-repos/cbeta-xml-p5a/T/07/T07n0220m.xml',
+  #     '/Users/ray/git-repos/cbeta-xml-p5a/T/07/T07n0220n.xml',
+  #     '/Users/ray/git-repos/cbeta-xml-p5a/T/07/T07n0220o.xml',
+  #   ]
+  #   
+  #   c = CBETA::P5aToEPUB.new(TEMP)
+  #   c.convert_sutra('T0220', '大般若經', xml_files, '/temp/cbeta-epub/T0220.epub')
+  def convert_sutra(book_id, title, xml_files, out)
+    @book_id = book_id
+    sutra_init
+    xml_files.each { |f| handle_file(f) }
+    
+    @title = title
+    create_epub(out)
+  end
+
+  private
+  
+  def copy_static_files(src, dest)
+    p1 = File.join(File.dirname(__FILE__), '../data', src)
+    p2 = File.join(@temp_folder, dest)
+    FileUtils.copy(p1, p2)
+  end
+  
+  def create_epub(output_path)
+    copy_static_files('epub-readme.xhtml', 'readme.xhtml')
+    copy_static_files('epub-donate.xhtml', 'donate.xhtml')
+    create_main_html
+    create_nav_html
+    
+    title = @title
+    book_id = @book_id
+    builder = GEPUB::Builder.new {
+      language 'zh-TW'
+      unique_identifier "http://www.cbeta.org/#{book_id}", 'BookID', 'URL'
+      title title
+
+      creator 'CBETA'
+
+      contributors 'DILA'
+
+      date Date.today.to_s
+    }
+
+    # in resources block, you can define resources by its relative path and datasource.
+    # item creator methods are: files, file.
+    builder.resources(:workdir => @temp_folder) {
+      glob 'img/*'
+      
+      # this is navigation document.
+      nav 'nav.xhtml'
+      
+      # ordered item. will be added to spine.
+      ordered {
+        file 'readme.xhtml'
+        file 'main.xhtml'
+        file 'donate.xhtml'
+      }
+    }
+    builder.generate_epub(output_path)
+    puts "output: #{output_path}"
+  end
+
+  def create_main_html
+    fn = File.join(@temp_folder, MAIN)
+    s = <<eos
+<html xmlns="http://www.w3.org/1999/xhtml">
+<head>
+  <meta charset="utf-8" />
+  <title>#{@title}</title>
+</head>
+<body>
+<div id='body'>
+eos
+    s += @main_text + "\n</div><!-- end of div[@id='body'] -->\n"
+    s += "<div id='back'>\n" + @back + "</div></body></html>\n"
+    File.write(fn, s)
+  end
+  
+  def create_nav_html
+    @nav_root_ol.add_child("<li><a href='donate.xhtml'>贊助資訊</a></li>")
+    
+    fn = File.join(@temp_folder, 'nav.xhtml')
+    s = NAV_TEMPLATE % to_html(@nav_root_ol)
+    File.write(fn, s)
+  end
+  
+  def handle_anchor(e)
+    id = e['id']
+    if e.has_attribute?('id')
+      if id.start_with?('nkr_note_orig')
+        note = @notes[id]
+        note_text = traverse(note)
+        n = id[/^nkr_note_orig_(.*)$/, 1]
+        @back += "<span class='footnote' id='n#{n}'>#{note_text}</span>\n"
+        return "<a class='noteAnchor' href='#n#{n}'></a>"
+      elsif id.start_with? 'fx'
+        return "<span class='star'>[＊]</span>"
+      end
+    end
+
+    if e.has_attribute?('type')
+      if e['type'] == 'circle'
+        return '◎'
+      end
+    end
+
+    ''
+  end
+
+  def handle_app(e)
+    r = ''
+    if e['type'] == 'star'
+      c = e['corresp'][1..-1]
+      r = "<a class='noteAnchor star' href='#n#{c}'></a>"
+    end
+    r + traverse(e)
+  end
+
+  def handle_byline(e)
+    r = '<p class="byline">'
+    r += "<span class='lineInfo'>#{@lb}</span>"
+    r += traverse(e)
+    r + '</p>'
+  end
+
+  def handle_cell(e)
+    doc = Nokogiri::XML::Document.new
+    cell = doc.create_element('td')
+    cell['rowspan'] = e['rows'] if e.key? 'rows'
+    cell['colspan'] = e['cols'] if e.key? 'cols'
+    cell.inner_html = traverse(e)
+    to_html(cell) + "\n"
+  end
+
+  def handle_corr(e)
+    r = ''
+    if e.parent.name == 'choice'
+      sic = e.parent.at_xpath('sic')
+      unless sic.nil?
+        @dila_note += 1
+        r = "<a class='noteAnchor dila' href='#dila_note#{@dila_note}'></a>"
+
+        note = @orig
+        sic_text = traverse(sic, 'back')
+        if sic_text.empty?
+          note += MISSING
+        else
+          note += sic_text
+        end
+        @back += "<span class='footnote_dila' id='dila_note#{@dila_note}'>#{note}</span>\n"
+      end
+    end
+    r + "<span class='cbeta'>%s</span>" % traverse(e)
+  end
+
+  def handle_div(e)
+    if e.has_attribute? 'type'
+      @open_divs << e
+      r = traverse(e)
+      @open_divs.pop
+      return "<div class='div-#{e['type']}'>#{r}</div>"
+    else
+      return traverse(e)
+    end
+  end
+
+  def handle_figure(e)
+    "<div class='figure'>%s</div>" % traverse(e)
+  end
+
+  def handle_g(e, mode)
+    # if 有 <mapping type="unicode">
+    #   if 不在 Unicode Extension C, D, E 範圍裡
+    #     直接採用
+    #   else
+    #     預設呈現 unicode, 但仍包缺字資訊，供點選開 popup
+    # else if 有 <mapping type="normal_unicode">
+    #   預設呈現 normal_unicode, 但仍包缺字資訊，供點選開 popup
+    # else if 有 normalized form
+    #   預設呈現 normalized form, 但仍包缺字資訊，供點選開 popup
+    # else
+    #   預設呈現組字式, 但仍包缺字資訊，供點選開 popup
+    gid = e['ref'][1..-1]
+    g = @gaijis[gid]
+    abort "Line:#{__LINE__} 無缺字資料:#{gid}" if g.nil?
+    zzs = g['zzs']
+    
+    if mode == 'txt'
+      return g['roman'] if gid.start_with?('SD')
+      if zzs.nil?
+        abort "缺組字式：#{g}"
+      else
+        return zzs
+      end
+    end
+
+    if gid.start_with?('SD')
+      case gid
+      when 'SD-E35A'
+        return '（'
+      when 'SD-E35B'
+        return '）'
+      else
+        return "<span class='siddam' roman='#{g['roman']}' code='#{gid}' char='#{g['sd-char']}'/>"
+      end
+    end
+    
+    if gid.start_with?('RJ')
+      return "<span class='ranja' roman='#{g['roman']}' code='#{gid}' char='#{g['rj-char']}'/>"
+    end
+   
+    default = ''
+    if g.has_key?('unicode')
+      #if @unicode1.include?(g['unicode'])
+      # 如果在 unicode ext-C, ext-D, ext-E 範圍內
+      if (0x2A700..0x2CEAF).include? g['unicode'].hex
+        default = g['unicode-char']
+      else
+        return g['unicode-char'] # 直接採用 unicode
+      end
+    end
+
+    nor = ''
+    if g.has_key?('normal_unicode')
+      nor = g['normal_unicode']
+      default = nor if default.empty?
+    end
+
+    if g.has_key?('normal')
+      nor += ', ' unless nor==''
+      nor += g['normal']
+      default = g['normal'] if default.empty?
+    end
+
+    default = zzs if default.empty?
+    
+    href = 'http://dict.cbeta.org/dict_word/gaiji-cb/%s/%s.gif' % [gid[2, 2], gid]
+    unless @back.include?(href)
+      @back += "<span id='#{gid}' class='gaijiInfo' figure_url='#{href}' zzs='#{zzs}' nor='#{nor}'>#{default}</span>\n"
+    end
+    "<a class='gaijiAnchor' href='##{gid}'>#{default}</a>"
+  end
+
+  def handle_graphic(e)
+    url = e['url']
+    url.sub!(/^.*figures\/(.*)$/, '\1')
+    
+    src = File.join(@graphic_base, url)
+    basename = File.basename(src)
+    dest = File.join(@temp_folder, 'img', basename)
+    FileUtils.copy(src, dest)
+    
+    "<img src='img/#{basename}' />"
+  end
+
+  def handle_head(e)
+    r = ''
+    unless e['type'] == 'added'
+      i = @open_divs.size
+      r = "<p class='head' data-head-level='#{i}'>%s</p>" % traverse(e)
+    end
+    r
+  end
+
+  def handle_item(e)
+    "<li>%s</li>\n" % traverse(e)
+  end
+
+  def handle_juan(e)
+    "<p class='juan'>%s</p>" % traverse(e)
+  end
+
+  def handle_l(e)
+    if @lg_type == 'abnormal'
+      return traverse(e)
+    end
+
+    @in_l = true
+
+    doc = Nokogiri::XML::Document.new
+    cell = doc.create_element('div')
+    cell['class'] = 'lg-cell'
+    cell.inner_html = traverse(e)
+    
+    if @first_l
+      parent = e.parent()
+      if parent.has_attribute?('rend')
+        indent = parent['rend'].scan(/text-indent:[^:]*/)
+        unless indent.empty?
+          cell['style'] = indent[0]
+        end
+      end
+      @first_l = false
+    end
+    r = to_html(cell)
+    
+    unless @lg_row_open
+      r = "\n<div class='lg-row'>" + r
+      @lg_row_open = true
+    end
+    @in_l = false
+    r
+  end
+
+  def handle_lb(e)
+    # 卍續藏有 X 跟 R 兩種 lb, 只處理 X
+    return '' if e['ed'] != @series
+
+    @lb = e['n']
+    r = ''
+    #if e.parent.name == 'lg' and $lg_row_open
+    if @lg_row_open && !@in_l
+      # 每行偈頌放在一個 lg-row 裡面
+      # T46n1937, p. 914a01, l 包雙行夾註跨行
+      # T20n1092, 337c16, lb 在 l 中間，不結束 lg-row
+      r += "</div><!-- end of lg-row -->"
+      @lg_row_open = false
+    end
+    unless @next_line_buf.empty?
+      r += @next_line_buf
+      @next_line_buf = ''
+    end
+    r
+  end
+
+  def handle_lem(e)
+    r = ''
+    w = e['wit']
+    if w.include? 'CBETA' and not w.include? @orig
+      @dila_note += 1
+      r = "<a class='noteAnchor dila' href='#dila_note#{@dila_note}'></a>"
+      r += "<span class='cbeta'>%s</span>" % traverse(e)
+
+      note = lem_note_cf(e)
+      note += lem_note_rdg(e)
+      @back += "<span class='footnote_dila' id='dila_note#{@dila_note}'>#{note}</span>\n"
+    else
+      r = traverse(e)
+    end
+    r
+  end
+
+  def handle_lg(e)
+    r = ''
+    @lg_type = e['type']
+    if @lg_type == 'abnormal'
+      r = "<p class='lg-abnormal'>" + traverse(e) + "</p>"
+    else
+      @first_l = true
+      doc = Nokogiri::XML::Document.new
+      node = doc.create_element('div')
+      node['class'] = 'lg'
+      if e.has_attribute?('rend')
+        rend = e['rend'].gsub(/text-indent:[^:]*/, '')
+        node['style'] = rend
+      end
+      @lg_row_open = false
+      node.inner_html = traverse(e)
+      if @lg_row_open
+        node.inner_html += '</div><!-- end of lg -->'
+        @lg_row_open = false
+      end
+      r = "\n" + to_html(node)
+    end
+    r
+  end
+
+  def handle_list(e)
+    "<ul>%s</ul>" % traverse(e)
+  end
+
+  def handle_milestone(e)
+    ''
+  end
+
+  def handle_mulu(e)
+    return '' if e['type'] == '卷'
+    
+    level = e['level'].to_i
+    while @current_nav.size > level
+      @current_nav.pop
+    end
+    
+    label = traverse(e, 'txt')
+    @mulu_count += 1
+    li = @current_nav.last.add_child("<li><a href='#{@main_html}#mulu#{@mulu_count}'>#{label}</a></li>").first
+    ol = li.add_child('<ol></ol>').first
+    @current_nav << ol
+    "<a id='mulu#{@mulu_count}' />"
+  end
+
+  def handle_node(e, mode)
+    return '' if e.comment?
+    return handle_text(e, mode) if e.text?
+    return '' if PASS.include?(e.name)
+    r = case e.name
+    when 'anchor'    then handle_anchor(e)
+    when 'app'       then handle_app(e)
+    when 'byline'    then handle_byline(e)
+    when 'cell'      then handle_cell(e)
+    when 'corr'      then handle_corr(e)
+    when 'div'       then handle_div(e)
+    when 'figure'    then handle_figure(e)
+    when 'foreign'   then ''
+    when 'g'         then handle_g(e, mode)
+    when 'graphic'   then handle_graphic(e)
+    when 'head'      then handle_head(e)
+    when 'item'      then handle_item(e)
+    when 'juan'      then handle_juan(e)
+    when 'l'         then handle_l(e)
+    when 'lb'        then handle_lb(e)
+    when 'lem'       then handle_lem(e)
+    when 'lg'        then handle_lg(e)
+    when 'list'      then handle_list(e)
+    when 'mulu'      then handle_mulu(e)
+    when 'note'      then handle_note(e)
+    when 'milestone' then handle_milestone(e)
+    when 'p'         then handle_p(e)
+    when 'rdg'       then ''
+    when 'reg'       then ''
+    when 'row'       then handle_row(e)
+    when 'sic'       then ''
+    when 'sg'        then handle_sg(e)
+    when 't'         then handle_t(e)
+    when 'tt'        then handle_tt(e)
+    when 'table'     then handle_table(e)
+    else traverse(e)
+    end
+    r
+  end
+
+  def handle_note(e)
+    n = e['n']
+    if e.has_attribute?('type')
+      t = e['type']
+      case t
+      when 'equivalent'
+        return ''
+      when 'orig'
+        return handle_note_orig(e)
+      when 'orig_biao'
+        return handle_note_orig(e, 'biao')
+      when 'orig_ke'
+        return handle_note_orig(e, 'ke')
+      when 'mod'
+        @pass << false
+        s = traverse(e)
+        @pass.pop
+        @back += "<span class='footnote_cb' id='n#{n}'>#{s}</span>\n"
+        return "<a class='noteAnchor' href='#n#{n}'></a>"
+      when 'rest'
+        return ''
+      else
+        return '' if t.start_with?('cf')
+      end
+    end
+
+    if e.has_attribute?('resp')
+      return '' if e['resp'].start_with? 'CBETA'
+    end
+
+    if e.has_attribute?('place') && e['place']=='inline'
+      r = traverse(e)
+      return "<span class='doube-line-note'>#{r}</span>"
+    else
+      return traverse(e)
+    end
+  end
+
+  def handle_note_orig(e, anchor_type=nil)
+    n = e['n']
+    @pass << false
+    s = traverse(e)
+    @pass.pop
+    @back += "<span class='footnote_orig' id='n#{n}'>#{s}</span>\n"
+
+    if @mod_notes.include? n
+      return ''
+    else
+      label = case anchor_type
+      when 'biao' then " data-label='標#{n[-2..-1]}'"
+      when 'ke'   then " data-label='科#{n[-2..-1]}'"
+      else ''
+      end
+      return "<a class='noteAnchor' href='#n#{n}'#{label}></a>"
+    end
+  end
+
+  def handle_p(e)
+    r = "<div class='p'>\n"
+    r += traverse(e)
+    r + "</div>\n"
+  end
+
+  def handle_row(e)
+    "<tr>" + traverse(e) + "</tr>\n"
+  end
+
+  def handle_sg(e)
+    '(' + traverse(e) + ')'
+  end
+
+  def handle_file(xml_fn)
+    puts "read #{xml_fn}"
+    @in_l = false
+    @lg_row_open = false
+    @mod_notes = Set.new
+    @next_line_buf = ''
+    @open_divs = []
+    
+    if @book_id.start_with? 'DA'
+      @orig = nil?
+    else
+      @orig = @cbeta.get_canon_abbr(@book_id[0])
+      abort "未處理底本: #{@book_id[0]}" if @orig.nil?
+    end
+
+    text = parse_xml(xml_fn)
+
+    # 註標移到 lg-cell 裡面，不然以 table 呈現 lg 會有問題
+    text.gsub!(/(<a class='noteAnchor'[^>]*><\/a>)(<div class="lg-cell"[^>]*>)/, '\2\1')
+    
+    @main_text += text    
+  end
+
+  def handle_t(e)
+    if e.has_attribute? 'place'
+      return '' if e['place'].include? 'foot'
+    end
+    r = traverse(e)
+
+    # <tt type="app"> 不是 悉漢雙行對照
+    return r if @tt_type == 'app'
+
+    # 處理雙行對照
+    i = e.xpath('../t').index(e)
+    case i
+    when 0
+      return r + '　'
+    when 1
+      @next_line_buf += r + '　'
+      return ''
+    else
+      return r
+    end
+  end
+
+  def handle_tt(e)
+    @tt_type = e['type']
+    traverse(e)
+  end
+
+  def handle_table(e)
+    "<table>" + traverse(e) + "</table>"
+  end
+
+  def handle_text(e, mode)
+    s = e.content().chomp
+    return '' if s.empty?
+    return '' if e.parent.name == 'app'
+
+    # cbeta xml 文字之間會有多餘的換行
+    r = s.gsub(/[\n\r]/, '')
+
+    # 把 & 轉為 &amp;
+    CGI.escapeHTML(r)
+  end
+
+  def lem_note_cf(e)
+    # ex: T32n1670A.xml, p. 703a16
+    # <note type="cf1">K30n1002_p0257a01-a23</note>
+    refs = []
+    e.xpath('./note').each { |n|
+      if n.key?('type') and n['type'].start_with? 'cf'
+        refs << n.content
+      end
+    }
+    if refs.empty?
+      ''
+    else
+      '修訂依據：' + refs.join('；') + '。'
+    end
+  end
+
+  def lem_note_rdg(lem)
+    r = ''
+    app = lem.parent
+    @pass << false
+    app.xpath('rdg').each { |rdg|
+      if rdg['wit'].include? @orig
+        s = traverse(rdg, 'back')
+        s = MISSING if s.empty?
+        r += @orig + s
+      end
+    }
+    @pass.pop
+    r += '。' unless r.empty?
+    r
+  end
+  
+  def sutra_init
+    s = NAV_TEMPLATE % '<ol></ol>'
+    @nav_doc = Nokogiri::XML(s)
+    
+    @nav_doc.remove_namespaces!()
+    @nav_root_ol = @nav_doc.at_xpath('//ol')
+    @current_nav = [@nav_root_ol]
+    
+    @nav_root_ol.add_child("<li><a href='readme.xhtml'>編輯說明</a></li>")
+    
+    @mulu_count = 0
+    @main_text = ''
+    @back = ''
+    @dila_note = 0
+    
+    FileUtils::mkdir_p File.join(@temp_folder, 'img')
+  end
+
+  def open_xml(fn)
+    s = File.read(fn)
+
+    if fn.include? 'T16n0657'
+      # 這個地方 雙行夾註 跨兩行偈頌
+      # 把 lb 移到 note 結束之前
+      # 讓 lg-row 先結束，再結束雙行夾註
+      s.sub!(/(<\/note>)(\n<lb n="0206b29" ed="T"\/>)/, '\2\1')
+    end
+
+    # <milestone unit="juan"> 前面的 lb 屬於新的這一卷
+    s.gsub!(%r{((?:<pb [^>]+>\n?)?(?:<lb [^>]+>\n?)+)(<milestone [^>]*unit="juan"[^/>]*/>)}, '\2\1')
+      
+    doc = Nokogiri::XML(s)
+    doc.remove_namespaces!()
+    doc
+  end
+
+  def read_mod_notes(doc)
+    doc.xpath("//note[@type='mod']").each { |e|
+      @mod_notes << e['n']
+    }
+  end
+
+  def parse_xml(xml_fn)
+    @pass = [false]
+
+    doc = open_xml(xml_fn)
+        
+    e = doc.xpath("//titleStmt/title")[0]
+    @title = traverse(e, 'txt')
+    @title = @title.split()[-1]
+    
+    read_mod_notes(doc)
+
+    root = doc.root()
+    body = root.xpath("text/body")[0]
+    @pass = [true]
+
+    text = traverse(body)
+    text
+  end
+  
+  def remove_empty_nav(node_list)
+    node_list.each do |n|
+      if n[:nav].empty?
+        n.delete(:nav)
+      else
+        remove_empty_nav(n[:nav])
+      end
+    end
+  end
+  
+  def to_html(e)
+    e.to_xml(encoding: 'UTF-8', pertty: true, :save_with => Nokogiri::XML::Node::SaveOptions::AS_XML)
+  end
+
+  def traverse(e, mode='html')
+    r = ''
+    e.children.each { |c| 
+      s = handle_node(c, mode)
+      r += s
+    }
+    r
+  end
+
+end

data/lib/cbeta/p5a_to_html.rb

@@ -5,18 +5,17 @@ require 'json'
 require 'nokogiri'
 require 'set'
 
-# 內容不輸出的元素
-PASS=['back', 'teiHeader']
-
-# 某版用字缺的符號
-MISSING = '－'
-
 # Convert CBETA XML P5a to HTML
 #
 # CBETA XML P5a 可由此取得: https://github.com/cbeta-git/xml-p5a
 #
 # 轉檔規則請參考: http://wiki.ddbc.edu.tw/pages/CBETA_XML_P5a_轉_HTML
 class CBETA::P5aToHTML
+  # 內容不輸出的元素
+  PASS=['back', 'teiHeader']
+
+  # 某版用字缺的符號
+  MISSING = '－'
 
   # @param xml_root [String] 來源 CBETA XML P5a 路徑
   # @param out_root [String] 輸出 HTML 路徑
@@ -25,11 +24,6 @@ class CBETA::P5aToHTML
     @out_root = out_root
     @cbeta = CBETA.new
     @gaijis = CBETA::Gaiji.new
-
-    # 載入 unicode 1.1 字集列表
-    #fn = File.join(File.dirname(__FILE__), 'unicode-1.1.json')
-    #json = File.read(fn)
-    #@unicode1 = JSON.parse(json)
   end
 
   # 將 CBETA XML P5a 轉為 HTML

data/lib/cbeta/p5a_to_simple_html.rb

@@ -22,11 +22,6 @@ class CBETA::P5aToSimpleHTML
     @output_root = output_root
     @cbeta = CBETA.new
     @gaijis = CBETA::Gaiji.new
-
-    # 載入 unicode 1.1 字集列表
-    fn = File.join(File.dirname(__FILE__), 'unicode-1.1.json')
-    json = File.read(fn)
-    @unicode1 = JSON.parse(json)
   end
 
   # 將 CBETA XML P5a 轉為 Text

data/lib/cbeta/p5a_to_text.rb

@@ -25,11 +25,6 @@ class CBETA::P5aToText
     @format = format
     @cbeta = CBETA.new
     @gaijis = CBETA::Gaiji.new
-
-    # 載入 unicode 1.1 字集列表
-    fn = File.join(File.dirname(__FILE__), 'unicode-1.1.json')
-    json = File.read(fn)
-    @unicode1 = JSON.parse(json)
   end
 
   # 將 CBETA XML P5a 轉為 Text

data/lib/data/epub-donate.xhtml

@@ -0,0 +1,23 @@
+<?xml version="1.0" encoding="utf-8" standalone="no"?>
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"
+  "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
+
+<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="zh-TW" xmlns:xml="http://www.w3.org/XML/1998/namespace">
+<head>
+  <link href="../Styles/stylesheet.css" rel="stylesheet" type="text/css" />
+
+  <title>贊助</title>
+</head>
+
+<body>
+  <p><b>歡迎隨喜贊助</b></p>
+
+  <p><b>劃撥捐款</b></p>
+
+  <p>郵政劃撥帳號：１９６２４２２４</p>
+
+  <p>戶名：財團法人智諭老和尚教育紀念基金會</p>
+
+  <p>若欲指定特殊用途者，請特別註明，我們會專款專用。</p>
+</body>
+</html>

data/lib/data/epub-nav.xhtml

@@ -0,0 +1,11 @@
+<html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops">
+<head>
+	<meta charset="utf-8" />
+</head>
+<body>
+<nav epub:type="toc" id="toc">
+  <h1>Table of contents</h1>
+	%s
+</nav>
+</body>
+</html>

data/lib/data/epub-readme.xhtml

@@ -0,0 +1,27 @@
+<?xml version="1.0" encoding="utf-8" standalone="no"?>
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"
+  "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
+
+<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="zh-TW" xmlns:xml="http://www.w3.org/XML/1998/namespace">
+<head>
+  <link href="../Styles/stylesheet.css" rel="stylesheet" type="text/css" />
+
+  <title>說明</title>
+</head>
+
+<body>
+  <div>
+    <h2>編輯說明</h2>
+
+    <ul>
+      <li>本電子書以<a href="http://www.seeland.org.tw/www/zhiyu/index.html">「西蓮淨苑智諭老和尚著作全集」</a>為資料來源。</li>
+
+      <li>漢字呈現以 Unicode 1.1 為基礎，不在此範圍的字則採用 <a href="http://www.cbeta.org/format/rare-rule.php">組字式</a> 表達。</li>
+
+      <li><span style="line-height: 1.6em;">若有發現任何問題，歡迎來函</span> <a href="mailto:seeland77@gmail.com" style="line-height: 1.6em;">seeland77@gmail.com</a> <span style="line-height: 1.6em;">回報。</span><br /></li>
+
+      <li>版權所有，歡迎自由流通，但禁止營利使用。</li>
+    </ul><br />
+  </div>
+</body>
+</html>

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: cbeta
 version: !ruby/object:Gem::Version
-  version: 0.4.3
+  version: 0.5.0
 platform: ruby
 authors:
 - Ray Chou
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-08-20 00:00:00.000000000 Z
+date: 2015-08-22 00:00:00.000000000 Z
 dependencies: []
 description: Ruby gem for use Chinese Buddhist Text resources made by CBETA (http://www.cbeta.org).
 email: zhoubx@gmail.com
@@ -16,16 +16,20 @@ executables: []
 extensions: []
 extra_rdoc_files: []
 files:
-- lib/canons.csv
 - lib/cbeta.rb
 - lib/cbeta/bm_to_text.rb
-- lib/cbeta/gaiji.json
 - lib/cbeta/gaiji.rb
 - lib/cbeta/html_to_text.rb
+- lib/cbeta/p5a_to_epub.rb
 - lib/cbeta/p5a_to_html.rb
 - lib/cbeta/p5a_to_simple_html.rb
 - lib/cbeta/p5a_to_text.rb
-- lib/cbeta/unicode-1.1.json
+- lib/data/canons.csv
+- lib/data/epub-donate.xhtml
+- lib/data/epub-nav.xhtml
+- lib/data/epub-readme.xhtml
+- lib/data/gaiji.json
+- lib/data/unicode-1.1.json
 homepage: https://github.com/RayCHOU/ruby-cbeta
 licenses:
 - MIT