cbeta 3.8.0 → 4.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: fdc6811da84a63c0696ba421fe426e21de327089cfa018cc8ec3252b7981caed
-  data.tar.gz: 77ee88f62ae4ac35506ca756c7df764a5d35fcf1b2d8ce98a05c578873ef3164
+  metadata.gz: 0a8826c19847f81602cd7ca3fb1e976fffc20b98976fde5844134b4672d60e63
+  data.tar.gz: 06a2dc151b6623439cca6bbbf4a913fb70c1ac1f6fe8a041e8b052471b08c202
 SHA512:
-  metadata.gz: 9d5e52c2087cc1d4c05fb8d0757dd1e1bb8d4e9a42a1d2aeb6c11f3455358d018380a777a7aeed914be3a4474dd12060222fbaa4118bf82fc69e9bf3cab7d26e
-  data.tar.gz: abb54c0eaa239d7545e02ec957ec0648bc776a3d4c9f27f18a130d7f09164ea9ed0a7fbf9a0cac4480bdf98eec2282f94ef6b737ab8af659e6cd564212ee1755
+  metadata.gz: 102a0d53249646d923f2b95c1b5906c23ad6a243b31827e8c663b795448277c1cc4295d9340cdc79f73fc5c29cfc03430c0698cbae69ad4f618d051a0eab8dd4
+  data.tar.gz: 48f777330445f72f0e80c8a02c862b05a36e1ba8da6c4509ac81be5a7cab77fed0b71c8dc392b66598f47eaf438e85fe4b44eeb0401e9e6b7571e4ced99501f8

data/lib/cbeta/p5a_checker.rb

@@ -21,6 +21,7 @@ require_relative 'cbeta_share'
 #   * [E14] <anchor type="right"> 不應直接出現在 div 或 body 下
 #   * [E15] <note> corresp 無對應的 <note>
 #   * [E16] app/@n 與 note/@n 不一致
+#   * [E17] list 下沒有任何有內容的 item
 #
 # * 警告類型
 #   * [W01] 夾注包夾注
@@ -29,6 +30,9 @@ require_relative 'cbeta_share'
 class CBETA::P5aChecker
   ALLOW_TAB = %w[change]
 
+  # 判斷 item 是否有內容時，不計入的元素（本身不產生內容，只是標記位置）
+  EMPTY_ELEMENTS = %w[lb pb milestone anchor mulu space]
+
   # @param xml_root [String] 來源 CBETA XML P5a 路徑
   # @param figures [String] 插圖 路徑 (可由 https://github.com/cbeta-git/CBR2X-figures 取得)
   # @param log [String] Log file path
@@ -119,6 +123,12 @@ class CBETA::P5aChecker
     end
   end
 
+  # 節點是否有內容（文字，或 EMPTY_ELEMENTS 以外的元素）
+  def content?(e)
+    return true unless e.text.strip.empty?
+    e.xpath('.//*').any? { |d| not EMPTY_ELEMENTS.include?(d.name) }
+  end
+
   def display_errors
     @g_errors.keys.sort.each do |k|
       s = @g_errors[k].to_a.join(',')
@@ -230,6 +240,20 @@ class CBETA::P5aChecker
     traverse(e)
   end
 
+  # list 下應該至少有一個 item 是有內容的
+  def e_list(e)
+    # item 可能包在 app/lem 之類的元素裡，所以用 descendant 找，
+    # 但要排除 巢狀 list 底下的 item
+    items = e.xpath('.//item').select { |item| nearest_list(item).equal?(e) }
+    unless items.any? { |item| content?(item) }
+      msg = "[E17] list 下沒有任何有內容的 item"
+      head = e.at_xpath('head')
+      msg += ", head: #{head.text.gsub(/\s+/, ' ').strip[0, 30]}" unless head.nil?
+      error msg
+    end
+    traverse(e)
+  end
+
   def e_note(e)
     error "[E11] note 直接出現在 div 下" if e.parent.name == 'div'
     error "[E12] note 直接出現在 lg 下" if e.parent.name == 'lg'
@@ -342,6 +366,7 @@ class CBETA::P5aChecker
     when 'item'    then e_item(e)
     when 'lb'      then e_lb(e)
     when 'lem'     then e_lem(e)
+    when 'list'    then e_list(e)
     when 'note'    then e_note(e)
     when 'p'       then e_p(e)
     when 'rdg'     then e_rdg(e)
@@ -360,6 +385,13 @@ class CBETA::P5aChecker
     end
   end
   
+  # 往上找最近的 list 祖先
+  def nearest_list(e)
+    r = e.parent
+    r = r.parent while r.element? and r.name != 'list'
+    r
+  end
+
   def read_notes(doc)
     @notes = Set.new
     doc.xpath('//note').each do |e|

data/lib/cbeta.rb

@@ -312,14 +312,8 @@ require 'cbeta/bm_to_text'
 require 'cbeta/canon'
 require 'cbeta/char_count'
 require 'cbeta/char_freq'
-require 'cbeta/html_to_pdf'
 require 'cbeta/p5a_checker'
-require 'cbeta/p5a_to_html'
-require 'cbeta/p5a_to_html_for_every_edition'
-require 'cbeta/p5a_to_html_for_pdf'
-require 'cbeta/p5a_to_simple_html'
 require 'cbeta/p5a_to_text'
 require 'cbeta/p5a_validator'
-require 'cbeta/html_to_text'
 require 'cbeta/unicode_service'
 require 'cbeta/xml_document'

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: cbeta
 version: !ruby/object:Gem::Version
-  version: 3.8.0
+  version: 4.1.0
 platform: ruby
 authors:
 - Ray Chou
@@ -109,13 +109,7 @@ files:
 - lib/cbeta/char_count.rb
 - lib/cbeta/char_freq.rb
 - lib/cbeta/gaiji.rb
-- lib/cbeta/html_to_pdf.rb
-- lib/cbeta/html_to_text.rb
 - lib/cbeta/p5a_checker.rb
-- lib/cbeta/p5a_to_html.rb
-- lib/cbeta/p5a_to_html_for_every_edition.rb
-- lib/cbeta/p5a_to_html_for_pdf.rb
-- lib/cbeta/p5a_to_simple_html.rb
 - lib/cbeta/p5a_to_text.rb
 - lib/cbeta/p5a_validator.rb
 - lib/cbeta/unicode_service.rb
@@ -124,9 +118,6 @@ files:
 - lib/data/categories.json
 - lib/data/cbeta_gaiji.json
 - lib/data/cbeta_sanskrit.json
-- lib/data/html-for-pdf.css
-- lib/data/pdf-template.htm
-- lib/data/unicode-1.1.json
 homepage: https://github.com/RayCHOU/ruby-cbeta
 licenses:
 - MIT

data/lib/cbeta/html_to_pdf.rb

@@ -1,75 +0,0 @@
-require 'fileutils'
-
-class CBETA::HTMLToPDF
-  # @param input [String] folder of source HTML, HTML can be produced by CBETA::P5aToHTMLForPDF.
-  # @param output [String] output folder
-  # @param converter [String] shell command to convert HTML to PDF
-  #   * suggestion: http://www.princexml.com/
-  #   * wkhtmltopdf has font problem to display unicode extb characters
-  #
-  # @example
-  #   c = CBETA::HTMLToPDF.new('/temp/cbeta-html', '/temp/cbeta-pdf', "prince %{in} -o %{out}")
-  def initialize(input, output, converter)
-    @input = input
-    @output = output
-    @converter = converter
-  end
-  
-  # Convert CBETA HTML to PDF
-  #
-  # @example for convert Taisho (大正藏) Volumn 1:
-  #
-  #   c = CBETA::HTMLToPDF.new('/PATH/TO/CBETA/XML/P5a', '/OUTPUT/FOLDER')
-  #   c.convert('T01')
-  #
-  # @example for convert all in Taisho (大正藏):
-  #
-  #   c = CBETA::HTMLToPDF.new('/PATH/TO/CBETA/XML/P5a', '/OUTPUT/FOLDER')
-  #   c.convert('T')
-  #
-  # @example for convert Taisho Vol. 5~7:
-  #
-  #   c = CBETA::P5aToHTMLForPDF.new('/PATH/TO/CBETA/XML/P5a', '/OUTPUT/FOLDER')
-  #   c.convert('T05..T07')
-  #
-  # T 是大正藏的 ID, CBETA 的藏經 ID 系統請參考: http://www.cbeta.org/format/id.php
-  def convert(target=nil)
-    return convert_all if target.nil?
-
-    arg = target.upcase
-    if arg.size <= 2
-      convert_collection(arg)
-    else
-      if arg.include? '..'
-        arg.match(/^([^\.]+?)\.\.([^\.]+)$/) {
-          convert_vols($1, $2)
-        }
-      else
-        convert_vol(arg)
-      end
-    end
-  end
-  
-  def convert_collection(c)
-    @canon = c
-    puts 'convert_collection ' + c
-    
-    output_folder = File.join(@output, @canon)
-    FileUtils.mkdir_p(output_folder) unless Dir.exist? output_folder
-    
-    folder = File.join(@input, @canon)
-    Dir.foreach(folder) { |f|
-      next if f.start_with? '.'
-      src = File.join(folder, f, 'main.htm')
-      dest = File.join(output_folder, "#{f}.pdf")
-      convert_file(src, dest)
-    }
-  end
-  
-  def convert_file(html_fn, pdf_fn)
-    puts "convert file: #{html_fn} to #{pdf_fn}"
-    cmd = @converter % { in: html_fn, out: pdf_fn}
-    `#{cmd}`
-  end
-    
-end

data/lib/cbeta/html_to_text.rb

@@ -1,151 +0,0 @@
-require 'fileutils'
-require 'nokogiri'
-
-# 將 CBETA HTML 轉為 純文字(含行首資訊)
-#
-# Example:
-#
-#   h2t = CBETA::HTMLToText.new('/temp/cbeta-html', '/temp/cbeta-text')
-#   h2t.convert("T01")  # 轉換大正藏第一冊
-class CBETA::HTMLToText
-  # @param html_root [String] 來源 HTML 路徑
-  # @param out_root [String] 輸出路徑
-  def initialize(html_root, out_root)
-    @html_root = html_root
-    @out_root = out_root
-  end
-
-  # @param arg [String] 要執行轉換的冊數
-  # @example
-  #   convert("T01")
-  def convert(arg)
-    @dirty = false
-    @vol = arg.upcase
-    @corpus = @vol[0]
-    handle_vol
-  end
-
-  private
-
-  def traverse(e)
-    r = ''
-    e.children.each { |c|
-      r += handle_node(c)
-    }
-    r.gsub('　', '')
-  end
-
-  def handle_text(e)
-    s = e.content().chomp
-    return '' if s.empty?
-    s.gsub(/[\n，、—！。：「]/, '')
-  end
-
-  def handle_span(e)
-    r = ''
-    case e['class']
-    when 'doube-line-note'
-      r = traverse(e)
-      unless r.start_with? '（'
-        r = "(#{r})"
-      end
-    when 'lb'
-      if @dirty
-        r += "\n"
-      else
-        @dirty = true
-      end
-      # 行首資訊 T05n0220a 改為 T05n0220
-      lb = e['id'].sub(/^(T0\dn0220)[a-z](.*)$/, '\1\2')
-      r += lb + '║'
-    when 'lineInfo'
-    when 'ranja'
-      r = '【◇】'
-    when 'siddam'
-      r = '【◇】'
-    when 'star'
-    else
-      r = traverse(e)
-    end
-    r
-  end
-
-  def handle_node(e)
-    return '' if e.comment?
-    return handle_text(e) if e.text?
-    r = ''
-    case e.name
-    when 'a'
-      if e['class'] == 'gaijiAnchor'
-        id = e['href'][1..-1]
-        r = @gaiji[id]
-      else
-        r = traverse(e)
-      end
-    when 'div'
-      if e['id'] != 'back'
-        r = traverse(e)
-      end
-    when 'head'
-    when 'p'
-      if e['class'] == 'figure'
-        r = '【圖】'
-      else
-        r = traverse(e)
-      end
-    when 'span'
-      r = handle_span(e)
-    else
-      r = traverse(e)
-    end
-    r
-  end
-
-  def prepare_folder()
-    folder = File.join(@out_root, @corpus, @vol)
-    FileUtils.remove_dir(folder, true)
-    FileUtils.mkdir_p(folder)
-    folder
-  end
-
-  def handle_file(path)
-    sutra = File.basename(path, ".*")
-    sutra.sub!(/^(.*)_.*$/, '\1')
-    sutra.sub!(/(T\d\dn0220).*$/, '\1') # T0220 BM 沒有分 a, b, c...
-
-    if sutra != @last_sutra
-      txt_fn = sutra + '.txt'
-      txt_path = File.join(@folder_out, txt_fn)
-      puts "h2t #{txt_path}"
-      @fo = File.open(txt_path, 'w')
-      @last_sutra = sutra
-      @dirty = false
-    end
-
-    f = File.open(path)
-    doc = Nokogiri::HTML(f)
-    f.close
-
-    @gaiji = {}
-    doc.css("span.gaijiInfo").each { |e|
-      @gaiji[e['id']] = e['zzs']
-    }
-
-    text = traverse(doc.root)
-
-    # 悉曇字
-    text.gsub!(/(\((【◇】)+\)|（【◇】）|【◇】)+/, '【◇】')
-
-    @fo.write(text)
-  end
-
-  def handle_vol()
-    folder_in = File.join(@html_root, @corpus, @vol)
-    @folder_out = prepare_folder
-    @last_sutra = ''
-    Dir["#{folder_in}/*"].each { |f|
-      handle_file(f)
-    }
-  end
-
-end