cbeta 3.7.10 → 4.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: fa8c0001a52f96df16ac52d46416bcf1630f841ea306940a3196e81ece5c1fac
-  data.tar.gz: 655f8c0438ff07cfd0b73f3ad1f749cde2b44c38c7567a11dc2dcd9d31b5219f
+  metadata.gz: 346145ac7ac767e90b736501a231bc7fb92d70eeff529dbc64f0d05c59bb4714
+  data.tar.gz: 9dd75ba7fbb1280e2cede956f6694193eb6f7e4d22d667a3c055dd35a7ced5c1
 SHA512:
-  metadata.gz: a4d29529c018af7658c30da3350e75e295181df0ea86693be3ed8f63257fdbe7d1efcf6fbf311965de718fbf589ea8971c20eca3f8a296cad5a1c4b2d8205871
-  data.tar.gz: b2f23a65ebc368bd5b7d19c21b1da96ba84fa4e21cebd7f2a32651f9ba571cc6fd76463d2483af619550a2cfcfafc2342ff07438a39e64246900b907904d7a24
+  metadata.gz: 521d9a7d25265961a650d7c1d8171bc3742872cca2bfdaa46d8fd93c2d07bd2ba10dd1f67822c6358c21a4cd050683ca1f63151998aac36c4a425e9e24dc6e57
+  data.tar.gz: e30e68df19b306e9472faa679da95ba4fc61862756748e0bebde6422cdeb3d4975b78b3850953dd0ef7af806b9e1c992daa6d4f293b59accc1ec7716d68763df

data/lib/cbeta/cbeta_share.rb

@@ -1,7 +1,7 @@
 module CbetaShare
   def each_canon(xml_root)
     Dir.entries(xml_root).sort.each do |c|
-      next unless c.match(/^#{CBETA::CANON}$/)
+      next unless c.match(CBETA::CANON_ID)
       yield(c)
     end
   end

data/lib/cbeta/p5a_checker.rb

@@ -192,9 +192,9 @@ class CBETA::P5aChecker
   
   def e_graphic(e)
     url = File.basename(e['url'])
-    fn = File.join(@figures, @canon, url)
-    unless File.exist? fn
-      error "[E05] 圖檔 不存在, url: #{url}"
+    path = File.join(@figures, @canon, url)
+    unless File.exist? path
+      error "[E05] 圖檔 不存在, url: #{e['url']}, path: #{path}"
     end
   end
 

data/lib/cbeta/p5a_to_text.rb

@@ -124,7 +124,7 @@ class CBETA::P5aToText
 
   def convert_all
     Dir.entries(@xml_root).sort.each do |c|
-      next unless c.match(/^#{CBETA::CANON}$/)
+      next unless c.match(CBETA::CANON_ID)
       handle_canon(c)
     end
   end

data/lib/cbeta.rb

@@ -6,22 +6,36 @@
 require 'csv'
 
 class CBETA
-  CANON = 'CC|DA|GA|GB|LC|TX|YP|ZS|ZW|[A-Z]'
+  # 藏經 ID, 例如 T, ZS
+  #
+  # 本身是 alternation, 已經包在 non-capturing group 裡,
+  # 所以插值到 regexp 時不需要 (也不應該) 再包一層 group:
+  #   正確: /\A#{CBETA::CANON}\z/
+  #   多餘: /\A(?:#{CBETA::CANON})\z/
+  # 若需要取出藏經 ID, 才用 capturing group 包起來: /^(#{CBETA::CANON})/
+  CANON = '(?:CC|DA|GA|GB|LC|TX|YP|ZS|ZW|[A-Z])'
   SORT_ORDER = %w(T X A K S F C D U P J L G M N ZS I ZW B GA GB Y LC TX YP CC)
   VOL3 = %w[A CC C G GA GB L M P U]
   DATA = File.join(File.dirname(__FILE__), 'data')
   PUNCS = ',.()[] 。‧．，、；？！：︰／（）「」『』《》＜＞〈〉〔〕［］【】〖〗〃…—─　～│┬▆△＊＋－＝'
-  
+
   # 經號 (不含 Canon ID):
   #   四碼數字 T0001
   #   四碼數字 + 英文字母 T0150A, T0128a
   #   英文字母 + 三碼數字 JA041, ZWa073
-  WORK_PART = '\d{4}[a-zA-Z]?|[ABa]\d{3}'
+  # 與 CANON 一樣已自帶 non-capturing group。
+  WORK_PART = '(?:\d{4}[a-zA-Z]?|[ABa]\d{3})'
 
   # XML file 主檔名
   # GA010n0009
-  BASENAME = "(?:#{CANON})\\d{2,3}n(?:#{WORK_PART})"
-  
+  BASENAME = "#{CANON}\\d{2,3}n#{WORK_PART}"
+
+  # 整個字串就是 藏經 ID, 例如 "T", "ZS"
+  CANON_ID = /\A#{CANON}\z/
+
+  # 整個字串就是 典籍編號, 例如 "T0001", "JB271", "ZWa073"
+  WORK_ID = /\A#{CANON}#{WORK_PART}\z/
+
   # 由 行首資訊 取得 藏經 ID
   # @param linehead[String] 行首資訊, 例如 "T01n0001_p0001a01" 或 "GA009n0008_p0003a01"
   # @return [String] 藏經 ID，例如 "T" 或 "GA"
@@ -142,7 +156,7 @@ class CBETA
   #   ex: J36nB348_p0284c01
   # @return [String] XML檔相對路徑，例如 "GA/GA009/GA009n0008.xml"
   def self.linehead_to_xml_file_path(linehead)
-    if m = linehead.match(/^(?<work>(?<vol>(?<canon>#{CANON})\d+)n(?:#{WORK_PART})).*$/)
+    if m = linehead.match(/^(?<work>(?<vol>(?<canon>#{CANON})\d+)n#{WORK_PART}).*$/)
       File.join(m[:canon], m[:vol], m[:work]+'.xml')
     else
       nil
@@ -172,7 +186,7 @@ class CBETA
   #   CBETA.linehead_to_s('T85n2838_p1291a03')
   #   # return "T85, no. 2838, p. 1291, a03"
   def self.linehead_to_s(linehead)
-    linehead.match(/^((?:#{CANON})\d+)n(.*)_p(\d+)([a-z]\d+)$/) {
+    linehead.match(/^(#{CANON}\d+)n(.*)_p(\d+)([a-z]\d+)$/) {
       return "#{$1}, no. #{$2}, p. #{$3}, #{$4}"
     }
     nil
@@ -298,14 +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: 3.7.10
+  version: 4.0.0
 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