cbeta 1.2.0 → 1.3.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 40faf97b4f5c3b0070145cefbdf86fe7eae39b72
-  data.tar.gz: ebe45fbfc2bf8eb3e025754234ec7e23f6107c49
+  metadata.gz: a21aa8cd67f37fc6d5ca9d87720b1d5d0da3f7f1
+  data.tar.gz: bb7fb6cb3ab5d8b36492cc96728aa5a7fc9a7286
 SHA512:
-  metadata.gz: 2d8bf5130fa58d65fe550c3855ba25b46714871e7e1abfe14d6b8546d3a2ca26e70ef1b23871cfc628b81f357fc83a84177a93a160ead4f6d75dd041c11ccd17
-  data.tar.gz: cc189891bd08585916c2fe6fb0c645d280558a1f214443a0116b5d9595707a6586323c37d5d886a25bbb3b83357e51fba71ccb8505ce053e543c1d7659225135
+  metadata.gz: cb8bbfd96ad22c331e01059303ca539cb6718aae0caa551eba4a8064a7854d9f29bbd3bdb447bd06dd6a6f14b14363d16edb7b4f0932bab7fcfb0feccb9ba9bf
+  data.tar.gz: 2a4394f918a235d08790409488e5dc1a1fd280239d9e1df2ece6dc99628ffc67b73a5beeef199d35a5fe10f4c781cf7ee19910585ff3aa13719b6163b0725d46

data/lib/cbeta.rb

@@ -7,6 +7,7 @@ require 'csv'
 
 class CBETA
   DATA = File.join(File.dirname(__FILE__), 'data')
+  PUNCS = '.[]。，、？「」『』《》＜＞〈〉〔〕［］【】〖〗'
   
   # 將行首資訊轉為引用格式
   #
@@ -23,6 +24,13 @@ class CBETA
     nil
   end
   
+  def self.open_xml(fn)
+    s = File.read(fn)
+    doc = Nokogiri::XML(s)
+    doc.remove_namespaces!()
+    doc
+  end
+  
   # 傳入 蘭札體 缺字碼，傳回 Unicode PUA 字元
   def self.ranjana_pua(gid)
     i = 0x10000 + gid[-4..-1].hex
@@ -100,10 +108,13 @@ class CBETA
   def get_category(book_id)
     @categories[book_id]
   end
+    
 end
 
 require 'cbeta/gaiji'
 require 'cbeta/bm_to_text'
+require 'cbeta/char_count'
+require 'cbeta/char_freq'
 require 'cbeta/html_to_pdf'
 require 'cbeta/p5a_to_epub'
 require 'cbeta/p5a_to_html'

data/lib/cbeta/char_count.rb

@@ -0,0 +1,99 @@
+class CBETA::CharCount
+  def initialize(xml_root)
+    @xml_root = xml_root
+    @result = {}
+  end
+  
+  def char_count(canon=nil)
+    stat_all if canon.nil?
+    stat_canon(canon)
+    @result
+  end
+  
+  private
+    
+  def handle_node(e)
+    return if e.comment?
+    return handle_text(e) if e.text?
+    return if %w(foreign mulu rdg reg sic).include? e.name
+    
+    case e.name
+    when 'g'    then @result[@work] += 1
+    when 'note' then handle_note(e)
+    when 't'    then handle_t(e)
+    else traverse(e)
+    end
+  end
+  
+  def handle_note(e)
+    if %w(inline interlinear).include? e['place']
+      traverse(e)
+    end
+  end
+  
+  def handle_t(e)
+    if e.has_attribute? 'place' and e['place'].include? 'foot'
+      return
+    end
+    traverse(e)
+  end
+  
+  def handle_text(e)
+    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
+    s.gsub!(/[\n\r]/, '')
+    
+    @result[@work] += s.size
+  end
+      
+  def stat_all
+    Dir.entries(@xml_root).sort.each do |canon|
+      next if canon.start_with? '.'
+      next if canon == 'schema'
+      stat_canon(canon)
+    end
+  end
+  
+  def stat_canon(canon)
+    return if canon.nil?
+    puts 'stat canon: ' + canon
+    folder = File.join(@xml_root, canon)
+    Dir.entries(folder).sort.each do |vol|
+      next if vol.start_with? '.'
+      p = File.join(folder, vol)
+      stat_vol(p)
+    end
+  end
+  
+  def stat_file(fn)
+    @work = File.basename(fn, '.xml')
+    @work.sub!(/^([A-Z])\d{2,3}n(.*)$/, '\1\2')
+    @work = 'T0220' if @work.start_with?('T0220')
+    unless @result.key? @work
+      puts "stat work: #{@work}"
+      @result[@work] = 0 
+    end
+    
+    doc = CBETA.open_xml(fn)
+    body = doc.at_xpath('/TEI/text/body')
+    traverse(body)
+  end
+  
+  def stat_vol(vol_folder)
+    Dir.entries(vol_folder).sort.each do |f|
+      next if f.start_with? '.'
+      p = File.join(vol_folder, f)
+      stat_file(p)
+    end
+  end
+  
+  def traverse(e)
+    e.children.each { |c| 
+      handle_node(c)
+    }
+  end
+  
+end

data/lib/cbeta/char_freq.rb

@@ -0,0 +1,123 @@
+class CBETA::CharFrequency
+  # @option opts [Integer] :top
+  def initialize(xml_root, opts={})
+    @xml_root = xml_root
+    @config = {
+      top: 10
+    }
+    @config.merge!(opts)
+    @result = {}
+    @current = @result
+  end
+  
+  def char_freq(canon=nil)
+    stat_all if canon.nil?
+    stat_canon(canon)
+    r = @result.sort_by {|k,v| v}
+    r[(0-@config[:top])..-1].reverse
+  end
+  
+  private
+  
+  def count(c)
+    if @current.key? c
+      @current[c] += 1
+    else
+      @current[c] = 1
+    end
+  end
+  
+  def handle_node(e)
+    return if e.comment?
+    return handle_text(e) if e.text?
+    return if %w(foreign mulu rdg reg sic).include? e.name
+    
+    case e.name
+    when 'g'    then count(e['ref'])
+    when 'note' then handle_note(e)
+    when 't'    then handle_t(e)
+    else traverse(e)
+    end
+  end
+  
+  def handle_note(e)
+    if %w(inline interlinear).include? e['place']
+      traverse(e)
+    end
+  end
+  
+  def handle_t(e)
+    if e.has_attribute? 'place' and e['place'].include? 'foot'
+      return
+    end
+    traverse(e)
+  end
+  
+  def handle_text(e)
+    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
+    s.gsub!(/[\n\r]/, '')
+
+    s.each_char do |c|
+      next if CBETA::PUNCS.include? c
+      count(c)
+    end
+  end
+      
+  def stat_all
+    Dir.entries(@xml_root).sort.each do |canon|
+      next if canon.start_with? '.'
+      next if canon == 'schema'
+      stat_canon(canon)
+    end
+  end
+  
+  def stat_canon(canon)
+    return if canon.nil?
+    puts 'stat canon: ' + canon
+    if @config[:group_by] == 'canon'
+      @result[canon] = {}
+      @current = @result[canon]
+    end
+    folder = File.join(@xml_root, canon)
+    Dir.entries(folder).sort.each do |vol|
+      next if vol.start_with? '.'
+      p = File.join(folder, vol)
+      stat_vol(p)
+    end
+  end
+  
+  def stat_file(fn)
+    if @config[:group_by] == 'work'
+      work = File.basename(fn, '.xml')
+      work.sub!(/^([A-Z])\d{2,3}n(.*)$/, '\1\2')
+      work = 'T0220' if work.start_with?('T0220')
+      puts "stat work: #{work}"
+      @result[work] = {}
+      @current = @result[work]
+    else
+      puts "stat file: #{fn}"
+    end
+    doc = CBETA.open_xml(fn)
+    body = doc.at_xpath('/TEI/text/body')
+    traverse(body)
+  end
+  
+  def stat_vol(vol_folder)
+    Dir.entries(vol_folder).sort.each do |f|
+      next if f.start_with? '.'
+      p = File.join(vol_folder, f)
+      stat_file(p)
+    end
+  end
+  
+  def traverse(e)
+    e.children.each { |c| 
+      handle_node(c)
+    }
+  end
+  
+end

data/lib/cbeta/p5a_to_html_for_pdf.rb

@@ -386,7 +386,7 @@ class CBETA::P5aToHTMLForPDF
   end
 
   def handle_p(e)
-    "<p>%s</p>\n" % traverse(e)
+    "<div class='p'>%s</div>\n" % traverse(e)
   end
 
   def handle_row(e)

data/lib/data/html-for-pdf.css

@@ -1,3 +1,11 @@
+@font-face {
+    font-family: 'Songti';
+    /* src: url('/Library/Fonts/Songti.ttc'); */
+		src: url('/Library/Fonts/华文仿宋.ttf');
+}
+body {
+	font-family: Songti, PMingLiU-ExtB;
+}
 div.lg {
 	display: table;
 }

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: cbeta
 version: !ruby/object:Gem::Version
-  version: 1.2.0
+  version: 1.3.1
 platform: ruby
 authors:
 - Ray Chou
@@ -18,6 +18,8 @@ extra_rdoc_files: []
 files:
 - lib/cbeta.rb
 - lib/cbeta/bm_to_text.rb
+- 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