cbeta 0.2.1 → 0.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 885182f24b433d1627754e78f5cd2cc21729122e
-  data.tar.gz: e02e81586409357be195a85e195e5f3db289a9d4
+  metadata.gz: f10484530e1c47830c305dfd75004aeae5fa24e5
+  data.tar.gz: e634ef5b83ce81564c1c9e916370a1b024a8a179
 SHA512:
-  metadata.gz: 1b589734bd657cce12e4f20cf1a66b8650cc09a788b2bed3dcef623ffb4a170c82c7e96b2c0211ce0bea543025583cb212fb1dcd779a3087488080ca7f9c4874
-  data.tar.gz: a839f8c28d5615832edebe4c7a87dd2ce064d0b9fd0ebdf5e8ca0f6211630697b2f4d0f0a2d18dc1305393edc2b531ef303ab6585c3541ed89c6e8c2f92b903f
+  metadata.gz: a498957ee9d982b5bde85ddd44a1671297c18acad54ffe2c1b9c48eb196f70a1916d2a7d769ec9c0b7a8a8b65c77f43d60e888df3dbb35fe5236af21eb748574
+  data.tar.gz: 173d4d21a6d36088ff57448e2da79b7a9f1bd5a6d5813cd23e06aabacfc39c83637cdb3248bc20b3aa15147b65fe726bff011f58a6a109b8ca2edcd1b360e939

data/lib/cbeta.rb

@@ -6,7 +6,6 @@
 require 'csv'
 
 class CBETA
-
   # 將行首資訊轉為引用格式
   #
   # @param linehead [String] 行首資訊, 例如：T85n2838_p1291a03

data/lib/cbeta/p5a_to_text.rb

@@ -8,13 +8,21 @@ require 'set'
 # Convert CBETA XML P5a to Text
 #
 # CBETA XML P5a 可由此取得: https://github.com/cbeta-git/xml-p5a
+#
+# @example for convert 大正藏第一冊 in app format:
+#
+#   c = CBETA::P5aToText.new('/PATH/TO/CBETA/XML/P5a', '/OUTPUT/FOLDER', 'app')
+#   c.convert('T01')
+#
 class CBETA::P5aToText
 
   # @param xml_root [String] 來源 CBETA XML P5a 路徑
   # @param output_root [String] 輸出 Text 路徑
-  def initialize(xml_root, output_root)
+  # @param format [String] 輸出格式，例：'app'
+  def initialize(xml_root, output_root, format=nil)
     @xml_root = xml_root
     @output_root = output_root
+    @format = format
     @cbeta = CBETA.new
     @gaijis = CBETA::Gaiji.new
 
@@ -61,6 +69,42 @@ class CBETA::P5aToText
 
   private
 
+  # 跨行字詞移到下一行
+  def appify(text)
+    r = ''
+    i = 0
+    app = ''
+    text.each_line do |line|
+      line.chomp!
+      if line.match(/^(.*)║(.*)$/)
+        r += $1
+        t = $2
+        r += "(%02d)" % i
+        r += "║#{app}"
+        app = ''
+        i = 0
+        chars = t.chars
+        until chars.empty?
+          c = chars.pop
+          if c == "\t"
+            break
+          elsif ' 　：》」』、；，！？。'.include? c
+            chars << c
+            break
+          elsif '《「『'.include? c  # 這些標點移到下一行
+            app = c + app
+            break
+          else
+            app = c + app
+          end
+        end
+        r += chars.join.gsub(/\t/, '') + "\n"
+        i = app.size
+      end
+    end
+    r
+  end
+
   def convert_all
     Dir.foreach(@xml_root) { |c|
       next unless c.match(/^[A-Z]$/)
@@ -83,11 +127,15 @@ class CBETA::P5aToText
   end
 
   def handle_byline(e)
-    traverse(e) + "\n"
+    r = traverse(e)
+    r += @format=='app' ? "\t" : "\n"
+    r
   end
 
   def handle_cell(e)
-    traverse(e) + "\n"
+    r = traverse(e)
+    r += @format=='app' ? "\t" : "\n"
+    r
   end
 
   def handle_collection(c)
@@ -109,11 +157,15 @@ class CBETA::P5aToText
   end
 
   def handle_docNumber(e)
-    traverse(e) + "\n"
+    r = traverse(e)
+    r += @format == 'app' ? "\t" : "\n"
+    r
   end
 
   def handle_figure(e)
-    traverse(e) + "\n"
+    r = traverse(e)
+    r += @format == 'app' ? "\t" : "\n"
+    r
   end
 
   def handle_g(e)
@@ -155,27 +207,37 @@ class CBETA::P5aToText
   end
 
   def handle_head(e)
-    traverse(e) + "\n"
+    r = traverse(e)
+    r += @format == 'app' ? "\t" : "\n"
+    r
   end
 
   def handle_item(e)
-    traverse(e) + "\n"
+    r = traverse(e)
+    r += @format == 'app' ? "\t" : "\n"
   end
 
   def handle_juan(e)
-    traverse(e) + "\n"
+    r = traverse(e)
+    r += @format == 'app' ? "\t" : "\n"
+    r
   end
 
   def handle_l(e)
     r = traverse(e)
-    unless @lg_type == 'abnormal'
-      r += "\n"
+    if @format == 'app'
+      r += "\t"
+    else
+      r += "\n" unless @lg_type == 'abnormal'
     end
     r
   end
 
   def handle_lb(e)
     r = ''
+    if @format == 'app'
+      r += "\n#{e['n']}║"
+    end
     unless @next_line_buf.empty?
       r += @next_line_buf + "\n"
       @next_line_buf = ''
@@ -197,7 +259,9 @@ class CBETA::P5aToText
   end
 
   def handle_list(e)
-    "\n" + traverse(e)
+    r = ''
+    r += "\n" unless @format == 'app'
+    r + traverse(e)
   end
 
   def handle_milestone(e)
@@ -247,6 +311,7 @@ class CBETA::P5aToText
     when 'row'       then handle_row(e)
     when 'sic'       then handle_sic(e)
     when 'sg'        then handle_sg(e)
+    when 'tt'        then handle_tt(e)
     when 't'         then handle_t(e)
     when 'table'     then handle_table(e)
     when 'teiHeader' then ''
@@ -264,7 +329,9 @@ class CBETA::P5aToText
   end
 
   def handle_p(e)
-    traverse(e) + "\n"
+    r = traverse(e)
+    r += @format == 'app' ? "\t" : "\n"
+    r
   end
 
   def handle_rdg(e)
@@ -337,6 +404,9 @@ class CBETA::P5aToText
     end
     r = traverse(e)
 
+    # 不是雙行對照
+    return r if @tt_type == 'app'
+
     # 處理雙行對照
     i = e.xpath('../t').index(e)
     case i
@@ -366,6 +436,11 @@ class CBETA::P5aToText
     CGI.escapeHTML(r)
   end
 
+  def handle_tt(e)
+    @tt_type = e['type']
+    traverse(e)
+  end
+
   def handle_vol(vol)
     puts "convert volumn: #{vol}"
 
@@ -427,13 +502,15 @@ class CBETA::P5aToText
           node.remove
         end
       end
+      text = frag.content
+      text = appify(text) if @format == 'app'
 
       folder = File.join(@out_sutra, ed)
       FileUtils.makedirs(folder)
 
       fn = "#{@sutra_no}_%03d.txt" % juan_no
       output_path = File.join(folder, fn)
-      File.write(output_path, frag.content)
+      File.write(output_path, text)
     end
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: cbeta
 version: !ruby/object:Gem::Version
-  version: 0.2.1
+  version: 0.3.0
 platform: ruby
 authors:
 - Ray Chou
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-07-13 00:00:00.000000000 Z
+date: 2015-07-14 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