unihan2 0.1.2 → 0.2.0

data/lib/unihan2.rb

@@ -1,24 +1,25 @@
+require 'csv'
+
 class Unihan2
   DATA_DIR = File.join(File.dirname(__FILE__), '../data')
   
   def initialize
-    fn = File.join(DATA_DIR, 'Unihan_DictionaryLikeData.txt')
-    @strokes = {}
-    File.foreach(fn) do |line|
-      next if line.start_with? '#'
-      line.chomp!
-      cells = line.split("\t")
-      if cells[1] == 'kTotalStrokes'
-        c = [cells[0].sub(/^U\+(.*)$/, '\1').hex].pack('U')
-        i = cells[2].to_i
-        @strokes[c] = i
-      end
-    end
+    read_strokes
+    read_version
+  end
+
+
+  # return total strokes of the character char
+  # @param char [String] the character
+  # @return [Integer] the total strokes
+  def strokes(char)
+    @strokes[char]
   end
 
-  # Listing of Characters Covered by the Unihan Database
-  #   https://www.unicode.org/reports/tr38/tr38-29.html#BlockListing
-  def self.ver(code)
+  # return unicode version of specific character
+  # @param code [String] character or codepoing
+  # @return [Float] unicode version
+  def ver(code)
     return nil if code.nil?
 
     if code.is_a? Integer
@@ -29,237 +30,53 @@ class Unihan2
       i = code.hex
     end
 
-    case i
-    when 0..0xFFFF        then uv0(i)
-    when 0x20000..0x2FFFF then uv2(i)
-    when 0x30000..0x3134A then 13.0
-    end
-  end
-
-  # return total strokes of the character char
-  # @param char [String] the character
-  # @return [Integer] the total strokes
-  def strokes(char)
-    @strokes[char]
+    ver_bsearch(i, 0, @vers.size-1)
   end
 
   private
 
-  def self.uv0(i)
-    case i
-    when 0..0x0FFF then uv00(i)
-    when 0x1E00..0x1EFF
-      case i
-      when 0x1E00..0x1E9A then 1.1
-      when 0x1E9B         then 2.0
-      when 0x1E9C..0x1E9F then 5.1
-      when 0x1EA0..0x1EF9 then 1.1
-      when 0x1EFA..0x1EFF then 5.1
-      end
-    when 0x2000..0x2FFF then uv02(i)
-    when 0x3000..0x33FF then uv03(i)
-    when 0x3400..0x9FA5
-      case i
-      when 0x3400..0x4DB5 then 3.0 # CJK Extension A 中日韓統一表意文字擴充 A 區
-      when 0x4DB6..0x4DBF then 13.0 # Extension A
-      when 0x4E00..0x9FA5 then 1.1 # CJK Unified Ideographs
-      end
-    when 0x9FA6..0x9FFF
-      case i
-      when 0x9FA6..0x9FBB then 4.1 # CJK Unified Ideographs
-      when 0x9FBC..0x9FC3 then 5.1 # CJK Unified Ideographs
-      when 0x9FC4..0x9FCB then 5.2 # CJK Unified Ideographs
-      when 0x9FCC         then 6.1 # CJK Unified Ideographs
-      when 0x9FCD..0x9FD5 then 8.0 # CJK Unified Ideographs
-      when 0x9FD6..0x9FEA then 10.0 # CJK Unified Ideographs
-      when 0x9FEB..0x9FEF then 11.0 # CJK Unified Ideographs
-      when 0x9FF0..0x9FFC then 13.0 # CJK Unified Ideographs
-      end
-    when 0xA000..0xDFFF
-      case i
-      when 0xA000..0xA48C then 3.0 # Yi Syllables 彝族文字區
-      when 0xAC00..0xD7A3 then 2.0 # Hangul Syllables 韓文拼音
-      end
-    when 0xF000..0xFFFF
-      case i
-      when 0xF900..0xFA2D then 1.1 # CJK Compatibility Ideographs
-      when 0xFA2E..0xFA2F then 6.1 # CJK Compatibility Ideographs
-      when 0xFA30..0xFA6A then 3.2 # CJK Compatibility Ideographs
-      when 0xFA6B..0xFA6D then 5.2 # CJK Compatibility Ideographs
-      when 0xFA70..0xFAD9 then 4.1 # CJK Compatibility Ideographs
-      when 0xFE10..0xFE19 then 4.1 # Vertical Forms 中文直排標點
-      when 0xFE20..0xFE23 then 1.1 # Combining Half Marks
-      when 0xFE24..0xFE26 then 5.1 # Combining Half Marks
-      when 0xFE30..0xFE44 then 1.0 # CJK Compatibility Forms 兼容性表格
-      when 0xFE45..0xFE46 then 3.2 # CJK Compatibility Forms 兼容性表格
-      when 0xFE47..0xFE48 then 4.0 # CJK Compatibility Forms 兼容性表格
-      when 0xFE49..0xFE4F then 1.0 # CJK Compatibility Forms 兼容性表格
-      when 0xFE50..0xFE52 then 1.0 # Small Form Variants
-      when 0xFE54..0xFE66 then 1.0 # Small Form Variants
-      when 0xFE68..0xFE6B then 1.0 # Small Form Variants
-      when 0xFF01..0xFF5E then 1.0 # Halfwidth and Fullwidth Forms
-      when 0xFF5F..0xFF60 then 3.2 # Halfwidth and Fullwidth Forms
-      when 0xFF61..0xFF9F then 1.0 # Halfwidth and Fullwidth Forms
-      end
-    end
-  end
-
-  def self.uv00(i)
-    case i
-    when 0..0x01FF
-      case i
-      when 0..0x017E then 1.0
-      when 0x017F    then 1.1
-      when 0x0180..0x01F0 then 1.0
-      when 0x01F1..0x01F5 then 1.1
-      when 0x01F6..0x01F9 then 3.0
-      when 0x01FA..0x0217 then 1.1
-      end
-    when 0x0200..0x02FF
-      case i
-      when 0x0218..0x021F then 3.0
-      when 0x0220         then 3.2
-      when 0x0221         then 4.0
-      when 0x0222..0x0233 then 3.0
-      when 0x0234..0x0236 then 4.0
-      when 0x0237..0x0241 then 4.1
-      when 0x0242..0x024F then 5.0
-      when 0x0250..0x02A8 then 1.0
-      when 0x02A9..0x02AD then 3.0
-      when 0x02AE..0x02AF then 4.0
-      when 0x02B0..0x02DE then 1.0
-      when 0x02DF         then 3.0
-      when 0x02E0..0x02E9 then 1.0
-      when 0x02EA..0x02EE then 3.0
-      when 0x02EF..0x02FF then 4.0
-      end
-    when 0x0300..0x0341 then 1.0
-    when 0x0400..0x04FF
-      case i
-      when 0x0401..0x040C then 1.0
-      when 0x040D         then 3.0
-      when 0x040E..0x044F then 1.0
+  def read_strokes
+    fn = File.join(DATA_DIR, 'Unihan_DictionaryLikeData.txt')
+    @strokes = {}
+    File.foreach(fn) do |line|
+      next if line.start_with? '#'
+      line.chomp!
+      cells = line.split("\t")
+      if cells[1] == 'kTotalStrokes'
+        c = [cells[0].sub(/^U\+(.*)$/, '\1').hex].pack('U')
+        i = cells[2].to_i
+        @strokes[c] = i
       end
     end
   end
 
-  def self.uv02(i)
-    case i
-    when 0x2000..0x20FF
-      case i
-      when 0x2000..0x202E then 1.0
-      when 0x2045..0x2046 then 1.1
-      when 0x2047         then 3.2
-      when 0x2048..0x204F then 3.0
-      end
-    when 0x2100..0x21FF
-      case i
-      when 0x2100..0x2138 then 1.0
-      when 0x2153..0x2182 then 1.0
-      when 0x2190..0x21EA then 1.0
-      end
-    when 0x2200..0x22F1 then 1.0
-    when 0x2400..0x24FF
-      case i
-      when 0x2460..0x24EA then 1.0
-      when 0x24EB..0x24FE then 3.2
-      when 0x24FF         then 4.0
-      end
-    when 0x2500..0x25FF
-      case i
-      when 0x2500..0x2595 then 1.0
-      when 0x2596..0x259F then 3.2
-      when 0x25A0..0x25EE then 1.0
-      when 0x25EF         then 1.1
-      end
-    when 0x2600..0x26FF
-      case i
-      when 0x2600..0x2613 then 1.0
-      when 0x261A..0x266F then 1.0
-      end
-    when 0x2E00..0x2FFF
-      case i
-      when 0x2E80..0x2EF3 then 3.0
-      when 0x2F00..0x2FD5 then 3.0
-      when 0x2FF0..0x2FFB then 3.0
-      end
+  def read_version
+    @vers = []
+    fn = File.join(DATA_DIR, 'unicode-chars-ver.csv')
+    CSV.foreach(fn, headers: true) do |row|
+      @vers << {
+        range: (row['cp1'].hex..row['cp2'].hex),
+        age: row['age'].to_f
+      }
     end
   end
 
-  def self.uv03(i)
-    case i
-    when 0x3000..0x30FF
-      case i
-      when 0x3000..0x3036 then 1.0 # CJK Symbols and Punctuation
-      when 0x3037         then 1.1 # CJK Symbols and Punctuation
-      when 0x3038..0x303A then 3.0 # CJK Symbols and Punctuation
-      when 0x303B..0x303D then 3.2 # CJK Symbols and Punctuation
-      when 0x303E         then 3.0 # CJK Symbols and Punctuation
-      when 0x303F         then 1.0 # CJK Symbols and Punctuation
-      when 0x3041..0x3094 then 1.0 # Hiragana 日文平假名
-      when 0x3095..0x3096 then 3.2 # Hiragana 日文平假名
-      when 0x3099..0x309E then 1.0 # Hiragana 日文平假名
-      when 0x309F         then 3.2 # Hiragana 日文平假名
-      when 0x30A0         then 3.2 # Katakana 日文片假名
-      when 0x30A1..0x30F6 then 1.0 # Katakana 日文片假名
-      when 0x30F7..0x30FA then 1.1 # Katakana 日文片假名
-      when 0x30FB..0x30FE then 1.0 # Katakana 日文片假名
-      when 0x30FF         then 3.2 # Katakana 日文片假名 (Unicode 3.2)
-      end
-    when 0x3100..0x31FF
-      case i
-      when 0x3105..0x312C then 1.0 # Bopomofo 注音符號
-      when 0x312D         then 5.1 # Bopomofo 上下顛倒的 'ㄓ'
-      when 0x3131..0x318E then 1.0 # Hangul Compatibility Jamo 韓文
-      when 0x3190..0x319F then 1.0 # Kanbun 在上方的小漢字
-      when 0x31A0..0x31B7 then 3.0 # Bopomofo Extended 注音擴展
-      when 0x31B8..0x31BA then 6.0 # Bopomofo Extended 注音擴展
-      when 0x31C0..0x31CF then 4.1 # CJK Strokes 筆劃 (基本筆劃, 如撇, 勾, 點...)
-      when 0x31D0..0x31E3 then 5.1 # CJK Strokes 筆劃 (基本筆劃, 如撇, 勾, 點...)
-      when 0x31F0..0x31FF then 3.2 # Katakana Phonetic Extensions 日文片假名語音擴展
-      end
-    when 0x3200..0x32FF
-      case i
-      when 0x3200..0x321C then 1.0 # Enclosed CJK Letters and Months 括號韓文
-      when 0x321D..0x321E then 4.0 # Enclosed CJK Letters and Months 括號韓文
-      when 0x3220..0x3243 then 1.0 # Enclosed CJK Letters and Months 括號一~十及漢字
-      when 0x3244..0x324F then 5.2 # Enclosed CJK Letters and Months 圓圈中有字及10~80
-      when 0x3250         then 4.0 # Enclosed CJK Letters and Months 'PTE' 組成一字
-      when 0x3251..0x325F then 3.2 # Enclosed CJK Letters and Months 圓圈 21~35
-      when 0x3260..0x327B then 1.0 # Enclosed CJK Letters and Months 圓圈韓文
-      when 0x327C..0x327D then 4.0 # Enclosed CJK Letters and Months 圓圈韓文
-      when 0x327E         then 4.1 # Enclosed CJK Letters and Months 圓圈韓文
-      when 0x327F..0x32B0 then 1.0 # Enclosed CJK Letters and Months 圓圈一~十及漢字
-      when 0x32B1..0x32BF then 3.2 # Enclosed CJK Letters and Months 圓圈 36~50
-      when 0x32C0..0x32CB then 1.1 # Enclosed CJK Letters and Months 1月~12月
-      when 0x32CC..0x32CF then 4.0 # Enclosed CJK Letters and Months 多英文組成一個字
-      when 0x32D0..0x32FE then 1.0 # Enclosed CJK Letters and Months 圓圈日文
-      end
-    when 0x3300..0x33FF
-      case i
-      when 0x3300..0x3357 then 1.0 # CJK Compatibility 多個日文組成一字
-      when 0x3358..0x3376 then 1.1 # CJK Compatibility 0点~24点 及多英文組成一字
-      when 0x3377..0x337A then 4.0 # CJK Compatibility 多英文組成一字
-      when 0x337B..0x33DD then 1.0 # CJK Compatibility 多日本漢字及多英文組成一字
-      when 0x33DE..0x33DF then 4.0 # CJK Compatibility 多英文組成一字
-      when 0x33E0..0x33FE then 1.1 # CJK Compatibility 1日~31日
-      when 0x33FF         then 4.0 # CJK Compatibility 'gal' 組成一字
+  def ver_bsearch(code, start, stop)
+    return nil if start > stop
+    middle = (stop - start) / 2 + start
+    h = @vers[middle]
+    if h[:range].include?(code)
+      return h[:age]
+    elsif middle == start
+      return nil if code < h[:range].begin
+      return ver_bsearch(code, middle+1, stop)
+    else
+      if code < h[:range].begin
+        return ver_bsearch(code, start, middle)
+      else
+        return ver_bsearch(code, middle, stop)
       end
     end
   end
 
-  def self.uv2(i)
-    case i
-    when 0x20000..0x2A6D6 then 3.1  # Extension B
-    when 0x2A6D7..0x2A6DD then 13.0 # Extension B
-    when 0x2A700..0x2B734 then 5.2  # extension C
-    when 0x2B740..0x2B81D then 6.0  # extension D
-    when 0x2B820..0x2CEA1 then 8.0  # extension E
-    when 0x2CEB0..0x2EBE0 then 10.0 # extension F
-    when 0x2F800..0x2FA1D then 3.1  # Unicode 3.1: CJK Compatibility Supplement
-    end
-  end
-
-  private_class_method :uv0, :uv00, :uv02, :uv03, :uv2
-
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: unihan2
 version: !ruby/object:Gem::Version
-  version: 0.1.2
+  version: 0.2.0
 platform: ruby
 authors:
 - Ray Chou
@@ -17,6 +17,7 @@ extensions: []
 extra_rdoc_files: []
 files:
 - data/Unihan_DictionaryLikeData.txt
+- data/unicode-chars-ver.csv
 - lib/unihan2.rb
 homepage: https://github.com/RayCHOU/unihan2
 licenses: