Method: ConverterBase#insert_word_separator

Defined in:
lib/converterbase.rb

#insert_word_separator(str) ⇒ Object

単語単位でzwsを挿入する



1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
 
# File 'lib/converterbase.rb', line 1212

def insert_word_separator(str)
  buffer = +""
  ss = StringScanner.new(str)
  before_symbol = false

  if @text_type == "textfile"
    buffer << ss.scan(/(.+\n){2}/)
  end

  while char = ss.getch
    symbol = false
    case char
    when "｜"
      ss.scan(/.+?》/)
    when "［"
      buffer << char
      if ss.scan(/^＃.+?］/)
        buffer << "#{ss.matched}"
      else
        before_symbol = false
      end
      next
    when "<"
      if ss.scan(/.+?>/)
        buffer << "<#{ss.matched}"
        next
      end
      symbol = true
    when /[\d０-９]/
      ss.scan(/[\d０-９]+/)
    when /[ぁ-んゝゞ]/
      ss.scan(/[ぁ-んゝゞー]+/)
    when /[ァ-ヶ]/
      ss.scan(/[ァ-ヶー・]+/)
    when /[Ａ-Ｚａ-ｚA-Za-z]/
      ss.scan(/[Ａ-Ｚａ-ｚA-Za-z ]+/)
    when /[一-龥朗-鶴]/
      ss.scan(/[一-龥朗-鶴]+/)
    when /[〔「『\(（【〈《≪〝]/
      buffer << char
      before_symbol = false
      next
    else
      symbol = true
    end
    if before_symbol && !symbol
      buffer << WORD_SEPARATOR
    end
    buffer << char
    unless symbol
      buffer << ss.matched if ss.matched?
      buffer << WORD_SEPARATOR
    end
    before_symbol = symbol
  end
  buffer
end