chinese string

gistfile1.rb

# regexps to check if a string is pure chinese
class String
  # 20k chars
  CHINESE_UCS2 = /^(?:
    [\x4e-\x9e][\x00-\xff]
    |\x9f[\x00-\xa5]
    )+$/xn

  # 20k chars
  CHINESE_UTF8 = /^(?:
    \xe4[\xb8-\xbf][\x80-\xbf]
    |[\xe5-\xe8][\x80-\xbf][\x80-\xbf]
    |\xe9[\x80-\xbd][\x80-\xbf]
    |\xe9\xbe[\x80-\xa5]
    )+$/xn

  # 27k chars
  CHINESE_GB_2000 = /^(?:  
    [\xB0-\xF7][\xA1-\xFE]  
    |[\x81-\xA0][\x40-\xFE]  
    |[\xAA-\xFE][\x40-\xA0]  
    |[\x81-\x82][\x30-\x39][\x81-\xFE][\x30-\x39]  
    )+$/xn  

  # 70k chars (including minorities chars)
  CHINESE_GB_2005 = /^(?:  
    [\xB0-\xF7][\xA1-\xFE]  
    |[\x81-\xA0][\x40-\xFE]  
    |[\xAA-\xFE][\x40-\xA0]  
    |[\x81-\x82][\x30-\x39][\x81-\xFE][\x30-\x39]  
    |[\x95-\x98][\x30-\x39][\x81-\xFE][\x30-\x39]  
    )+$/xn
end