Class: GIGO::CharDet::Big5DistributionAnalysis

Inherits:
CharDistributionAnalysis show all
Defined in:
lib/gigo/rchardet/chardistribution.rb

Instance Method Summary collapse

Methods inherited from CharDistributionAnalysis

#feed, #get_confidence, #got_enough_data, #reset

Constructor Details

#initialize ⇒ Big5DistributionAnalysis

Returns a new instance of Big5DistributionAnalysis.



164
165
166
167
168
169
 
# File 'lib/gigo/rchardet/chardistribution.rb', line 164

def initialize
  super
  @_mCharToFreqOrder = Big5CharToFreqOrder
  @_mTableSize = BIG5_TABLE_SIZE
  @_mTypicalDistributionRatio = BIG5_TYPICAL_DISTRIBUTION_RATIO
end

Instance Method Details

#get_order(aStr) ⇒ Object 



171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
 
# File 'lib/gigo/rchardet/chardistribution.rb', line 171

def get_order(aStr)
  # for big5 encoding, we are interested 
  #   first  byte range: 0xa4 -- 0xfe
  #   second byte range: 0x40 -- 0x7e , 0xa1 -- 0xfe
  # no validation needed here. State machine has done that
  if aStr[0..0] >= "\xA4"
	if aStr[1..1] >= "\xA1"
	  return 157 * (aStr[0].ord - 0xA4) + aStr[1].ord - 0xA1 + 63
	else
	  return 157 * (aStr[0].ord - 0xA4) + aStr[1].ord - 0x40
	end
  else
	return -1
  end
end