class CharDet::CharDistributionAnalysis

def feed(aStr, aCharLen)
  # # """feed a character with known length"""
  if aCharLen == 2
    # we only care about 2-bytes character in our distribution analysis
    order = get_order(aStr)
  else
    order = -1
  end
  if order >= 0
    @totalChars += 1
    # order is valid
    if order < @tableSize
      if 512 > @charToFreqOrder[order]
        @freqChars += 1
      end
    end
  end
end

def get_confidence
  # """return confidence based on existing data"""
  # if we didn't receive any character in our consideration range, return negative answer
  if @totalChars <= 0
    return SURE_NO
  end
  if @totalChars != @freqChars
    r = @freqChars / ((@totalChars - @freqChars) * @typicalDistributionRatio)
    if r < SURE_YES
      return r
    end
  end
  # normalize confidence (we don't want to be 100% sure)
  return SURE_YES
end

def get_order(aStr)
  # We do not handle characters based on the original encoding string, but 
  # convert this encoding string to a number, here called order.
  # This allows multiple encodings of a language to share one frequency table.
  return -1
end

def got_enough_data
  # It is not necessary to receive all data to draw conclusion. For charset detection,
  # certain amount of data is enough
  return @totalChars > ENOUGH_DATA_THRESHOLD
end

def initialize
  @charToFreqOrder = nil # Mapping table to get frequency order from char order (get from GetOrder())
  @tableSize = nil # Size of above table
  @typicalDistributionRatio = nil # This is a constant value which varies from language to language, used in calculating confidence.  See http://www.mozilla.org/projects/intl/UniversalCharsetDetection.html for further detail.
  reset()
end

def reset
  # # """reset analyser, clear any state"""
  @done = false # If this flag is set to constants.True, detection is done and conclusion has been made
  @totalChars = 0 # Total characters encountered
  @freqChars = 0 # The number of characters whose frequency order is less than 512
end