Class EUCJPDistributionAnalysis

thirdparty/chardet/chardistribution.py:217–233

class EUCJPDistributionAnalysis(CharDistributionAnalysis):
    def __init__(self):
        super(EUCJPDistributionAnalysis, self).__init__()
        self._char_to_freq_order = JIS_CHAR_TO_FREQ_ORDER
        self._table_size = JIS_TABLE_SIZE
        self.typical_distribution_ratio = JIS_TYPICAL_DISTRIBUTION_RATIO

    def get_order(self, byte_str):
        # for euc-JP encoding, we are interested
        #   first  byte range: 0xa0 -- 0xfe
        #   second byte range: 0xa1 -- 0xfe
        # no validation needed here. State machine has done that
        char = byte_str[0]
        if char >= 0xA0:
            return 94 * (char - 0xA1) + byte_str[1] - 0xa1
        else:
            return -1