autotagger.py 文件源码

def unicode_width(string):
    def char_width(char):
        # ('F', 'W', 'A', 'H', 'N', 'Na')
        # Ref: http://www.unicode.org/reports/tr11/tr11-14.html
        w2 = ('F', 'W', 'A')
        w1 = ('H', 'Na')
        w = unicodedata.east_asian_width(char)
        if w in w2:
            return 2
        elif w in w1:
            return 1
        else:
            return 0
    length = sum([char_width(c) for c in string])
    return length