Python中使用bidict模块双向字典结构的奇技淫巧 快速入门 模块提供三个类来处理一对一映射类型的一些操作 'bidict', 'inverted', 'namedbidict' >>> import bidict >>> dir(bidict) ['MutableMapping', '_LEGALNAMEPAT', '_LEGALNAMERE', '__builtins__', '__doc__', '__file__', '__name__', '__package__', 'bidict', 'inverted', 'namedbidict', 're', 'wraps'] 1.bidict类: >>> from bidict import bidict >>> D=bidict({'a':'b'}) >>> D['a'] 'b' >>> D[:'b'] 'a' >>> ~D #反转字典 bidict({'b': 'a'}) >>> dict(D) #转为普通字典 {'a': 'b'} >>> D['c']='c' #添加元素,普通字典的方法都可以用 >>> D bidict({'a': 'b', 'c': 'c'}) 2.inverted类,反转字典的键值 >>> seq = [(1, 'one'), (2, 'two'), (3, 'three')] >>> list(inverted(seq)) [('one', 1), ('two', 2), ('three', 3)] 3.namedbidict(mapname, fwdname, invname): >>> CoupleMap = namedbidict('CoupleMap', 'husbands', 'wives') >>> famous = CoupleMap({'bill': 'hillary'}) >>> famous.husbands['bill'] 'hillary' >>> famous.wives['hillary'] 'bill' >>> famous.husbands['barack'] = 'michelle' >>> del famous.wives['hillary'] >>> famous CoupleMap({'barack': 'michelle'}) 更多内容 如果你不喜欢冒号的方式,可以使用namedbidict类给双向字典起2个别名。这样对外会提供正向和逆向的2个子字典。实际上还是以一个双向 字典的形式存在: >>> HTMLEntities = namedbidict('HTMLEntities', 'names', 'codepoints') >>> entities = HTMLEntities({'lt': 60, 'gt': 62, 'amp': 38}) # etc >>> entities.names['lt'] 60 >>> entities.codepoints[38] 'amp' 还可以使用一元的逆运算符"~"获取bidict逆映射字典。 >>> import bidict >>> from bidict import bidict >>> husbands2wives = bidict({'john': 'jackie'}) >>> ~husbands2wives bidict({'jackie': 'john'}) 以下情况注意添加括号,因为~的优先级低于中括号: >>> import bidict >>> from bidict import bidict >>> husbands2wives = bidict({'john': 'jackie'}) >>> ~husbands2wives bidict({'jackie': 'john'}) 以下情况注意添加括号,因为~的优先级低于中括号: >>> (~bi)['one'] 1 bidict不是dict的子类,但它的API的是dict的超集(但没有fromkeys方法,改用了MutableMapping接 口)。 迭代器类inverted会翻转key和value,如: >>> seq = [(1, 'one'), (2, 'two'), (3, 'three')] >>> list(inverted(seq)) [('one', 1), ('two', 2), ('three', 3)] bidict的invert()方法和inverted类似。依赖模块:collections中的MutableMapping,functools中的wraps,re。 bidict可以和字典进行比较 >>> bi == bidict({1:'one'}) >>> bi == dict([(1, 'one')]) True 其他字典通用的方法,bidict也支持: >>> bi.get('one') 1 >>> bi.setdefault('one', 2) 1 >>> bi.setdefault('two', 2) 2 >>> len(bi) # calls __len__ 2 >>> bi.pop('one') 1 >>> bi.popitem() ('two', 2) >>> bi.inv.setdefault(3, 'three') 'three' >>> bi bidict({'three': 3}) >>> [key for key in bi] # calls __iter__, returns keys like dict ['three'] >>> 'three' in bi # calls __contains__ True >>> list(bi.keys()) ['three'] >>> list(bi.values()) [3] >>> bi.update([('four', 4)]) >>> bi.update({'five': 5}, six=6, seven=7) >>> sorted(bi.items(), key=lambda x: x[1]) [('three', 3), ('four', 4), ('five', 5), ('six', 6), ('seven', 7)]