John Henry napisal(a): > I believe the standard dictionary should be amened to allow the use of > case insensitive keys - as an option. I found some work done by others > to do that at: > > http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/283455 > > but the problem with that approach is that they lowercase the keys > immediately when you create the dictionary and so the true identity of > the key is lost. > > Of course, I can subclass it and save a copy of the "real" key but > that's kind of messcy. > > In other words: > > If I have: > > pets=caselessDict() > > pets["Cat"] = 3 > pets["Dog"] = 2 > > I would like to see: > > pets["cat"] prints 3 > pets["DOG"] prints 2 > > but > > print pets.keys() > > should print: > > "Cat", "Dog" > > not: > > "cat", "dog"
How about: >>> class Dictstr(str): ... def __hash__(self): ... return str.__hash__(self.lower()) ... def __eq__(self,other): ... return self.lower()==other.lower() ... >>> class dictt(dict): ... def __setitem__(self, key, value): ... if isinstance(key, str): ... dict.__setitem__(self, Dictstr(key), value) ... else: ... dict.__setitem__(self, key, value) ... def __getitem__(self, key): ... if isinstance(key, str): ... return dict.__getitem__(self, Dictstr(key)) ... else: ... return dict.__getitem__(self, key) ... >>> d=dictt() >>> d[1]=1 >>> d['Cat']='Cat' >>> d['Dog']='Dog' >>> d['dOg'] 'Dog' >>> d.keys() [1, 'Dog', 'Cat'] Note that you would need to redefine also __init__, __contains__ and other methods. -- http://mail.python.org/mailman/listinfo/python-list
