最近任务上有去重,随手翻了下发现有不少好方法啊~
记录一下,转自http://yxmhero1989.blog.163.com/blog/static/112157956201381443244790/:
一.{}.fromkeys(list).keys()
二.setlist2 ={}.fromkeys(list1).keys()
三.itertools.groubylist2 = list(set(list1))
ids = [1,4,3,3,4,2,3,4,5,6,1] ids.sort() it = itertools.groupby(ids) for k, g in it: print k
四,笨方法
ids = [1,2,3,3,4,2,3,4,5,6,1] news_ids = [] for id in ids: if id not in news_ids: news_ids.append(id) print news_ids
这四种都有个特点,去重后元素排序变了,效率 据说第一种比第二种快一点
五.索引再次排序 这种可以去重并且保持元素顺序
#要结果是[1, 4, 3, 2, 5, 6] 不要[1, 2, 3, 4, 5, 6] ids = [1,4,3,3,4,2,3,4,5,6,1] news_ids = list(set(ids)) news_ids.sort(key=ids.index) print news_ids #[1, 4, 3, 2, 5, 6]
六:Reduce
ids = [1,4,3,3,4,2,3,4,5,6,1] func = lambda x,y:x if y in x else x + [y] print reduce(func, [[], ] + ids)#[1, 4, 3, 2, 5, 6]
参考:
http://the5fire.com/python-remove-duplicates-in-list.html
http://xcw31.diandian.com/post/2012-11-28/40042801718
http://www.benben.cc/blog/?p=386
http://blog.csdn.net/zhengnz/article/details/6265282