作者ikari512 ( //StrategyBass)
看板Google
标题Re: [问题] 0rz.net??
时间Mon Jul 24 09:36:51 2006
1F：推 ikari512:现在的问题是如果九亿个都被用光是否会override的问题^^" 07/24 01:22
2F：→ ikari512:不是hash的问题喔.. 07/24 01:23
3F：推 cpcmce:你的第一行没办法推到第二行的结论吧 07/24 04:19
4F：推 proLIONS:The same string of course will output the same hash 07/24 07:11
5F：→ proLIONS:value. The problem is will two different string map 07/24 07:12
6F：→ proLIONS:to ONE hash value, and it's so called hash collision. 07/24 07:13
我想请教一下
如果说今天是参照别的资讯(例如DB中最久没用缩址的日期)
而造成的Hash collision还算是Hash collision吗？
wikipedia的定义是：
In computer science, a hash collision is a situation that occurs when
two distinct inputs into a hash function produce identical outputs.
就是说 明明output不同 但是为了别的理由把他map到之前用过的output
不知道这样算不算？
抱歉....借用google版讨论一下 ><
--
※ 发信站: 批踢踢实业坊(ptt.cc)
◆ From: 59.112.49.249
7F：推 xanthous2:既然讨论hash collision 先搞清楚"hash function"是啥 07/24 11:44
8F：→ xanthous2:如果额外的mapping是对hash output做的 那不是hash问题 07/24 11:49
9F：推 proLIONS:Your thinking is quiet interesting. This extra 07/25 02:12
10F：→ proLIONS:processing may somehow relieve the hash collision 07/25 02:13
11F：→ proLIONS:problem. However, just as what 1st floor says, 07/25 02:13
12F：→ proLIONS:it beyonds the operation of a basic hash function. 07/25 02:13
13F：→ proLIONS:and may take extra computational cost due to 07/25 02:14
14F：→ proLIONS:breaking the regularity of hash functions. 07/25 02:15
15F：→ ikari512:感谢 所以说对hash collision做处理不属於hash function? 07/25 17:41