Imprecise running time for topN?

[Magnus Lie Hetland]

On 2011-02-01 16:29:56 +0100, Andrei Alexandrescu said:

> On 2/1/11 8:12 AM, Magnus Lie Hetland wrote:
[snip]
>> I'm not objecting to the use of algorithm -- it's a good choice in
>> practice -- but the docs should probably specify that the linear
>> guarantee does not hold in the worst case?
> 
> You're right (and randomization should be there, too). Could you please 
> add a bugzilla entry so we don't forget about this? 
> http://d.puremagic.com/issues. Thanks!

Will do. :)

> Andrei


-- 
Magnus Lie Hetland
http://hetland.org