topN using a heap

[Andrei Alexandrescu via Digitalmars-d]

On 01/19/2016 08:08 PM, Ivan Kazmenko wrote:
> On Tuesday, 19 January 2016 at 13:52:08 UTC, Andrei Alexandrescu wrote:
>> On 01/18/2016 09:21 PM, Ivan Kazmenko wrote:
>> Do you think sort and topN would be attackable if they used a
>> per-process-seeded RNG as per Xinok's idea? -- Andrei
>
> Yes, but with a little interactivity (not generating the input in
> advance) and more labor (finding the state of RNG).
[snip]

Thanks! -- Andrei