topN using a heap

Ilya via Digitalmars-d digitalmars-d at puremagic.com
Mon Jan 18 15:18:03 PST 2016


On Monday, 18 January 2016 at 20:45:56 UTC, Ivan Kazmenko wrote:
> On Monday, 18 January 2016 at 12:00:10 UTC, Ivan Kazmenko wrote:
>> Here goes the test which shows quadratic behavior for the new 
>> version:
>> http://dpaste.dzfl.pl/e4b3bc26c3cf
>> (dpaste kills the slow code before it completes the task)
>
> Correction: this is the result of removing a uniform call in 
> pull 3921.  Since we are supposed to discuss the improvements 
> related to pull 3934 here, I created a separate entry for the 
> issue: https://issues.dlang.org/show_bug.cgi?id=15583.

A RNGs don't improve worst case. It only changes an permutation 
for worst case. --Ilya


More information about the Digitalmars-d mailing list