Re: random index in array

An option is to use splice. It's probably not the most efficient (you can always benchmark to make sure), but otherwise you can't get away with a single pass (at least not using rand, there probably is a perl module to only spit out unique random values, but its destined to be less efficient than just keeping track yourself).

______crazyinsomniac_____________________________
Of all the things I've lost, I miss my mind the most.
perl -e "$q=$_;map({chr unpack qq;H*;,$_}split(q;;,q*H*));print;$q/$q;"

Comment on Re: random index in array - no duplicates

Replies are listed 'Best First'.
Re: random index in array - no duplicates by Abigail-II (Bishop) on Jun 07, 2002 at 11:28 UTC
A `splice` isn't efficient. It's essentially quadratic in the size of the array, while a shuffle is linear. Or to be more precise. If you have an array of size N and you want to draw M elements from it (without duplicates), `splice` will take time proportional to *N M, while the suffle will take time proportional to N + M** (or just M if you get a reference to the array). Abigail	[reply] [d/l] [select]