Re: Answer: How do I find unique Array in Array of Array?

Since this stores the Data::Dumper output for each array in %seen as a key, wouldn't it be vastly inefficient in terms of memory use? It might be better to md5 your Dumper output before using it as a key:

use strict;
use warnings;
use Data::Dumper;
use Digest::MD5 qw/md5/;

my @a = (
    ['a','b','c'],
    ['a','b','c'],
    ['a','b','d'],
    ['a','b','d'],
);

my %seen;
my @b = grep { !$seen{md5 Dumper($_)}++ } @a;

print Dumper(\@b);
[download]

Comment on Re: Answer: How do I find unique Array in Array of Array? Download Code

Replies are listed 'Best First'.
Re^2: Answer: How do I find unique Array in Array of Array? by jdporter (Paladin) on May 12, 2006 at 19:32 UTC
Good point; but as with anything, there's a time/space tradeoff, and it's the engineer's call. I would say that for something like your sample data, the time it takes to calculate the MD5 would not be worth it, especially given that the memory savings would be neglible. In really extreme cases, you'd probably want a function that could hash a complex data structure directly, rather than a stringification of it. We're building the house of the future together.	[reply]

Replies are listed 'Best First'.

Re^2: Answer: How do I find unique Array in Array of Array?
by jdporter (Paladin) on May 12, 2006 at 19:32 UTC

Good point; but as with anything, there's a time/space tradeoff, and it's the engineer's call.

I would say that for something like your sample data, the time it takes to calculate the MD5 would not be worth it, especially given that the memory savings would be neglible.

In really extreme cases, you'd probably want a function that could hash a complex data structure directly, rather than a stringification of it.

We're building the house of the future together.

[reply]