17 replies to this topic

[Jojan]

Hi. Decided to learn some Perl today and bumped in to something odd.

This code

my %lot = ("a" => 1, "b" => 2, "c" => 3, "d" => 4);

for my $i (%lot) {
    print $i . "\n";
}

results in

c
3
a
1
b
2
d
4

on my home machine (Ubuntu 10.04) with Perl version 5.10. But on my schools Solaris machine with Perl version 5.004 I get more of the result I expect, which is

a
1
b
2
c
3
d
4

So is there a good explanation for this, or is it just some strange feature?

I moved c => 3 to other places in lot (for example my %lot = ("a" => 1, "b" => 2, "d" => 4, "c" => 3); ) but get the same result.

[Alexander]

In PERL a hash defines no ordering properties. The order in which things come out will be unpredictable.

[Jojan]

Thank you for the reply!

Is there a "deeper thought" about this feature? It does not seem to be random, so how have this been implemented? It differs, as I wrote, between versions.

[Alexander]

I did some digging, perldoc -f keys returns:

The keys of a hash are returned in an apparently random order. The actual random order is subject to change in future versions of Perl, but it is guaranteed to be the same order as either the values or each function produces (given that the hash has not been modified). Since Perl 5.8.1 the ordering is different even between different runs of Perl for security reasons (see Algorithmic Complexity Attacks in perlsec).

It appears since your 5.004 the internal storing retrieval method has changed a little, but remember to never rely on it (as I hinted before), use sort on the hash:

for my $i (sort keys %hash) {
    print "$i -> $hash{$i}\n";
}

[Jojan]

Interesting.

Since I am new to Perl I just wanted to see what happened.

[bythos]

I've always through that the hash behavior was a pain if you wanted to iterate through a hash. I wonder how perl6 will handle it?

[Alexander]

I've always through that the hash behavior was a pain if you wanted to iterate through a hash. I wonder how perl6 will handle it?

Welcome to the forums!

There isn't going to be much change in that aspect, although I found a nice compiled list of differences in PERL 6 that may be of interest:
Perl6 :: Perl5 :: Differences

[bythos]

cheers thanks for the welcome. But i do have to ask the question why is the perl forum under the web development? i don't think perl is used much for web development anymore. more for system administration these days.

[Alexander]

Hmmm you know what I wonder the same thing too, although now I look at it there aren't any sections it'd "obviously" fit in, I'd vote for the Linux section but that's sort of tight already. Webdev more up top so it gets enough traffic, can't complain there.

[Jojan]

I assume most people use Ctrl + f (or equivalent) to search for the language they want, and don't really care where it is placed.

[lor]

Um... Pretty sure any hash set has unreliable ordering, not just in Perl.

[bythos]

Yeah its depends on the how a hash is implemented in the languages library in most cases.