Technically there is no such thing as an array of hashes. When we speak of an array of hashes what we really have is an array of references to hashes so the assignment:
@array2 = @array1
will only assign the contents of @array1 (the hash references) to @array2. Here is some code that demonstrates the difference between assignment and completely cloning @array1
print "Here is what happens with assignment\n";
my @array1 = ( {foo=>'foo1', bar=>'bar1' } , {foo=>'foo2', bar=>'bar2'
+ } );
my @array2 = ();
@array2 = @array1;
print "\@array1: @array1\n";
print "\@array1: @array2\n";
$array1[0]{foo} = 'new';
print "\$array1[0]{foo}: $array1[0]{foo}\n";
print "\$array2[0]{foo}: $array2[0]{foo}\n";
print "Here is how to clone \@array1\n";
my @array1 = ( {foo=>'foo1', bar=>'bar1' } , {foo=>'foo2', bar=>'bar2'
+ } );
my @array2 = ();
for (@array1) {
my %new_hash = %$_;
push @array2, \%new_hash;
}
print "\@array1: @array1\n";
print "\@array1: @array2\n";
$array1[0]{foo} = 'new';
print "\$array1[0]{foo}: $array1[0]{foo}\n";
print "\$array2[0]{foo}: $array2[0]{foo}\n";
Run this code to see the differences. In the first case when we change one of the hash elements both @array1[0]{foo} and @array2[0]{foo} are changed as they both point to the same anonymous hash. In the second case we generate a new anon hash and push references to it into @array2. We now have a true independent clone as you can see by the different references in the array and the different result of changing @array1[0]{foo}.
cheers
