matching array element to hash key

[ccc31807]

This is probably a real stupid question, but it's late and I'm running
behind, and quite frankly have run out of mental energy.

I have an array with about 5,000 elements that looks like this:
0002793 0005095 0093350 0134143 0145740 0146854 0150248 0151827
0156424 0161321 0186959 0198460 0218115

I have a four level hash that looks like this:
$hash{$s}{$l}{$h}{$id} = $terms
This has contains about 8,000 discrete elements.

The $id key in the hash either matches an array element or it does
not. I need to write the hash keys and the values to one file if the
$id matches the array element. I need to write the unmatching array
elements to a second file, and the unmatching hash elements to a third
file.

Here's what I'm trying to do:

foreach my $sid (@array)
{
if ($sid == ???)
{
print FIRST qq($s $l $h $id $hash{$s}{$l}{$h}{$id}\n);
delete $hash{$s}{$l}{$h}{$id};
shift @array;
}
}
foreach (@array) { print SECOND qq{$_\n); }

Then, print the remaining hash elements to THIRD

I'm sorry, but for the life of me I can't figure out how to access the
last level of the hash, at least without iterating through the hash.

CC.

 

[sln]

This is probably a real stupid question, but it's late and I'm running
behind, and quite frankly have run out of mental energy.

I have an array with about 5,000 elements that looks like this:
0002793 0005095 0093350 0134143 0145740 0146854 0150248 0151827
0156424 0161321 0186959 0198460 0218115

I have a four level hash that looks like this:
$hash{$s}{$l}{$h}{$id} = $terms
This has contains about 8,000 discrete elements.

The $id key in the hash either matches an array element or it does
not. I need to write the hash keys and the values to one file if the
$id matches the array element. I need to write the unmatching array
elements to a second file, and the unmatching hash elements to a third
file.

Here's what I'm trying to do:

foreach my $sid (@array)
{
if ($sid == ???)
{
print FIRST qq($s $l $h $id $hash{$s}{$l}{$h}{$id}\n);
delete $hash{$s}{$l}{$h}{$id};
shift @array;
}
}
foreach (@array) { print SECOND qq{$_\n); }

Then, print the remaining hash elements to THIRD

I'm sorry, but for the life of me I can't figure out how to access the
last level of the hash, at least without iterating through the hash.

CC.

There is no shame in itterating.

-sln
=====================
use strict;
use warnings;

my %hash = (
s1 => {
l1 => {
h1 => {
id01 => 9487,
id02 => 9488,
bad => { id2a => 666 }
},
h2 => {
id03 => 9489,
id04 => 9490
}
},
l2 => {
h3 => {
id05 => 9491,
id06 => 9492
},
h4 => {
id07 => 9493,
id08 => 9494
}
}
},
s2 => {
l3 => {
h5 => {
id09 => 9495,
id10 => 9496,
bad => { id10a => 666 }
},
h6 => {
id11 => 9497,
id12 => 9498
}
},
bad => 3982
}
);


my $level = 0;
my @keystack = ();
my @found = ();

_GVM(\%hash);

print "$_\n" for sort @found;

exit 0;


sub _GVM {
my $var = shift;
return if ( ref ($var) ne 'HASH');

for (keys %{$var})
{
push @keystack, $_;
if (ref ($var->{$_}) eq 'HASH') {
_GVM ($var->{$_}) if ++$level < 4;
--$level;
}
elsif (ref (\$var->{$_}) eq 'SCALAR' && $level == 3) {
push @found, "@keystack $var->{$_}";
#print "@keystack $var->{$_}\n";
#print THIRD "@keystack $var->{$_}\n";
}
pop @keystack;
}
return;
}


__END__

Output:

s1 l1 h1 id01 9487
s1 l1 h1 id02 9488
s1 l1 h2 id03 9489
s1 l1 h2 id04 9490
s1 l2 h3 id05 9491
s1 l2 h3 id06 9492
s1 l2 h4 id07 9493
s1 l2 h4 id08 9494
s2 l3 h5 id09 9495
s2 l3 h5 id10 9496
s2 l3 h6 id11 9497
s2 l3 h6 id12 9498

 

[ccc31807]

You can't. Iterate over the whole structure and build a new hash keyed
by $id only (or, obviously, modify whatever code builds %hash to do that
in the first place).

Haste makes waste, as they say. I was trying to do this on the quick
and dirty. Yes, you are correct, I need to put the IDs from the hash
into a separate hash. Not only is it not a big deal, it will improve
my output since I will have access to some details that I was losing
trying to do it the way I posted.

Presumably all your $ids are unique? If they aren't, you will need to
decide what to do with the duplicates, but you needed to do that anyway.

Yes, all the IDs are unique as to the person, but (obviously) not to
the data sets. This is trouble shooting problem with data from a
database. The two files are data files that should return exactly the
same data set, and about 85% of the data is the same. However, the
data set I'm putting in the hash has about 20% more records that the
data set I'm putting in the array, and the array has about 30 or 40
records that are unique to it, and we are trying to discover what the
difference is that would cause certain records to be returned by one
query but not the other.

Thanks, CC.