Perl Hashes (associative arrays) are most like arrays but there are some significant differences, too. An array has a number of elements (scalars) and to access one particular element we use the index (or subscript) of that element, where the first element has the index 0, the next one 1, and so on. A hash contains a number of scalars, too. But in this case an element of the structure represents a pair of two parts – a key and a value. Whenever we refer to an element of a Perl hash structure, we mean a pair which links a value to a key. We have access at the elements of a hash variable by a scalar key and not by an index (like in arrays).

In Perl language, the hash structure is based upon a Perl hash table that associates keys with values. In order o retrieve an element of a hash variable, we must supply a key and find the corresponding value. The keys are unique but for different keys we can have identical values.

Note: I tested all the examples below on a Windows machine, Perl v5.8.8. If you want to run them on Linux, don’t forget to insert at the beginning of the script the following line:

#!/usr/bin/perl

back to top

%ages = ("John", 43, "Paul", 25, "Marie", 22);
%members = (John => "father", Paul => "son", 
                        Marie => "daughter");

back to top

%destHash = %sourceHash;  # copy a hash
$destHasRef = $sourceHashRef;  # copy a hash ref

This is a shallow method to copy Perl hashes, and works well when we speak about one–dimensional hashes, with no references to other structures.

back to top

%ages = ("John", 43, "Paul", 25, "Marie", 22);
%members = (John => "father", Paul => "son", 
                      Marie => "daughter");
$agesRef = \%ages;
print $$agesRef{Marie}; #prints 22 

In the above example, we used two ways to create/initialize a hash variable. In the first line we initialized the %ages hash variable with a list composed by pairing of keys and values, separated by commas. But Perl has a special "comma-arrow" (=>) operator that we used to initialize the %members hash variable in the second line of code. The left side of the comma-arrow operator is expected to be a simple string and therefore it is not necessary to be quoted.

The third line shows you how to define a hash reference in Perl. You can dereference Perl hashes using the notation %$hashRef. If you want to get a certain hash value, you can dereference it by using the notation $$hashRef{key} – the fourth line of the above snippet code shows you how to do that.

I’ll show you below the Perl hash table for both the %ages and %members hash variables.

%ages

%members

%hash = ();

back to top

# define a hash
%cds = ("Dr. Hook"         => "Sylvias Mother", 
            "Rod Stewart"    => "Maggie May", 
            "Andrea Bocelli"  => "Romanza",
            "Kenny Rogers"  => "For the good times",
            "Bee Gees"        => "One night only",
            "Bonnie Tyler"    => "Hide your heart"); 

# we used the @ prefix because the return value
# represents a list
# we need to supply the keys for which we wanted 
# to extract the values 
@slice = @cds{"Dr. Hook","Kenny Rogers","Bee Gees"}; 
# we used the join function to print each string
# on a new line 
print join("\n", @slice);

Running the code it will produce the output:

Sylvias Mother
For the good times
One night only

back to top

ref EXPR

If EXPR is not specified, $_ will be used. This function returns an empty string if the EXPR/$_ is not a reference, and a string (the type of reference) otherwise. Look at the code below to see how it works:

# define some simple variables
# we can use the same name – colors for different things
$colors = "darkblue";
@colors = qw(blue, red, yellow, brown);
%colors = (c1 => "blue", c2 => "green", c3 => "yellow");

# define a hash with references
%moreColors = (1 => "white", 2 => \$colors, 
                      3 => \@colors, 4 => \%colors);

# loop through %moreColors hash to find out the references
foreach $key (keys %moreColors)
{
  if (ref $moreColors{$key} eq "SCALAR") {
    print "$key: SCALAR\n";
  } elsif (ref $moreColors{$key} eq "ARRAY") {
    print "$key: ARRAY\n";
  } elsif (ref $moreColors{$key} eq "HASH") {
    print "$key: HASH\n";
  } elsif (ref $moreColors{$key} eq "") {
    print "$key: NO REFERENCE\n";
  }
}

And here is what you get as output if you run the code:

4: HASH
1: NO REFERENCE
3: ARRAY
2: SCALAR

Do you see the order (4 1 3 2) in which the keys were printed? Perl doesn’t care about the sequential order (1 2 3 4) in which I initialized %moreColors hash! Remember, if you want to print Perl hashes in some order, you must sort the hashes yourself. (see "Sort or order a hash").

back to top

%time = (hour => 12, min =>25, sec => 32, msec =>75);
$size = scalar keys %time;   # explicit scalar context - or
$size = keys %time;   # implicit scalar context - or

$timeRef = \%time;    # using a reference
$size = keys %$timeRef;   

print "Hash size: $size\n";  # prints Hash size: 4

back to top

$hash{"key"} = "value";

If hash doesn’t exist, it will be created automatically by Perl. Please have a look at the following code to see some examples:

%ages = ("John", 43, "Paul", 25, "Marie", 22);
$ages{"Anne"} = 22;
$ages{"Paul"} = 24;
$ages{"Marie"} = 24 unless exists $ages{"Marie"} ;
$ages{"George"} = 51 unless exists $ages{"George"} ;
print %ages, "\n";

Anne22George51John43Paul24Marie22

Please note that the hash is not ordered and you couldn’t expect at any order when you display its elements. Because Perl hashes are not ordered, I preferred to use the insert/add verb instead of append (the end of a hash has not particular meaning).

back to top

%phoneNumbers = ("Anne", "432566", "Paul", "231275", "Marie", 
                 "299302");
while (($key, $value) = each %phoneNumbers)
{
  print "$key has the phone number $phoneNumbers{$key}\n";
}

The code shown above will extract the keys and values from the %phoneNumbers hash and it will display as output:

Anne has the phone number 432566
Paul has the phone number 231275
Marie has the phone number 299302

(see Print a hash for another way to loop through Perl hashes)

back to top

# I’ll create a new hash - %notebookPrices 
# by assigning some elements to it:

$notebookPrices{"Toshiba"}=650;
$notebookPrices{"HP"}=550;
$notebookPrices{"Acer"}=750;

# the next loop prints the keys/values of the hash
foreach $key (keys %notebookPrices)
{
  print "Key: $key, Value: \$$notebookPrices{$key}\n";
}

After running this code, you’ll get the output:

Key: Acer, Value: $750
Key: Toshiba, Value: $650
Key: HP, Value: $550

Note that Perl does not have a default order for the elements of a hash.

back to top

delete $hash{$key};
delete $hashRef->{$key};   # if you have a hash reference
delete $$hashRef{$key};     # if you have a hash reference

See the examples below about how to implement it:

%phoneNumbers = ("Anne", "408566", "Paul", "233375", "Marie", 
                 "217302");
delete $phoneNumbers {"Paul"};
delete $phoneNumbers {"John"};
$phoneNumbersRef = \%phoneNumbers;
delete $$phoneNumbersRef{"Anne"};

#print the hash (keys, values)
foreach $key (keys % phoneNumbers)
{
  print "Key: $key, Value: $phoneNumbers{$key}\n";
}

The second line of this snippet code will have as effect the deletion of the "Paul" key from the %phoneNumbers hash. In the third line of the code we try to delete an inexistent key from the hash, the script will not return any error and the %phoneNumbers hash will remain unchanged. The last line of the above code shows you how to delete a key when you have a hash reference (we used $$hashRef{$key} notation to dereference a certain hash value, but you can use $hashRef->{$key}, too).

back to top

To sort a hash by key, you can use the following code lines:

# define a hash
%time = (hour => 12, min => 25, sec => 32, msec =>75);

# foreach loop
foreach $key (sort keys %time)
{
  print "$key: $time{$key}\n";
}

To sort a hash by value, in a descending order, you can use the following code:

# define a hash
%v = (v1 => 75, v2 => 251, v3 => 3, v4 => 12);

# sort by value and put the keys in an array 
@keys = sort {$v{$b} <=> $v{$a}} keys %v;

# loop through array to print the hash pairs ordered
foreach $key (@keys)
{
  print "$key: $v{$key}\n";
}

If you will run the code, it will produce the following output:

v1: 251
v3: 75
v2: 12
v4: 3

In the example above, our hash has numerical values so we used the numerical operator "<=>" for the comparisons. If you want to sort the values lexically, you need to use the stringwise operator "cmp" instead of "<=>".

back to top

# define a hash
%cds = ("Dr. Hook"        => "Sylvias Mother", 
            "Andrea Bocelli" => "Romanza",
            "Bonnie Tyler"   => "Hide your heart"); 
%cdsReversed = reverse %cds;

# we’ll use Data::Dumper module to see 
# what it is in the hash
use Data::Dumper;
print Dumper(%cdsReversed);

after running this code, we’ll get as output:

$VAR1 = 'Sylvias Mother';
$VAR2 = 'Dr. Hook';
$VAR3 = 'Hide your heart';
$VAR4 = 'Bonnie Tyler';
$VAR5 = 'Romanza';
$VAR6 = 'Andrea Bocelli';

back to top

If you want to download the Perl hashes script with all the above examples included, please click here: Script download