Sorting Hash in Perl

Prerequisite: Perl | Hashes Set of key/value pair is called a Hash. Each key in a hash structure are unique and of type strings. The values associated with these keys are scalar. These values can either be a number, string or a reference. A Hash is declared using my keyword. Let us consider an example to understand the concept of sorting the hash. 

Example: Consider a hash with the names of the students in a class and with their average score. Here, the name of the students are the keys and their average scores are the values. Print this hash using foreach loop on the student names returned by the keys function. 

Perl

# Perl program to demonstrate
# the concept of hash
use strict;
use warnings;
use 5.010;

# Creating a hash of studentnames
# and their average score
my %studentnames = (
Martha => 14,
Vivek => 27,
Earl => 31,
Marty => 16.5,
Jason => 25.2,
Socrates => 29.5,
Uri => 19.6,
Nitin => 30,
Plato => 39,
);

# displaying the keys and values of Hash
# using the foreach loop and keys function
# output may be different each
# time when you run the code
foreach my $name (keys %studentnames) {

    # printing the keys and values of hash
    printf "%-8s %s\n", $name, $studentnames{$name};
}

Output:

Marty    16.5
Jason    25.2
Plato    39
Vivek    27
Socrates 29.5
Martha   14
Earl     31
Uri      19.6
Nitin    30

Time Complexity: O(n)
Auxiliary Space: O(n)

Note: The output may contain any random order depends on the system and the version of Perl. Hashes can be sorted in many ways as follows:

• Sorting the Hash according to the ASCII values of its keys: Generally, sorting is based on ASCII table. It means sorting will keep all the upper-case letters in front of all the lower-case letters. This is the default behavior of the sorting. The above example code is sorted according to the ASCII values of its keys.
• Sorting the Hash according to the alphabetical order of its keys: Here, the keys are sorted alphabetically. 

Example:

Perl

# Perl program to demonstrate
# sorting of the hash according
# alphabetical order of its keys
use strict;
use warnings;
use 5.010;

# Creating a hash of studentnames
# and their average score
my %studentnames = (
Martha => 14,
Vivek => 27,
Earl => 31,
Marty => 16.5,
Jason => 25.2,
Socrates => 29.5,
Uri => 19.6,
Nitin => 30,
Plato => 39,
);

# sorting the hash according
# alphabetical order of its keys
foreach my $name (sort {lc $a cmp lc $b} keys %studentnames)
{
    printf "%-8s %s\n", $name, $studentnames{$name};
}

Output:

Earl     31
Jason    25.2
Martha   14
Marty    16.5
Nitin    30
Plato    39
Socrates 29.5
Uri      19.6
Vivek    27

Time Complexity: O(n logn)
Auxiliary Space: O(n)

Sorting according to the values of Hash: You can also sort the hash according to the values of hash as follows: 
Example 1: Sorting the hash values according to the ASCII table. 

Perl

# Perl program to demonstrate
# sorting of the hash according
# to the values of Hash
use strict;
use warnings;
use 5.010;

# Creating a hash of studentnames
# and their average score
my %studentnames = (
Martha => 14,
Vivek => 27,
Earl => 31,
Marty => 16.5,
Jason => 25.2,
Socrates => 29.5,
Uri => 19.6,
Nitin => 30,
Plato => 39,
);

# sorting of the hash according
# to ASCII code of values of Hash
foreach my $name (sort values %studentnames)
{
    say $name;
}

• Output:

14
16.5
19.6
25.2
27
29.5
30
31
39

Time Complexity: O(n logn)
Auxiliary Space: O(n)

Example 2: Sorting according to the numerical value of Values of hash as follows: 

Perl

# Perl program to demonstrate
# sorting of the hash according
# to the values of Hash
use strict;
use warnings;
use 5.010;

# Creating a hash of studentnames
# and their average score
my %studentnames = (
Martha => 14,
Vivek => 27,
Earl => 31,
Marty => 16.5,
Jason => 25.2,
Socrates => 29.5,
Uri => 19.6,
Nitin => 30,
Plato => 39,
);

# sorting of the hash according
# to the numerical value of
# Values of hash
foreach my $name (sort {$a<=>$b} values %studentnames)
{
    say $name;
}

• Output:

14
16.5
19.6
25.2
27
29.5
30
31
39

Time Complexity: O(n logn)
Auxiliary Space: O(n)

Sort the keys of the hash according to the values: You can also sort the keys of hash according to the given values. Example 1: In the below program, <=> is termed as the spaceship operator. If you will run the code again and again then you can notice the difference in the output. Sometimes you find Plato before the Nitin and vice-versa. 

Perl

# Perl program to demonstrate the
# Sorting of keys of the hash
# according to the values
use strict;
use warnings;
use 5.010;

# Creating a hash of studentnames
# and their average score
my %studentnames = (
Martha => 14,
Vivek => 27,
Earl => 31,
Marty => 16.5,
Jason => 25.2,
Socrates => 29.5,
Uri => 19.6,
Nitin => 45,
Plato => 45,
);

# Sort the keys of the hash
# according to the values
# Here $a and $b are the
# placeholder variable of sort
foreach my $name (sort {$studentnames{$a} <=>
       $studentnames{$b}} keys %studentnames)
{
    printf "%-8s %s\n", $name, $studentnames{$name};
}

• Output:

Martha   14
Marty    16.5
Uri      19.6
Jason    25.2
Vivek    27
Socrates 29.5
Earl     31
Plato    45
Nitin    45

Time Complexity: O(n logn)
Auxiliary Space: O(n)

Example 2: To solve the above code, keys having the same values can be sorted according to the ASCII table as follows: 

Perl

# Perl program to demonstrate the
# Sorting of keys of the hash
# according to the values
use strict;
use warnings;
use 5.010;

# Creating a hash of studentnames
# and their average score
my %studentnames = (
Martha => 14,
Vivek => 27,
Earl => 31,
Marty => 16.5,
Jason => 25.2,
Socrates => 29.5,
Uri => 19.6,
Nitin => 45,
Plato => 45,
);

# keys that have the same value
# will be sorted according the
# ASCII table
foreach my $name (sort { $studentnames{$a} <=> $studentnames{$b} or
                                $a cmp $b } keys %studentnames) {
    printf "%-8s %s\n", $name, $studentnames{$name};
}

Output:

Martha   14
Marty    16.5
Uri      19.6
Jason    25.2
Vivek    27
Socrates 29.5
Earl     31
Nitin    45
Plato    45

Explanation: Here you can see the key Nitin and Plato are sorted according to the ASCII table. It doesn’t matter how many times you run the code, the output will remain the same.

Time Complexity: O(n logn)
Auxiliary Space: O(n)