Read space delimited text file into array of hashes [Perl] - arrays

Question

I have text file that matches the following format:
1 4730 1031782 init
4 0 6 events
2190 450 0 top
21413 5928 1 sshd
22355 1970 2009 find
And I need to read it into a data structure in perl that will allow me to sort and print according to any of those columns.
From left to right the columns are process_id, memory_size, cpu_time and program_name.
How can I read a text file with formatting like that in a way that allows me to sort the data structure and print it according to the sort?
My attempt so far:
my %tasks;
sub open_file{
if (open (my $input, "task_file" || die "$!\n")){
print "Success!\n";
while( my $line = <$input> ) {
chomp($line);
($process_id, $memory_size, $cpu_time, $program_name) = split( /\s/, $line, 4);
$tasks{$process_id} = $process_id;
$tasks{$memory_size} = $memory_size;
$tasks{$cpu_time} = $cpu_time;
$tasks{$program_name} = $program_name;
print "$tasks{$process_id} $tasks{$memory_size} $tasks{$cpu_time} $tasks{$program_name}\n";
}
This does print the output correctly, however I can't figure out how to then sort my resulting %tasks hash by a specific column (i.e. process_id, or any other column) and print the whole data structure in a sorted format.

Answer 1

You're storing the values under keys that are equal to the values. Use Data::Dumper to inspect the structure:
use Data::Dumper;
# ...
print Dumper(\%tasks);
You can store the pids in a hash of hashes, using the value of each column as the inner key.
#!/usr/bin/perl
use strict;
use warnings;
use feature qw{ say };
my #COLUMNS = qw( memory cpu program );
my %sort_strings = ( program => sub { $a cmp $b } );
my (%process_details, %sort);
while (<DATA>) {
my ($process_id, $memory_size, $cpu_time, $program_name) = split;
$process_details{$process_id} = { memory => $memory_size,
cpu => $cpu_time,
program => $program_name };
undef $sort{memory}{$memory_size}{$process_id};
undef $sort{cpu}{$cpu_time}{$process_id};
undef $sort{program}{$program_name}{$process_id};
}
say 'By pid:';
say join ', ', $_, #{ $process_details{$_} }{#COLUMNS}
for sort { $a <=> $b } keys %process_details;
for my $column (#COLUMNS) {
say "\nBy $column:";
my $cmp = $sort_strings{$column} || sub { $a <=> $b };
for my $value (sort $cmp keys %{ $sort{$column} }
) {
my #pids = keys %{ $sort{$column}{$value} };
say join ', ', $_, #{ $process_details{$_} }{#COLUMNS}
for #pids;
}
}
__DATA__
1 4730 1031782 init
4 0 6 events
2190 450 0 top
21413 5928 1 sshd
22355 1970 2009 find
But if the data aren't really large and the sorting isn't time critical, just sorting the whole array of arrays by a given column is much easier to write and read:
#!/usr/bin/perl
use strict;
use feature qw{ say };
use warnings;
use enum qw( PID MEMORY CPU PROGRAM );
my #COLUMN_NAMES = qw( pid memory cpu program );
my %sort_strings = ((PROGRAM) => 1);
my #tasks;
push #tasks, [ split ] while <DATA>;
for my $column_index (0 .. $#COLUMN_NAMES) {
say "\nBy $COLUMN_NAMES[$column_index]:";
my $sort = $sort_strings{$column_index}
? sub { $a->[$column_index] cmp $b->[$column_index] }
: sub { $a->[$column_index] <=> $b->[$column_index] };
say "#$_" for sort $sort #tasks;
}
__DATA__
...
You need to install the enum distribution.

Answer 2

I can't figure out how to then sort my resulting %tasks hash by a specific column
You can't sort a hash. You need to convert each of your input rows in a hash (which you're doing successfully) and then store all of those hashes in an array. You can then print the contents of the array in a sorted order.
This seems to do what you want:
#!/usr/bin/perl
use strict;
use warnings;
use feature 'say';
my #cols = qw[process_id memory_size cpu_time program_name];
#ARGV or die "Usage: $0 [sort_order]\n";
my $sort = lc shift;
if (! grep { $_ eq $sort } #cols ) {
die "$sort is not a valid sort order.\n"
. "Valid sort orders are: ", join('/', #cols), "\n";
}
my #data;
while (<DATA>) {
chomp;
my %rec;
#rec{#cols} = split;
push #data, \%rec;
}
if ($sort eq $cols[-1]) {
# Do a string sort
for (sort { $a->{$sort} cmp $b->{$sort} } #data) {
say join ' ', #{$_}{#cols};
}
} else {
# Do a numeric sort
for (sort { $a->{$sort} <=> $b->{$sort} } #data) {
say join ' ', #{$_}{#cols};
}
}
__DATA__
1 4730 1031782 init
4 0 6 events
2190 450 0 top
21413 5928 1 sshd
22355 1970 2009 find
I've used the built-in DATA filehandle to make the code simpler. You would need to replace that with some code to read from an external file.
I've used a hash slice to simplify reading the data into a hash.
The column that you want to sort by is passed into the program as a command-line argument.
Note that you have to sort the last column (the program name) using string comparison and all other columns using numeric comparison.

Answer 3

This decides how to sort using the first argument the script receives.
#!/usr/bin/env perl
use strict;
use warnings;
use feature 'say';
open my $fh, '<', 'task_file';
my #tasks;
my %sort_by = (
process_id=>0,
memory_size=>1,
cpu_time=>2,
program_name=>3
);
my $sort_by = defined $sort_by{defined $ARGV[0]?$ARGV[0]:0} ? $sort_by{$ARGV[0]} : 0;
while (<$fh>) {
push #tasks, [split /\s+/, $_];
}
#tasks = sort {
if ($b->[$sort_by] =~ /^[0-9]+$/ ) {
$b->[$sort_by] <=> $a->[$sort_by];
} else {
$a->[$sort_by] cmp $b->[$sort_by];
}
} #tasks;
for (#tasks) {
say join ' ', #{$_};
}