use warnings;
use strict;

while (<DATA>) {
    tr/-abcdef/9012345/;
    print substr($_, 0, 10);
    print "\n";
}

__DATA__
a81f9284-b353-430a-a8ce-85b8118d3f3e
[download]

• perldoc perlop
• perldoc perlre

To grab the output of an external command, such as uuidgen, you should use qx instead of system.

You don't want system('uuidgen') here. system executes a command, but doesn't return the output of the command (it returns the command's exit status - the equivalent of $? in shell scripts). You want backticks.

A fairly direct translation would be:

my $string = `uuidgen`;
$string =~ tr [\-abcdef] [9012345];
my $first_10 = substr($string, 0, 10);
[download]

The tr operator (is it technically an operator? it's documented in perlop) acts much like the 'tr' command in many shells. substr extracts a substring from a string. The first argument is the string to extract from; the second argument is the character to start at (in this example, 0, so start at the beginning of the string) and the third is the number of characters to take (10 here). The third argument can be omitted, in which case substr extracts to the end of the string.

A slightly better way though might be to avoid spawning the external program and use Data::UUID (or a similar module like UUID::Tiny).

use Data::UUID;
my $string = Data::UUID->new->create_str;
$string =~ tr [\-abcdef] [9012345];
my $first_10 = substr($string, 0, 10);
[download]

use UUID::Tiny;
my $string = create_UUID();
$string =~ tr [\-abcdef] [9012345];
my $first_10 = substr($string, 0, 10);
[download]

Thanks to toolic and tobyink, it worked like a charm!

yes, tr/// (also known as y///) is the transliteration operator, much like m// is the match operator, and s/// is the substitution operator, and =~ is the binding operator

See also: My Favourite Regex Tools