Use Perl to search for a match between two files - regex

Question

I have two files: file1.txt and file2.txt. Both contains lines in this format:
file1
name1:value1
file2
name2:value2
I want to check if value1 is found in list2 (in the name2 string)
I have this function:
#!/usr/bin/perl
use warnings;
use Parallel::ForkManager;
sub loadf($);
print "Starting main program\n";
my #list1 = loadf("list1.txt");
my #list2 = loadf("list2.txt");
my $workernum = 10;
open(OK, '>>', 'valid.txt');
open(ER, '>>', 'invalid.txt');
$pm = new Parallel::ForkManager($workernum);
my $cnt = 0;
foreach $line (#list1) {
$cnt++;
my $pid = $pm->start and next;
my #data1 = split(":", $line);
my $name1 = $data1[0];
my $value1 = $data1[1];
my #data2 = split(":", $list2);
my $name2 = $data2[0];
my $value2 = $data2[1];
if (/$value1/i ~~ #list2)
{
print OK $name1 . " - " . $value2 . "\n";
print " [+] Found: " . $name1 . " - " . $value2 . "\n";
}
else
{
print ER $name1 . "\n";
print " [x] Unknown: " . $name1 . " - " . $value1 . "\n";
}
$pm->finish;
}
close(OK);
close(ER);
print "\n*** Finished ***\n";
sub loadf($) {
my #file;
open(FILE, $_[0] . "\n") or die("[+] Couldn't open " . $_[0] . "\n");
#file = <FILE>;
close(FILE);
return #file;
}
__END__
which is not working. What am I doing wrong ?

Answer 1

#!/usr/bin/perl
open(F,'list1.txt');
my #list1=<F>;
close(F);
open(F,'list2.txt');
my #list2=<F>;
close(F);
chomp(#list1,#list2);
foreach my $line (#list1)
{
if ($line=~/.+\:.+/)
{
my #data1 = split(":", $line);
if (my #d2=grep /$data1[1]\:/i,#list2){print " [+] Found: " . $data1[0] . " - " . [split(':',$d2[0])]->[1] . "\n"; }
else { print " [x] Unknown: " . $data1[0] . " - " . $data1[1] . "\n"; }
}
}

Answer 2

Erm... I really don't get your algorithm, sorry. First you read all the lines from compared files into two arrays, then you forking the processor loop but, as I see it, make no attempts to split these workers' job into chunks (to parallel it).
I'd suggest trying a bit different approach: slurp only the second file, then process the first file line by line. You didn't mention whether names and values of the second file are unique; I suppose they're not, but the program could be made even simpler if they are.
open my $caf, '<', 'list2.txt' or die $!, "\n";
my $checkedAgainst = do { local $/; <$caf>; };
open my $cf, '<', 'list1.txt' or die $!, "\n";
my $workernum = 10;
$pm = new Parallel::ForkManager($workernum);
while (<$cf>) {
my $pid = $pm->start and next;
my ($nameToCheck, $valueToCheck) = split /:/;
if ($checkedAgainst =~ /^\Q$valueToCheck\E:(.+)$/m) {
print " [+] Found: $nameToCheck - $1", "\n";
}
else {
print " [x] Unknown: $nameToCheck - $valueToCheck", "\n";
}
}
$pm->finish;
In other words, I first load the second file into one big string, then try to match it with the lines from the first file (line by line). I don't know, what symbols could appear in your values, that's why \Q-\E (quotemeta operators) are used there.
UPDATE: attempted to make this code fork, have no means to test it where I stand though.