perl

hardbop200 · 10-27-2008, 05:11 PM

Hello all,

I am working on a project where I read incoming data from a RFID reader, and spit the data into a database. The data coming from the device looks like this:

Tag:4845 4930 3030 3030 3030 3033, Disc:2008/10/27 14:37:02.015, Last:2008/10/27 14:37:02.015, Count:1, Ant:0, Proto:2

My script reads this data into an array, splitting on the comma:

Code:

#!/usr/bin/perl -w
use strict;
use warnings;
use IO::Socket;

my $port = 4000;
my @data;
my $server = IO::Socket::INET->new (
	Proto => 'tcp',
	LocalPort => $port,
	Listen => SOMAXCONN,
	);

print "Server running on port $port...\n";
while (my $connection = $server->accept) {
	print "Client connected at ", scalar(localtime), "\n";
	while (<$connection>) {
		@data = split (/,/);
		print @data;
	}
	close $connection;
	print "Client disconnected\n";
}

My problem is this: I *know* that I need only 6 fields in my array: tag, disc, last, count, ant and proto; however, the script doesn't allow me to define where an array stops and a new one begins. For example, I would like to do something like this:

my @data = $tag, $disc, $count...etc

So that when this part of the script reads:

Code:

while (<$connection>) {
		@data = split (/,/);
		print @data;
	}

...it stops on the 6th field and creates a new array.

Is this making sense? All of the textbook examples have the array's data defined at the beginning of the script, which obviously isn't going to happen here; I need to insert the data as it comes in, and access it (so I can then insert it).

Help!

Thanks,

Josh

chrism01 · 10-27-2008, 05:30 PM

Actually, its not entirely clear to me, but I'll have a go.
Are you getting 6 fields at a time into the array, or 1 field at a time ie a 6-field rec, or an unending stream of individual fields, 1 at a time that happen to be csv?
What do you need to do with the data?
Are you going to process each array / 6-fields as they arrive, or are you planning to store a load of them and process them later?

PS please use code tags, it makes it easier to read your code.

keefaz · 10-27-2008, 07:50 PM

Maybe insert a newline character after Proto:xx

Code:

while (<$connection>) {
    s/(Proto:\d)/$1\n/g;
    print;
    # if you need data to process...
    @data = split (/\n/);
}

Edit: not sure it is a good idea with a live stream from connection...
Better to loop and check end of infos blocks with /Proto:\d/ or similar

estabroo · 10-27-2008, 08:08 PM

A q&d way of doing this would be to have a counter that increases whenver you see the proto tag and push the data into an anonymous array in a hash

kind of like

Code:

$counter = 0;
while (<$connection>) {
    @data = split(/,/);
    foreach $item (@data) {
        push @{$h_data{$counter}}, $item;
        if ($item =~ /Proto: /) {
            $counter++;
        } 
    }   
}
# counter ends up with the total number of records read

Then to access any particular record

Code:

@data = @{$h_data{$record_number}};
$tag = $data[0];

Alternatively you could use an anonymous hash and then pull out the information by name, just split each data item on the :

Code:

while (<$connection>) {
    @data = split(/,/);
    foreach $item (@data) {
        ($key, $val) = split(/:/,$item,2);
        $h_data{$counter}->{$key} = $val;
        if ($item =~ /Proto: /) {
            $counter++;
        }
    }
}

Now if it's a continuous stream then instead of just incrementing the counter you could call a function with the data, just make sure you delete the entry when your finished so you don't run out of memory.

hardbop200 · 10-27-2008, 09:35 PM

Quote:

Originally Posted by estabroo

Now if it's a continuous stream then instead of just incrementing the counter you could call a function with the data, just make sure you delete the entry when your finished so you don't run out of memory.

Thanks, I'll try all of these suggestions and see which one works out best.

You were right about the continuous stream; I'll give this approach a try and see if I can figure this out and I'll post my results.

Thanks for all of the information, and apologies for my post being a little hazy...I'm very new to Perl, just trying to use it to "get the job done!"

Thanks again,

Josh

Sergei Steshenko · 10-28-2008, 03:24 AM

The following example might help:

Code:

sergei@amdam2:~/junk> cat -n arrays_and_vars.pl
     1  #!/usr/bin/perl -w
     2
     3  use strict;
     4
     5  my @array = (0, 1, 2, 3, 4);
     6
     7  my ($field0, $field1, $field2) = @array;
     8
     9  warn "\$field0=$field0 \$field1=$field1 \$field2=$field2";
sergei@amdam2:~/junk> ./arrays_and_vars.pl
$field0=0 $field1=1 $field2=2 at ./arrays_and_vars.pl line 9.
sergei@amdam2:~/junk>

hardbop200 · 10-28-2008, 10:46 AM

OK, lots of good stuff happening with this script!

Here's what ended up working:

Code:

while (my $data = <$connection>) {
		foreach (split(/\n/,$data)) {
			#print $data;
			my ($field0, $field1, $field2, $field3, $field4, $field5) = (split(/\,/,$_))[0,1,2,3,4,5];

This returns nicely formatted columns like:

Code:

Tag:4845 4930 3030 3030 3030 3032

Now I would like to run each item ($field0, $field1, etc.) through another parser in a subroutine that will split each column by the colon ":", with the first part of the split being $key (which will be discarded), and $value (which will actually be inserted; Here is the part of the code that sends $field0...$field5 to the parser:

Code:

my ($field0, $field1, $field2, $field3, $field4, $field5) = (split(/\,/,$_))[0,1,2,3,4,5];
			&parse_text (@data);
			print "$field0\n";

...and here is the parser:

Code:

sub parse_text {
	foreach (@_) {
		my ($key, $var) = (split(/:/,@_))[0,1];
		$var = @_;
		return @_;
	}
}

Basically, I want to send the parser $field0, split it, assign the result of $var = $field0, and return the new-and-improved $field0 back to my main program. Yet, the $var = @_ doesn't seem to work, and return @_ sure doesn't work. Any ideas?

hardbop200 · 10-28-2008, 10:56 AM

I spoke too soon, this works great:

Code:

#!/usr/bin/perl -w
#use strict;
use warnings;
use IO::Socket;

my $port = 4000;
my $server = IO::Socket::INET->new (
	Proto => 'tcp',
	LocalPort => $port,
	Listen => SOMAXCONN,
	);

print "Server running on port $port...\n";
while (my $connection = $server->accept) {
	print "Client connected at ", scalar(localtime), "\n";
	while (my $data = <$connection>) {
		foreach (split(/\n/,$data)) {
			#print $data;
			#my ($field0, $field1, $field2, $field3, $field4, $field5) = (split(/\,/,$_))[0,1,2,3,4,5];
			my (@data) = (split(/\,/,$_))[0,1,2,3,4,5];
			foreach (@data) {
				my ($key, $var) = (split(/:/,$_))[0,1];
				print "$var\n";
				}
			}
	}
	close $connection;
	print "Client disconnected\n";
}

Sergei Steshenko · 10-28-2008, 11:20 AM

Quote:

Originally Posted by hardbop200

I spoke too soon, this works great:

Code:

#!/usr/bin/perl -w
#use strict;
use warnings;
use IO::Socket;

my $port = 4000;
my $server = IO::Socket::INET->new (
	Proto => 'tcp',
	LocalPort => $port,
	Listen => SOMAXCONN,
	);

print "Server running on port $port...\n";
while (my $connection = $server->accept) {
	print "Client connected at ", scalar(localtime), "\n";
	while (my $data = <$connection>) {
		foreach (split(/\n/,$data)) {
			#print $data;
			#my ($field0, $field1, $field2, $field3, $field4, $field5) = (split(/\,/,$_))[0,1,2,3,4,5];
			my (@data) = (split(/\,/,$_))[0,1,2,3,4,5];
			foreach (@data) {
				my ($key, $var) = (split(/:/,$_))[0,1];
				print "$var\n";
				}
			}
	}
	close $connection;
	print "Client disconnected\n";
}

Whatever you are writing, "while (my $data = <$connection>)" is bad,
it should rather be "while(defined(my $data = <$connection>))".

It's written in the manual, FAQ, etc.

The point is relationship between TRUE, FALSE, 0, undef.