line number of the current file handle is stored in the special variable $.

You may note that $. is reset only by an explicit close of the filehandle, so if the same filehandle is reopened $. continue grows between files. This can be avoided with close ARGV if eof that is very useful for oneliners.

perl -lne "print qq($.)" two_lines.log 13_lines.log
1
2
3  #<--here starts second file          
4
5
6
7
8
9
10
11
12
13
14
15

perl -lne "print qq($.); close ARGV if eof" two_lines.log 13_lines.log
1
2
1  #<--here starts second file
2
3
4
5
6
7
8
9
10
11
12
13
[download]

L*

$. perlvar:

test.txt:

line 1
line 2
line 3
line 4
line 5
line 99
[download]

ln.pl:

#!/usr/bin/perl

use strict;
use warnings;

open(my $fh, "<", "test.txt")
    or die "cannot open < test.txt: $!";

while (my $row = <$fh>) {
  chomp $row;
  # do something 
  print "line number $.\n";
}
[download]

output:

line number 1
line number 2
line number 3
line number 4
line number 5
line number 6
[download]

update: forgot to add the output.

what I would do is use signals to give a nudge to the long running process...

I have included a print statement (you can make that to log one single entry in a logfile) OR, also fancy, modify the name of the script $0 directly, hence, you get the information through ps -ef (used by some programs). There are more ways, like named pipes, etc.

#!/usr/bin/perl

my $SHOUT = 0;
$SIG{USR1} = sub {
        ++$SHOUT;
};

my $PROGRAM = $0;
print "$PROGRAM started with pid $$\n";

my $counter = 0;
while(1) {
        $counter++;
        if($SHOUT){
                print "$0 says: $counter \n";
                $0 = $PROGRAM . " counter=$counter";
                $SHOUT = 0;
        }
        sleep 1;
}
[download]

The nudge is given with a kill -USR1 $PID

Update:

The normal way is to have a second program that asks for ETA and status. This second program send the signal, then both programs then initiate a temporal TCP connection to send/receive the information. But from what I read in your post, you want to keep it simple. The other way is to only update your linenumber logfile each 10 seconds. you will need $NEXTUPDATE = time()+1000*10 and do compares if(time()>$NEXTUPDATE) This will use much less IO than logging each line.

edit2: added missing my's. And Do'h, suggesting same things as peers... sorry for that.

*important note*: Do not use IO in signals. Your script could coredump if you open files and such. Do the work outside.

You could maintain a status file. Open it once, at the beginning, then seek($stat_hf, 0, 0); just before you write the current status to the file. This is a lot of extra IO, so you might want to do it once every 5 or 10 input lines.

Unfortunately, some operating systems might auto-lock the status file against even reads from other processes. The preferred "fix" for that is to open it with options to only lock the file for writes. (but I don't know how to do that.) A common work-around is to close and re-open the file for each update. This incurs much more extra IO, so you'd only want to update once for every 50 or 100 input lines.