Another option is to use list for the line numbers. The algorithm scans only once through data file

# first read and numericaly sort the lines
my @lines = sort {$a <=> $b} <>;

# then loop over data on STDIN 
# and write it to STDOUT
my $line = shift @lines;
while (<>){
   if ($.==$line){
       s/foo/bar/g;
       $line = shift @lines;
   }
   print;
}
[download]

cat data | ./script index-file > result
[download]

Best regards

P.S. Code is not tested

cat data | ./script index-file > result
[download]

To avoid the "useless use of cat award"

./script index-file < data > result
[download]

;-)

--shmem

_($_=" "x(1<<5)."?\n".q·/)Oo.  G°\        /
                              /\_¯/(q    /
----------------------------  \__(m.====·.(_("always off the crowd"))."·
");sub _{s./.($e="'Itrs `mnsgdq Gdbj O`qkdq")=~y/"-y/#-z/;$e.e && print}

++karavelov

I didn't realise how deep the magic of <> was.

perlop says:

The <> symbol will return undef for end-of-file only once. If you call it again after this, it will assume you are processing another @ARGV list, and if you haven't set @ARGV, will read input from STDIN.

---

Unless I state otherwise, all my code runs with strict and warnings

Read your index file into a hash:

open my $idx, '<', 'index.file' or die "Can't open 'index.file' for re
+ading: $!";
my %index;
while (<$idx>){
    chomp;
    $index{$_} = 1;
}
[download]

And then go through your data file, and replace as needed while copying to a temporary file:

while (<$in>) {
    if ($index{$.}){
        s/foo/bar/g;
    }
    print $tmp $_;
}
[download]

When you're done, override the data file with the temporary file.

Use Tie::File, sort your index file by line number and then do the replacement. That way, you will progress forward through your large file once without having to start over from the beginning.