Sed in perl

corfuitl · 08-05-2013, 12:38 PM

hi,

Could you tell me how can I use the sed command in a perl script? Basically, my perl script extracts a variable which is a number of line ($line) of a i file. I want print the same line of j file. In bash the command should be

Code:

sed -n "${line}p" j.txt >> output.txt

Thanks!

codeguy · 08-05-2013, 01:21 PM

Not sure if you want a single line (perl golf style) or a script. I'll go script.

Code:

#!/usr/bin/perl

# this is the line to print
my $line = 42;

# we'll read from stdin
while (<>)
{
  if ($. == $line) {
     print;
     last;
  }
}

corfuitl · 08-05-2013, 01:27 PM

hi and thanks for your reply.

I want a single line or more but not a script. I am not sure if your lines will work for me since need to import the line from another file.
Thanks!

konsolebox · 08-05-2013, 01:35 PM

codeguy's script for one line could be like this:

Code:

perl -ne 'if ($. == 42) { print; last; }' file

As for your original sed code it's equivalent to this:

Code:

perl -ne "if ($. == ${line}) { print; last; }" j.txt >> output

TB0ne · 08-05-2013, 02:21 PM

Quote:

Originally Posted by corfuitl

hi,
Could you tell me how can I use the sed command in a perl script? Basically, my perl script extracts a variable which is a number of line ($line) of a i file. I want print the same line of j file. In bash the command should be

Code:

sed -n "${line}p" j.txt >> output.txt

I'm a bit confused here....if the script already extracts a variable which is a line number, why do you need sed to do anything, since you've already got the variable???

The way I'm reading it, you either get the line number from file "i", which is the $line variable, and want to print the same line number from file "j":

Code:

open (FILEJ, $filename);
@lines = <FILEJ>;
close(FILEJ);

#Get the incoming line from whatever, subtract 1 for the offset.
$line = $line-1;
print $lines[$line];

So you just read the file into an array, and hork out the line you want. Of course, with a huge file that can be problematic.

Or you want to search for a particular pattern in file j which you get from file i, using sed operators in the variable definition:

Code:

my $find = "<some string or regex goes here>";
open FILE, "<filej";
my @line = <FILE>;
print "Lines that match $find\n";
for (@lines) {
    if ($_ =~ /$find/) {
        print "$_\n";
    }
}

corfuitl · 08-05-2013, 02:45 PM

I think that both work perfectly. I will test them and I will be back soon.
Thank you