Delete the last line of a file using perl

Question

sed '$d' $file; 

Using this command doesn't seem to work, as $ is a reserved symbol in Perl.

Answer 1

Don't know why are you using sed into Perl. Perl itself have standard module to delete last line from a file.

Use the standard (as of v5.8) Tie::File module and delete the last element from the tied array:

use Tie::File;

tie @lines, Tie::File, $file or die "can't update $file: $!";
delete $lines[-1];

Answer 2

Last line only

The closest syntax seem to be:

perl -ne 'print unless eof()'

This will act like sed, ie: without the requirement of reading the whole file into memory and could work with FIFO like STDIN.

See:

perl -ne 'print unless eof()' < <(seq 1 3)
1
2

or maybe:

perl -pe '$_=undef if eof()' < <(seq 1 3)
1
2

First and last lines

perl -pe '
    BEGIN {
        chomp(my $first= <>);
        print "Something special with $first\n";
    };
    do {
        chomp;
        print "Other speciality with $_\n";
        undef $_;
    } if eof();
  ' < <(seq 1 5)

will render:

Something special with 1
2
3
4
Other speciality with 5

Shortest: first and last line:

perl -pe 's/^/Something... / if$.==1||eof' < <(seq 1 5)

will render:

Something... 1
2
3
4
Something... 5

Try this:

perl -pe 'BEGIN{$s=join"|",qw|1 3 7 21|;};
          if ($.=~/^($s)$/||eof){s/^/---/}else{s/$/.../}' < <(seq 1 22)

... something like sed command:

sed '1ba;3ba;7ba;21ba;$ba;s/$/.../;bb;:a;s/^/---/;:b' < <(seq 1 22)

In a script file:

#!/usr/bin/perl -w

use strict;

sub something {
    chomp;
    print "Something special with $_.\n";
}

$_=<>;
something;

while (<>)  {
    if (eof) { something; }
    else { print; };
}

will give:

/tmp/script.pl < <(seq 1 5)
Something special with 1.
2
3
4
Something special with 5.

Answer 3

Hope you are trying to execute the 'sed' command from the middle of a perl script. I would recommend not to use this approach because it will work only in non-windows systems. Below is a perl'ish approach in which you can process first and last lines only rather than spending effort to delete file contents. Thank you.

Assuming "myfile.txt" as the input file:

open (FH, "<", "myfile.txt") or die "Unable to open \"myfile.txt\": $! \n";

$i = 1;

while(<FH>){
    next if ($i++ == 1 or eof);
    # Process other lines
    print "\nProcessing Line: $_";    
}

close (FH);

print "\n";
1;

myfile.txt -

# First line
This is the beginning of comment free file.
Hope this is also the line to get processed!
# last line

Result -

Processing Line: This is the beginning of comment free file.

Processing Line: Hope this is also the line to get processed!

Answer 4

This works for me, and it also removes unwanted empty lines at start and end.

use File::Slurp;
my $fileContent = read_file($csv_file);
my @lines = grep /\S/, split /\n/, $fileContent;
pop @lines;
$fileContent = join '', map "$_\n", @lines;
print $fileContent;
write_file($csv_file, $fileContent);