Perl code that breaks a line after fixed number of columns

Question

Here is a simple program that I created to print a file by breaking it after a fixed number of columns. I feel it can be optimized but not sure how. Any feedback is great.

#This is a simple program that takes a file and the limits the columns length to
#the number passed in as second argument
if ($#ARGV < 1) {
   printf("Usage: perl %s <name of the input file> <column length>\n", $0);
   exit(-1);
}
$file=$ARGV[0];
open(INFO, $file) or die("Could not open  file.");
$LINE_LENGTH = $ARGV[1];
$count = 0;
foreach $line (<INFO>)  {
    #print $line;
    if (length($line) <= $LINE_LENGTH) {
      printf("%s\n", $line);
    } else {
      $remaining_chars  = length ($line);
      $remaining_line = $line;
      while ($remaining_chars >= $LINE_LENGTH) {
          $cur_str = substr($remaining_line, 0, $LINE_LENGTH - 1 );
          printf("%s\n", $cur_str );
          $remaining_line = substr($remaining_line, $LINE_LENGTH - 1);
          $remaining_chars = $remaining_chars - length($cur_str);
      }
      printf("%s",$remaining_line );
    }
    if ($++counter == 2){
      last;
    }
}
printf("\n");
close(INFO);

Answer 1

I am not sure it can be optimized in a sense of better performance. It could be streamlined though. The initial test for length($line) <= $LINE_LENGTH is just redundant. Compare:

    while (length ($line) > LINE_LENGTH) {
        $prefix = substr($line, 0, $LINE_LENGTH - 1 );
        $line = substr($line, $LINE_LENGTH - 1);
        printf("%s\n", $prefix);
    }
    printf("%s",$line );

PS: I am not familiar with perl enough. If there is a split at position function, you may use it instead of calling substr twice.

Answer 2

You can use regular expression instead of substr(),

use strict;
use warnings;

if (@ARGV != 2) {
   die("Usage: perl $0 <name of the input file> <column length>\n");
}    
my ($file, $LINE_LENGTH) = @ARGV;

open(my $INFO, "<", $file) or die($!);
my $count = 0;
while (my $line = <$INFO>)  {

    print $1, "\n" while $line =~ /( .{1,$LINE_LENGTH} )/xg;

    # ...
    if (++$count == 2) {
      last;
    }
}
print "\n";
close($INFO);

Answer 3

use strict;

^^^ always.

You will be saved from embarassing things, like this:

if ($++counter == 2){
  last;
}

Seriously, what is that? ;-)

With perl it's often best to put required arguments first, and then use ARGV magic to handle the input files. In other words, put the column width first, and then the input files... The ARGV magic is powerful, it will read all the files given on the commandline, or, read the STDIN if no files are given. By shifting the column width off the @ARGV you can let the magic do its thing.

chomp is another nice trick, it removes any trailing newlines, if any.

The lines can be printed with brute-force substrings in a while loop then.

I would write your code as:

#!/usr/bin/perl

my $cols = shift @ARGV;
# Set a default, if empty
$cols ||= 80;

while (my $line = <>) {
    chomp $line;
    while (length($line) > $cols) {
        print substr($line, 0, $cols) . "\n";
        $line = substr($line, $cols);
    }
    print $line . "\n";
}

Note, assuming this is saved as a file called "narrow", then this can now be called in many ways:

• narrow 80 *.java
• curl http://example.com | narrow
• curl http://example.com | narrow 120
• etc.