Anonymous Monk has asked for the wisdom of the Perl Monks concerning the following question:

Hi, problem I have been having is following, I know solution is there but I cant seem to get there.... I am reading by user defined file, which looks like: 
jan 2.76
feb 3.20
mar 1.6
apr 4.00

[download]
In the user file there might be empty lines which I wish to ignore, I would like to know how many lines actually have text on.......one of several versions of code attempts 
$empty[0] ="";
open (VER,$f_name);
while (defined ($rows = <VER>))
    {  
    chomp($rows); 
    ($rows1,$rows2)=split(/\ /,$rows);  
    $names[$l]=$rows1;
    $versions[$l]=$rows2;
    
    if ($names[$l] != $empty[0] ) #HOW should this be done???
        {
    $l++;
        }                    
    } 
close VER;

[download]
How to compare if the line is empty so that $l does not increase in case of empty table member.... BR Hewarn

Replies are listed 'Best First'.
Re: How to find empty in empty
by grep (Monsignor) on Nov 19, 2002 at 07:20 UTC
    use next inside your loop to skip the blanks. next will skip the current iteration and move on to the next for any loop structure. 
    while (defined ($rows = <VER>)) {  
    chomp($rows);
    next if ($rows eq '');   # this'll skip blank lines 
    ($rows1,$rows2)=split(/\ /,$rows);  
    $names[$l]=$rows1;
    $versions[$l]=$rows2;
}

    [download]


    grep
    Mynd you, mønk bites Kan be pretti nasti...
[reply]
[d/l]
Re: How to find empty in empty
by Monky Python (Scribe) on Nov 19, 2002 at 07:58 UTC
    Hi, some remarks:
    - use "use strict"
    - initialize your data before using it ($l is not initialized)
    - you don't need rows, row1, row2 as temporary variables

    The following script checks for an empty line with the
    mighty regexp /^[\s]+$/ instead of
    comparing with "". 

    #!/usr/local/bin/perl -w

use strict;

my (@names, @versions,$l,$e);

$l=0;
$e=0;
while (<DATA>) {
  chomp;
  if (! /^[\s]+$/) {
    ($names[$l],$versions[$l])=split(/[\s]+/);  
    $l++;
  }
  else {
    $e++;
  }
}

$,=", ";
print "@names @versions\n";
print "empty lines=$e\n";

__DATA__

jan 2.76
feb 3.20

mar 1.6
apr 4.00

    [download]
[reply]
[d/l]
[select]
      Um, '+' means at least one (and you don't need the box): ! /^\s*$/ Anyway, it's even easier just to check for a non-space: /\S/ But, as long as we're at it, why not accept only well-formed lines: 
      if ( /^(\w+)\s+([\d.]+)/ ) {

   ($names[$l],$versions[$l])=($1,$2);
   ++$l;

else { . . .

      [download]

        p
[reply]
[d/l]
[select]

Back to Seekers of Perl Wisdom