in reply to (jeffa) 3Re: Searching for variable then blank lines
in thread Searching for variable then blank lines

SUBJECT WAS: Searching for variable then blank lines

Thanks to everyone who has helped me on this. I must be dense. :-\ Now I am getting my script to read in my file and go down to the beginning of the line starting with ARCSERVE.NLM, and it gives me *every* line after that. My script is not not exiting when it finds the first 2 returns after ARCSERVE.NLM for the first time. Below is my script and a sample of the output file.

use strict;
use File::Basename;

##********  Variables  ********##
my $filepath = 'c:\a\netware';
my $dump = 'Netware_hw.txt';
my $arc = 'ARCSERVE.NLM'; 
my $serverfile = 'dallas.txt';


    open (SERVER, "<$serverfile") or die "Can't open the file\n";
    open (OUTPUT, ">>$filepath\\$dump");

    my @lines= <SERVER>;
    my ($s, $b);
    foreach my $line (@lines)
    {
      if ($line =~ /^\n$/) {
        $b++;
        last if (($b == 2) && $s);
      } else {
        $b = 0;
  }

  if ($line =~ /$arc/) {
    $s = 'true';
  }

  if ($s) {
    print OUTPUT $line;
  }

}



close (OUTPUT);
close (SERVER);

===========
OUTPUT FILE
===========

ARCSERVE.NLM
  Version:            1
  Date:                10/22/2000
  ID:                000
  Parent ID:            x
  Type:                UNKNOWN
  Description:            ARCserve
  Flags:            

  
ASDB.NLM
  Version:            2
  Date:                12/8/2000


ASPIBD.NLM
  Version:            4
  Date:                12/15/2000
  ID:                0

[download]
This is just an example. My file was much larger than this. I was not wanting the information starting with ASDB.NLM to the end of the file. What do I need to do to make this happen?

Thx again!

Replies are listed 'Best First'.
(jeffa) Re: Exiting when parsing files
by jeffa (Bishop) on Mar 15, 2002 at 15:05 UTC
    Hi rscott212. Sorry for the delay ... but you don't want to hear my problems ... ;)

    Ok, i changed this line: 

    # match a line that only contains one new line
if ($line =~ /^\n$/) {

    [download]
    To this: 
    # match a line that only contains one or more whitespace
if ($line =~ /^\s+$/) {

    [download]
    And it worked ... there was one extra newline at the end of the extracted data, but i'll get to that in a moment. (And for the record, one of your 'blank lines' had a single space in it.)

    Now, some thoughts on your thoughts. First, please learn how to indent properly. When i first glanced at your code, i really believed that the last two if blocks were OUTSIDE the foreach loop. Please, for the sanity of those trying to help, use good formatting: 

    foreach (@blah) {
   if ($foo) {
      # foo stuff
   }
   else {
      # other stuff
   }
}

    [download]
    That makes all the difference in the world. Second, the integer 1 (one) makes a great true value. Instead of assigning a boolean flag the scalar value 'TRUE', just use 1 (one).

    Last, you have a lot of unecessary code - here is a slight improvement for you to ponder: 

    # open filehandles ...

my ($s, $b) = (0,0);
my $arc     = 'ARCSERVE.NLM'; 
my @lines   = <DATA>;

foreach my $line (@lines) {
    if ($line =~ /^\s+$/ and $s) {
        $b++;
        last if $b == 2;
    }
    $s = 1 if $line =~ /$arc/;

    print OUTPUT $line if $s and $line !~ /^\s+$/;
}

    [download]
    This produces (using your data file) the following output: 
    ARCSERVE.NLM
  Version:            1
  Date:                10/22/2000
  ID:                000
  Parent ID:            x
  Type:                UNKNOWN
  Description:            ARCserve
  Flags:
    Hope this helps. :)

    jeffa

    L-LL-L--L-LL-L--L-LL-L--
-R--R-RR-R--R-RR-R--R-RR
B--B--B--B--B--B--B--B--
H---H---H---H---H---H---
(the triplet paradiddle with high-hat)
[reply]
[d/l]
[select]

In Section Seekers of Perl Wisdom