in reply to Re^4: Checking LInes in Text File
in thread Checking LInes in Text File

The following is ok for reasonable size files but may bog down when things get huge.

use strict;
use warnings;
use Data::Dump::Streamer;

my %firstLines;
my @lines;

while (<DATA>) {
    chomp;
    my ($data, $type) = /(.*)\s+TYPE:\s+(\w+)$/;
    
    next if ! defined $type; # ignore malformed line
    if (exists $firstLines{$data}) {
        $lines[$firstLines{$data}] .= ", $type";
    } else {
        $firstLines{$data} = @lines;
        push @lines, $_;
    }
}

print join "\n", @lines;

__DATA__
MCAT: 0xf30cbe01 PCAT: 0xcda2b409 LMAT: 0x00100000 TYPE: KA0
MCAT: 0xcc3fbed1 PCAT: 0x000fb109 LMAT: 0x00000800 TYPE: KA1
MCAT: 0xeeccbe01 PCAT: 0xcda2b409 LMAT: 0x00100000 TYPE: KA1
MCAT: 0xf30cbe91 PCAT: 0xafaddd09 LMAT: 0x00040000 TYPE: KA0
MCAT: 0xeeecbe01 PCAT: 0xcda2b409 LMAT: 0x00100000 TYPE: KA0
MCAT: 0xcc000331 PCAT: 0x000fb109 LMAT: 0x00000800 TYPE: KA1
MCAT: 0xe554be01 PCAT: 0xcda2b409 LMAT: 0x00100000 TYPE: KA1
MCAT: 0xf30cbe91 PCAT: 0xafaddd09 LMAT: 0x00040000 TYPE: KA1

[download]

Prints:

MCAT: 0xf30cbe01 PCAT: 0xcda2b409 LMAT: 0x00100000 TYPE: KA0
MCAT: 0xcc3fbed1 PCAT: 0x000fb109 LMAT: 0x00000800 TYPE: KA1
MCAT: 0xeeccbe01 PCAT: 0xcda2b409 LMAT: 0x00100000 TYPE: KA1
MCAT: 0xf30cbe91 PCAT: 0xafaddd09 LMAT: 0x00040000 TYPE: KA0, KA1
MCAT: 0xeeecbe01 PCAT: 0xcda2b409 LMAT: 0x00100000 TYPE: KA0
MCAT: 0xcc000331 PCAT: 0x000fb109 LMAT: 0x00000800 TYPE: KA1
MCAT: 0xe554be01 PCAT: 0xcda2b409 LMAT: 0x00100000 TYPE: KA1

[download]
DWIM is Perl's answer to Gödel

Replies are listed 'Best First'.
Re^6: Checking LInes in Text File
by Anonymous Monk on Jun 02, 2006 at 16:52 UTC
    Grandfather,

    Thank you very much!!! I really appreciate you solving that problem for me. I'm very new to PERL and have to admit that your code took a while to make sense to me. I didn't realize that you could access an array by the data value, I thought you had to access it by location (0,1,2,3... etc). Thanks again for your help!
[reply]

      Just in case there is some confusion or misunderstanding of some of the Perl tricks used I better go through some of that code and elaborate on what's happening. Note that I've taken interesting lines in processing order rather that the order they are coded.

      $firstLines{$data} = @lines;

      this is a little tricksy. It creates a new entry in %firstLines that contains the index to the new line as the value and is keyed by the unique part of the line contents. @lines in scalar context returns the number of elements in the array.

      if (exists $firstLines{$data})

      checks to see if we've already seen a specific line.

      $lines[$firstLines{$data}] .= ", $type";

      builds the multiple entries for duplicated lines. Note that $firstLines{$data} returns the index number that was stored earlier.

      DWIM is Perl's answer to Gödel
[reply]
[d/l]
[select]
        Thanks much for the added commentary! Now I think I actually understand what you did.
[reply]

In Section Seekers of Perl Wisdom