Parse txt file using perl

Question

I would like to parse a file containing data as below:

05\/26\/2013 06:09:47 \-0700 - AUTHN_SUCCESS - GET - ddsbcggio_ac  - 200.12.33.44 - abcweb.eegeserv.com\/abcweb\/abcwebInitialize.do?PORT=SPQ  - [email protected]\,ou=People\,o=zeb.com - 06:09:47 - http - uizweb_zam -  - [email protected]
05\/26\/2013 06:09:48 \-0700 - AUTHN_SUCCESS - GET - ddsbcggio_ac  - 200.12.33.44 - abcweb.eegeserv.com\/abcweb\/abcwebInitialize.do?PORT=SPQ  - [email protected]\,ou=People\,o=zeb.com - 06:09:48 - http - uizweb_zam -  - [email protected]
05\/26\/2013 06:09:49 \-0700 - AUTHN_SUCCESS - GET - ddsbcggio_ac  - 200.12.33.43 - abcweb.eegeserv.com\/abcweb\/abcwebInitialize.do?PORT=SPQ  - [email protected]\,ou=People\,o=zeb.com - 06:09:49 - http - uizweb_zam -  - [email protected]

and get:

05/26/2013 06:09:49  and [email protected],ou=People,o=zeb.com 
05/26/2013 06:09:48  and [email protected],ou=People,o=zeb.com

I tried split('-') but it does not work split('-') because as you can see : Some lines like the second one above has: [email protected] ( a '-' ) in between. Also some times, there are '-' in other part of data too.

Please help.

@amon I'll try. Presently i'm using split("-",$line), not how shall i change it to have space? — david
– david, Commented Jun 5, 2013 at 17:02
On the second line after "and get", you had backslash in front of ,ou and ,o, is that a typo? — doubleDown
– doubleDown, Commented Jun 5, 2013 at 17:25

David W. · Accepted Answer · 2013-06-05 18:34:07Z

You maybe better off using a regular expression. With a regular expression, I can quickly grab the part of the string I want with (...). See the Perldoc on Regular expressions to see what the various regular expression meta-characters mean.

#! /usr/bin/env perl

use 5.12.0;
use warnings;
use autodie;

while ( my $line = <DATA> ) {
    chomp $line;
    $line =~ s/\\//g;   #Remove all backslashes
    $line =~ /^(.+?) -.+?(uid=\S+)/;
    my $date = $1;
    my $uid = $2;
    say qq($date and $uid);
}

__DATA__
05\/26\/2013 06:09:47 \-0700 - AUTHN_SUCCESS - GET - ddsbcggio_ac  - 200.12.33.44 - abcweb.eegeserv.com\/abcweb\/abcwebInitialize.do?PORT=SPQ  - [email protected]\,ou=People\,o=zeb.com - 06:09:47 - http - uizweb_zam -  - [email protected]
05\/26\/2013 06:09:48 \-0700 - AUTHN_SUCCESS - GET - ddsbcggio_ac  - 200.12.33.44 - abcweb.eegeserv.com\/abcweb\/abcwebInitialize.do?PORT=SPQ  - [email protected]\,ou=People\,o=zeb.com - 06:09:48 - http - uizweb_zam -  - [email protected]
05\/26\/2013 06:09:49 \-0700 - AUTHN_SUCCESS - GET - ddsbcggio_ac  - 200.12.33.43 - abcweb.eegeserv.com\/abcweb\/abcwebInitialize.do?PORT=SPQ  - [email protected]\,ou=People\,o=zeb.com - 06:09:49 - http - uizweb_zam -  - [email protected]

Borodin · Accepted Answer · 2013-06-05 17:07:51Z

0

This program does what you ask. It looks like the field separator is ' - ', i.e. a hyphen with a space either side, giving a blank penultimate field (the eleventh).

This program expects the name of the input file as a parameter on the command line.

use strict;
use warnings;

while (<>) {
  chomp;
  tr/\\//d;
  my @fields = split /\x20-\x20/;
  printf "%s and %s\n", @fields[0,6];
}

With your own data, this produces

05/26/2013 06:09:47 -0700 and [email protected],ou=People,o=zeb.com
05/26/2013 06:09:48 -0700 and [email protected],ou=People,o=zeb.com
05/26/2013 06:09:49 -0700 and [email protected],ou=People,o=zeb.com

answered Jun 5, 2013 at 17:07

Borodin

127k9 gold badges72 silver badges146 bronze badges

5 Comments

david Over a year ago

What if the file has: uid=rad - [email protected],ou=People,o=zeb.com. Will this work?

Borodin Over a year ago

No. You are asking too much. Is that likely? Or even possible?

Borodin Over a year ago

If they are valid email addresses then they cannot contain whitespace.

david Over a year ago

Yes, i have something like [email protected],ou=Peo - ple,o=zeb.com. Please note ou=Peo - ple

Borodin Over a year ago

@david: Then please describe how to distinguish between " - " as a separator and the same thing within a field.

Collectives™ on Stack Overflow

Parse txt file using perl

2 Answers 2

Comments

5 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

5 Comments

Your Answer

Sign up or log in

Post as a guest

Related