Regex Doubt

[Anonymous Monk]

Hello Monks,

I have a doubt regarding how to extract the locations from the data. Pls hav a look at the below code,

$data = "Part Time - Weekend Receptionist - Barton le Clay - Barton-le
+-Clay (near Luton)";
$data = "Part-Time Weekend Day Veterinary Nurse - Milton Keynes";
$data = "Receptionist - Barton-le-Clay (near Luton)";
$data = "Full Time - Veterinary Nurse - Glasgow Referral - Glasgow";

The above are the various types. The below is the code I am currently using.!!!

$location = $1  if ($data =~ /-(.*)/i);
$location = $1  if ($location =~ /-(.*)/i);

Now I am using the above weird technique to extract this. But its noe reliable. Any Monks with a Helping hand. Many thanks in advance.

[Athanasius]

Perhaps you’re looking for something along these lines?

#! perl
use strict;
use warnings;
use Data::Dump;

my @records;

while (my $line = <DATA>)
{
    chomp(my @r = split / - /, $line);
    push @records, [ @r ];
}

dd @records;

__DATA__
Part Time - Weekend Receptionist - Barton le Clay - Barton-le-Clay (ne
+ar Luton)
Part-Time Weekend Day Veterinary Nurse - Milton Keynes
Receptionist - Barton-le-Clay (near Luton)
Full Time - Veterinary Nurse - Glasgow Referral - Glasgow

Output:

18:58 >perl 381_SoPW.pl
(
  [
    "Part Time",
    "Weekend Receptionist",
    "Barton le Clay",
    "Barton-le-Clay (near Luton)",
  ],
  ["Part-Time Weekend Day Veterinary Nurse", "Milton Keynes"],
  ["Receptionist", "Barton-le-Clay (near Luton)"],
  ["Full Time", "Veterinary Nurse", "Glasgow Referral", "Glasgow"],
)

18:59 >

Hope that helps,

[Anonymous Monk]

Thanks Athanasius. :)

[space_monk]

The only reliable way to detect the location appears to be to lock the regex to the end of the string and then do a non-greedy match from the last " - " (dash with spaces)

[pvaldes]

I love to cheat (and I love to check too). Look at the last line of data

/i is said yet, so its use here is intentionally omitted

use strict; use warnings;

while (<DATA>){
if (/([Bb]arton-[Ll]e-[Cc]lay|[Mm]ilton\s+[Kk]eynes|[Gg]lasgow|[Bb]art
+on-le-[cC]lay \(near [Ll]utton\))\s?$/){ print $1,"\n"}
else {print "-- what? --\n"}
}

__DATA__
Part Time - Weekend Receptionist - Barton le Clay - Barton-le-Clay (ne
+ar Luton)
Part-Time Weekend Day Veterinary Nurse - Milton keynes 
Receptionist - barton-le-clay (near luton) 
Full time liar - Glasgow (Australia)

[2teez]

Hi,
Playing around with regex, I came up with this. This is not absolute but it solves this problem, and some others.
Assuming that the locations you want to extract are the names of places after job titles.

use warnings;
use strict;

while (<DATA>) {
    chomp;
    my $str = reverse $_;
    if ( $str =~ m/((.*?)(\s\-\s)?(.*?))\3/ ) {
        my $matched_str = scalar reverse $1;
        $matched_str =~ s/^(-\s?)?//;
        print $matched_str, $/;
    }
}

__DATA__
Part Time - Weekend Receptionist - Barton le Clay - Barton-le-Clay (ne
+ar Luton)
Part-Time Weekend Day Veterinary Nurse - Milton Keynes
Receptionist - Barton-le-Clay (near Luton)
Full Time - Veterinary Nurse - Glasgow Referral - Glasgow
Full Time - Professional House wife - Umuezebi Family House - New Have
+n-Upper-Chime-Enugu ( near 9th Mile )
Full time liar - Glasgow (Australia)

Output:
Barton le Clay - Barton-le-Clay (near Luton)
Milton Keynes
Barton-le-Clay (near Luton)
Glasgow Referral - Glasgow
Umuezebi Family House - New Haven-Upper-Chime-Enugu ( near 9th Mile )
Glasgow (Australia)