Regular expression in Perl

[sakshismriti]

I have a regex which extratcs the name from it.

The code is given below:

# Extract some parts from regex
$input = "Hello, Shane !";
$pattern = "^Hello, \\w+\\s*!\$";
 
if($input =~ /$pattern/) {
	print "\nThe input matches the pattern The dollar variables are\n";  
	print " \nONE == $1";
	print "\n";
}

The input matches the pattern The dollar variables are  
ONE == Shane

Now, if I want to extract the name, like this :

Problem:
Now, I'd like to match the earlier, and get the extracted data ("Shane" here in our example) by a designated name, not a number set by the regex.

Eg.:

^Hello, (?{surname}\w+) \s*!$

then I'll have a way to get the extracted data into $surname or maybe there is a way to do something like this:

my $input = "Hello, Shane !";
my $regex = '^Hello, (?{surname}\w+)\s*!$'
my %match = Regex.match($regex, $input);
$match{"surname"} == "Shane"; # true

Regex.match() is the sub routine name given by me(dont confuse with any existing in built module in Perl).
It is this sub routine which has the functionality of matching the regex and returning the extracted data in the parameter surname.


Any help is highly appreciated.

Thanks.

Regards,
Sakshi

[toolic]

Update: As Sakshismriti discovered, named regex captures are now possible as of Perl version 5.10.0.


Original reply:
I think you must use the special number variables, but that's not a bad thing:

use strict;
use warnings;
use Data::Dumper;

my $input   = 'Hello, Shane !';
my $pattern = qr{^Hello, (\w+)\s*!$};
my %match;
if ($input =~ /$pattern/) {
    $match{surname} = $1;
}

print Dumper(\%match);

__END__

$VAR1 = {
          'surname' => 'Shane'
        };

[sakshismriti]

Thanks for the help.
I have also made another solution for me and it is working fine :

#!/opt/perl/bin/perl

# Test regex with surname 
my $input_string = "Hello, Shane !";
my $regex_string = "^Hello, (?{surname}\\w+)\\s*!\$";

my  %match = match($regex_string, $input_string);
if ($match{"code"} != 0) { 
	print $match{"error_msg"};
} else {
	print "\nExtracted string is: [". $match{"surname"}. "]\n";
}

# Test regex with another word called color in place of surname    
# I made another input string and regex for it 
$input_string = "The, red cherries !";
$regex_string = "^The, (?{color}\\w+)\\s*\\w+\\s*!\$";

%match =  match($regex_string, $input_string);
if ($match{"code"} != 0) { 
	print $match{"error_msg"};
} else {
	print "\nExtracted string is: [". $match{"color"}. "]\n";
}

# Test above regex non matching pattern      
# I made another input string and regex for it 
$input_string = "The red cherries !";
$regex_string = "^The, (?{color}\\w+)\\s*\\w+\\s*!\$";

%match =  match($regex_string, $input_string);
if ($match{"code"} != 0) { 
	print $match{"error_msg"};
} else {
	print "\nExtracted string is: [". $match{"color"}. "]\n";
}
 
sub match () {
	my $regex = shift();
	my $input = shift();
	
	my %return = ();
	my $param_name;
	# Extract the name of return type from regex
	if($regex =~ s/\?{(.+)}//) { 
		$param_name = $1; 
	
	} else {
		$return{"code"} = 1;
        $return{"error_msg"} = "Improper regex [$regex] given as input.";
        return %return;
	}
	 
	# Compare input with the regex
	if($input =~ /$regex/) {
		$return{"code"} = 0;
		$return{$param_name} = $1;
		return %return;
	} else {
		$return{"code"} = 1;
		$return{error_msg} = "There are no match corresponding to your search.";
		return %return;
	}
}

Regards,
Sakshi

[sakshismriti]

I have a regex which extratcs the name from it.

The code is given below:

# Extract some parts from regex
$input = "Hello, Shane !";
$pattern = "^Hello, \\w+\\s*!\$";
 
if($input =~ /$pattern/) {
	print "\nThe input matches the pattern The dollar variables are\n";  
	print " \nONE == $1";
	print "\n";
}

The input matches the pattern The dollar variables are  
ONE == Shane

Now, if I want to extract the name, like this :

Problem:
Now, I'd like to match the earlier, and get the extracted data ("Shane" here in our example) by a designated name, not a number set by the regex.

Eg.:

^Hello, (?{surname}\w+) \s*!$

then I'll have a way to get the extracted data into $surname or maybe there is a way to do something like this:

my $input = "Hello, Shane !";
my $regex = '^Hello, (?{surname}\w+)\s*!$'
my %match = Regex.match($regex, $input);
$match{"surname"} == "Shane"; # true

Regex.match() is the sub routine name given by me(dont confuse with any existing in built module in Perl).
It is this sub routine which has the functionality of matching the regex and returning the extracted data in the parameter surname.


Any help is highly appreciated.

Thanks.

Regards,
Sakshi

Hey,

I got the more appropriate solution, I have to use labelled capturing groups in perl which is new in Perl 5.0
Read it here :

http://books.google.com/books?id=lNVHi3TunxsC&lpg=PA127&ots=2ZWVY5581W&dq=labelled%20capturing%20groups%20perl&pg=PA126#v=onepage&q&f=false

The solution is as follows:

#!/usr/bin/perl -w
use strict;
use Data::Dumper; # for debug only

my $input = 'Hello, Shane !';
my $regex = qr/^Hello, (?<surname>\w+)\s*!$/;
my %match = match($regex, $input);
# print("\nthis is match" . Dumper(%match));
print ("\n The matched value ios === ". $match{"surname"});	
# $match{surname} eq 'Shane'; # true


sub match {
   $_[1] =~ $_[0];
   map { $_ => $+{$_} } keys %+;
}

Cheers!!!
Sakshi