regrex pattern matching

[johnhelen]

Hello

I want to parse a String with empty spaces to separate fields into a hash:

[code=perl]
use strict;
use warnings;
my $data = 'ACCESS:NO PROXYNAME:local host URL:http://localhost/ MDN:none FREE:YES PAID:NO';

my %values = split(/[: ]/, $data);
foreach my $k (keys %values)
{
print "$k: $values{$k}\n";
}
[/code]

However, the way above did not work as there is two ":" in the pattern "URL:http://localhost/".

It only works when I change it into "URL:http//localhost/" (remove one ":" character). Could you please help

Many thanks
john

[KevinADC]

I misread the question so I removed my comments

[KevinADC]

one way:

[CODE=perl]
use strict;
use warnings;
my $data = 'ACCESS:NO PROXYNAME:local host URL:http://localhost/ MDN:none FREE:YES PAID:NO';
my %values = ();
my @data = split(/\s+/,$data); # first split fields on the spaces
foreach my $var (@data) {
my($k,$v) = split(/:/,$var,2); # split each field into two string only
$values{$k} = $v;
}
foreach my $k (keys %values){
print "$k = $values{$k}\n";
}[/CODE]

You could also use a regular expression to split the string into key/value pairs and build the hash but the above should be fast and efficient as long as there are no spaces in the keys or values themselves.

[numberwhun]

one way:

[CODE=perl]
use strict;
use warnings;
my $data = 'ACCESS:NO PROXYNAME:local host URL:http://localhost/ MDN:none FREE:YES PAID:NO';
my %values = ();
my @data = split(/\s+/,$data); # first split fields on the spaces
foreach my $var (@data) {
my($k,$v) = split(/:/,$var,2); # split each field into two string only
$values{$k} = $v;
}
foreach my $k (keys %values){
print "$k = $values{$k}\n";
}[/CODE]

You could also use a regular expression to split the string into key/value pairs and build the hash but the above should be fast and efficient as long as there are no spaces in the keys or values themselves.

Its funny, I was wondering how to exclude the second :, but it looks like split will do that since there are only two variables to populate. Very interesting. Thanks for the lesson Kevin!

Jeff

[KevinADC]

It can be done in "one shot" but I think this is probably less efficient but without benchmarkling I am not sure:

[CODE=perl]
my $data = 'ACCESS:NO PROXYNAME:local host URL:http://localhost/ MDN:none FREE:YES PAID:NO';
my %values = map{ $_->[0],$_->[1] } map {[split(/:/,$_,2)]} split(/\s+/,$data);
foreach my $k (keys %values){
print "$k = $values{$k}\n";
}[/CODE]

it certainly is less readable/understandable for the casual perl coder.

[johnhelen]

thank you very much, Kevin ...