Regex placeholders

I was recently parsing some data with perl regexes. Each match provided number of found items that I needed to extract. The problem was slightly different, but say we have following input and a variable:

id: 1   label: Button
id: 2   label: Radiobutton
id: 3   label: Checkbox

If I want to create mapping between label and the id, I can use something like this:

my %label_to_id = ();
while($items =~ /id: \s* (?<id>\d+) \s+ label: \s* (?<label>.*?)$/gmx) {
    $label_to_id{ $+{label} } = $+{id};
}

The regex is pretty straightforward. For better readability it uses x modifier that allows us to put in whitespace and even comments. The spacing then needs to be explicitly written as you can see with \s* entries.

The captures can be done with parenthesis and address them with $1 and $2 variables, but here we use named capture with (?<name> ...). All results are then available in %+ hash.

Last thing worth noting is g modifier. It tries to find all matches, so we can iterate over them with while and store every match.

When we dump the %label_to_id contents, we get following as expected:

{ Button => 1, Checkbox => 3, Radiobutton => 2 }