I haven't been doing a lot of perl lately, but a lot of my stuff at work is still perl, so every now and then I have to do some maintenance or add new features. Today I was working on new features.

One issue with perl is that there's no way to get an array of regular expression matches without building it yourself in a really ugly way. I had a situation where I was getting a '1' returned in array of match captures when doing this:

1
2
3

if (my @matches = ($url =~ $test)) {
        ...
}

If there are captures in $test, you get the list of them, but if not, you'll get a '1' in there, in some cases. Frustrating.

You can do this (Thanks Paul!):

1
2
3

if (my @matches = ($url =~ $test)) {
        @matches = () if @+ < 2;
}

This works great... but in working out what I needed, I decided I wanted to handle both normal captures and named captures. So here's what I got:

1
2
3
4
5
6
7
8
9

if ($uri =~ $test) {
        # store named matches
        %matches = map {$_ => $+{$_}} keys(%+);

        # store not-named matches by number
        for (my $i=1; $i<= $#-; $i++) {
           $matches{$i} = substr($uri, $-[$i], $+[$i] - $-[$i]);
        }
}

And that is some ugly perl. It works though!

I use Finance::Currency::Convert::XE for the old perl version of an IRC bot, and recently xe.com updated their website, which broke XE.pm. This patch fixes it.

XE.pm.diff

UPDATE: I submitted this as a bug, and it got merged into version 0.21:

bug #78433