The Perl and Raku pro­gram­ming lan­guages have a com­pli­cat­ed his­to­ry togeth­er. The lat­ter was envi­sioned in the year 2000 as Perl 6, a com­plete redesign and rewrite of Perl to solve its prob­lems of dif­fi­cult main­te­nance and the bur­den of then-​13 years of back­ward com­pat­i­bil­i­ty. Unfortunately, the devel­op­ment effort towards a first major release dragged on for ten years, and some devel­op­ers began to believe the delay con­tributed to the decline of Perl’s market- and mind­share among pro­gram­ming languages.

In the inter­ven­ing years work con­tin­ued on Perl 5, and even­tu­al­ly, Perl 6 was posi­tioned as a sis­ter lan­guage, part of the Perl fam­i­ly, not intend­ed as a replace­ment for Perl.” Two years ago it was renamed Raku to bet­ter indi­cate it as a dif­fer­ent project.

Although the two lan­guages aren’t source-​compatible, the Inline::Perl5 mod­ule does enable Raku devel­op­ers to run Perl code and use Perl mod­ules with­in Raku, You can even sub­class Perl class­es in Raku and call Raku meth­ods from Perl code. I hadn’t real­ized until recent­ly that the Perl sup­port was so strong in Raku despite them being so dif­fer­ent, and so I thought I’d take the oppor­tu­ni­ty to write some sam­ple code in both lan­guages to bet­ter under­stand the Raku way of doing things.

Rather than a sim­ple Hello World” pro­gram, I decid­ed to write a sim­ple syn­di­cat­ed news read­er. The Raku mod­ules direc­to­ry didn’t appear to have any­thing com­pa­ra­ble to Perl’s WWW::Mechanize and XML::RSS mod­ules, so this seemed like a great way to test Perl-​Raku interoperability.

Perl Feed Finder

First, the Perl script. I want­ed it smart enough to either direct­ly fetch a news feed or find it on a site’s HTML page.

#!/usr/bin/env perl

use v5.24;    # for strict, say, and postfix dereferencing
use warnings;
use WWW::Mechanize;
use XML::RSS;
use List::Util 1.33 qw(first none);

my @rss_types = qw<
    application/rss+xml
    application/rdf+xml
    application/xml
    text/xml
>;

my $mech = WWW::Mechanize->new;
my $rss  = XML::RSS->new;

my $url = shift @ARGV || 'https://phoenixtrap.com';

my $response = $mech->get($url);

# If we got an HTML page, find the linked RSS feed
if ( $mech->is_html
    and my @alt_links = $mech->find_all_links( rel => 'alternate' ) )
{
    for my $rss_type (@rss_types) {
        $url = ( first { $_->attrs->{type} eq $rss_type } @alt_links )->url
            and last;
    }
    $response = $mech->get($url);
}

die "$url does not have an RSS feed\n"
    if none { $_ eq $response->content_type } @rss_types;

binmode STDOUT, ':encoding(UTF-8)';    # avoid wide character warnings

my @items = $rss->parse( $mech->content )->{items}->@*;
say join "\t", $_->@{qw<title link>} for @items;

In the begin­ning, you’ll notice there’s a bit of boil­er­plate: use v5.24 (released in 2016) to enable restrict­ing unsafe code, the say func­tion, and post­fix deref­er­enc­ing to reduce the noise from nest­ed curly braces. I’m also bring­ing in the first and none list pro­cess­ing func­tions from List::Util as well as the WWW::Mechanize web page retriev­er and pars­er, and the XML::RSS feed parser.

Next is an array of pos­si­ble media (for­mer­ly MIME) types used to serve the RSS news feed for­mat on the web. Like Perl and Raku, RSS for­mats have a long and some­times con­tentious his­to­ry, so a news­read­er needs to sup­port sev­er­al dif­fer­ent ways of iden­ti­fy­ing them on a page.

The pro­gram then cre­ates new WWW::Mechanize (called a mech for short) and XML::RSS objects for use lat­er and gets a URL to browse from its command-​line argu­ment, default­ing to my blog if it has none. (My site, my rules, right?) It then retrieves that URL from the web. If mech believes that the URL con­tains an HTML page and can find link tags with rel="alternate" attrib­ut­es pos­si­bly iden­ti­fy­ing any news feeds, it then goes on to check the media types of those links against the ear­li­er list of RSS types and retrieves the first one it finds.

Next comes the only error check­ing done by this script: check­ing if the retrieved feed’s media type actu­al­ly match­es the list defined ear­li­er. This pre­vents the RSS pars­er from attempt­ing to process plain web pages. This isn’t a large and com­pli­cat­ed pro­gram, so the die func­tion is called with a trail­ing new­line char­ac­ter (\n) to sup­press report­ing the line on which the error occurred.

Finally, it’s time to out­put the head­lines and links, but before that hap­pens Perl has to be told that they may con­tain so-​called wide char­ac­ters” found in the Unicode stan­dard but not in the plain ASCII that it nor­mal­ly uses. This includes things like the typo­graph­i­cal curly quotes’ that I some­times use in my titles. The last two lines of the script loop through the parsed items in the feed, extract­ing their titles and links and print­ing them out with a tab (\t) sep­a­ra­tor between them:

Output from feed_finder.pl

Raku Feed Finder

Programming is often just stitch­ing libraries and APIs togeth­er, so it shouldn’t have been sur­pris­ing that the Raku ver­sion of the above would be so sim­i­lar. There are some sig­nif­i­cant (and some­times wel­come) dif­fer­ences, though, which I’ll go over now:

#!/usr/bin/env raku

use WWW::Mechanize:from<Perl5>;
use XML::RSS:from<Perl5>;

my @rss_types = qw<
  application/rss+xml
  application/rdf+xml
  application/xml
  text/xml
>;

my $mech = WWW::Mechanize.new;
my $rss  = XML::RSS.new;

sub MAIN($url = 'https://phoenixtrap.com') {
    my $response = $mech.get($url);

    # If we got an HTML page, find the linked RSS feed        
    if $mech.is_html {
        my @alt_links = $mech.find_all_links( Scalar, rel => 'alternate' );
        $response = $mech.get(
            @alt_links.first( *.attrs<type> (elem) @rss_types ).url
        );
    }

    if $response.content_type(Scalar) !(elem) @rss_types {
        # Overriding Raku's `die` stack trace is more verbose than we need
        note $mech.uri ~ ' does not have an RSS feed';
        exit 1;
    }

    my @items = $rss.parse( $mech.content ).<items>;
    put join "\t", $_<title link> for @items;
}

The first thing to notice is there’s a bit less boil­er­plate code at the begin­ning. Raku is a younger lan­guage and doesn’t have to add instruc­tions to enable less backward-​compatible fea­tures. It’s also a larg­er lan­guage with func­tions and meth­ods built-​in that Perl needs to load from mod­ules, though this feed find­er pro­gram still needs to bring in WWW::Mechanize and XML::RSS with anno­ta­tions to indi­cate they’re com­ing from the Perl5 side of the fence.

I decid­ed to wrap the major­i­ty of the pro­gram in a MAIN func­tion, which hand­i­ly gives me command-​line argu­ments as vari­ables as well as a usage mes­sage if some­one calls it with a --help option. This is a neat quality-​of-​life fea­ture for script authors that clev­er­ly reuses func­tion sig­na­tures, and I’d love to see this avail­able in Perl as an exten­sion to its sig­na­tures feature.

Raku and Perl also dif­fer in that the for­mer has a dif­fer­ent con­cept of con­text, where an expres­sion may be eval­u­at­ed dif­fer­ent­ly depend­ing upon whether its result is expect­ed to be a sin­gle val­ue (scalar) or a list of val­ues. Inline::Perl5 calls Perl func­tions in list con­text by default, but you can add the Scalar type object as a first argu­ment to force scalar con­text as I’ve done with calls to find_​all_​links (to return an array ref­er­ence) and content_​type (to return the first para­me­ter of the HTTP Content-​Type header).

Another inter­est­ing dif­fer­ence is the use of the (elem) oper­a­tor to deter­mine mem­ber­ship in a set. This is Raku’s ASCII way of spelling the ∈ sym­bol, which it can also use; !(elem) can also be spelled . Both are hard to type on my key­board so I chose the more ver­bose alter­na­tive, but if you want your code to more close­ly resem­ble math­e­mat­i­cal nota­tion it’s nice to know the option is there.

I also didn’t use Raku’s die rou­tine to exit the pro­gram with an error, main­ly because of its method of sup­press­ing the line on which the error occurred. It requires using a CATCH block and then key­ing off of the type of excep­tion thrown in order to cus­tomize its behav­ior, which seemed like overkill for such a small script. It would have looked some­thing like this:

{
    die $mech.uri ~ ' does not have an RSS feed'
        if $response.content_type(Scalar) !(elem) @rss_types;
    CATCH {
        default {
            note .message;
            exit 1;
        }
    }
}

Doubtless, this could be golfed down to reduce its ver­bosi­ty at the expense of read­abil­i­ty, but I didn’t want to resort to clever tricks when try­ing to do a one-​to-​one com­par­i­son with Perl. More expe­ri­enced Raku devel­op­ers are wel­come to set me straight in the com­ments below.

The last dif­fer­ence I’ll point out is Raku’s wel­come lack of deref­er­enc­ing oper­a­tors com­pared to Perl. This is due to the former’s con­cept of con­tain­ers, which I’m still learn­ing about. It seems to be fair­ly DWIMmy so I’m not that wor­ried, but it’s nice to know there’s an under­stand­able mech­a­nism behind it.

Overall I’m pleased with this first ven­ture into Raku and I enjoyed what I’ve learned of the lan­guage so far. It’s not as dif­fer­ent with Perl as I antic­i­pat­ed, and I can fore­see cod­ing more projects as I learn more. The com­mu­ni­ty on the #raku IRC chan­nel was also very friend­ly and help­ful, so I’ll be hang­ing out there as time permits.

What do you think? Can Perl and Raku bet­ter learn to coex­ist, or are they des­tined to be rivals? Leave a com­ment below.


Cover image: Frenemies 2” by Mauricio Delgado is licensed under CC BY-​SA 2.0.