root/lang/perl/WWW-CommentGetter/trunk/WWW-CommentGetter/lib/WWW/CommentGetter/Plugin.pm @ 1934

package WWW::CommentGetter::Plugin;
use strict;
use warnings;
use utf8;
use Web::Scraper;

sub import {
    my $pkg = caller(0);
    my $class = shift;

    for my $name ( qw/trim new scrape_by_siteinfo/) {
        no strict 'refs'; ## no critic.
        *{"$pkg\::$name"} = *{"$class\::$name"};
    }
}

sub new { bless {}, shift }

sub trim {
    local $_ = shift;
    s/^\s*//;
    s/\s*$//;
    return $_;
}

sub scrape_by_siteinfo {
    my ($url, $siteinfo) = @_;

    return unless $url =~ $siteinfo->{url};

    my $ret = scraper {
        process $siteinfo->{context}, 'comments[]' => scraper {
            my $x = shift;
            while (my ($key, $xpath) = each %{$siteinfo->{attributes}}) {
                process $xpath, $key, sub {
                    my $element = shift;
                    trim $element->as_text;
                };
            }
        };
    }->scrape(URI->new($url));

    return [ @{ $ret->{comments} } ];
}

1;