File Coverage

blib/lib/WWW/Scraper/Delicious.pm

Criterion	Covered	Total	%
statement	9	77	11.6
branch	0	58	0.0
condition	0	8	0.0
subroutine	3	10	30.0
pod	6	6	100.0
total	18	159	11.3

package WWW::Scraper::Delicious;

use strict;
use warnings;

use LWP::UserAgent;

our $VERSION = '0.10';

sub new {
    my ($class, %args) = @_;
    my $self = {};
    limit($self, $args{limit}) if $args{limit};
    ua($self, $args{ua}) if $args{ua};
    $self->{ua} = LWP::UserAgent->new() unless $self->{ua};
    bless($self, $class);
    return $self;
}

sub getlinks {
    my ($self, $path) = @_;
    return unless $path;
    my $url = ($path =~ /^htt/ ? $path
      : ( $path =~ /^\// ? "http://del.icio.us$path"
            : "http://del.icio.us/$path" ) );
    my $limit = $self->{limit} || 0;
    my %linkset = _scrape($self->{ua}, $url, $limit);
    return %linkset;
}

sub getlinksarray {
    my ($self, $path) = @_;
    my %linkset = getlinks($self, $path);
    return unless keys %linkset;
    my @table = sort { $b->[5] cmp $a->[5] }
      map { [ $linkset{$_}{id},
              $linkset{$_}{url},
              $linkset{$_}{desc},
              $linkset{$_}{notes},
              $linkset{$_}{pop},
              $linkset{$_}{date},
              $linkset{$_}{tag}, $linkset{$_}  ]
      } keys %linkset;
    return @table;
}

sub ua {
    my ($self, $ua) = @_;
    $self->{ua} = $ua if defined $ua
      && ref($ua) eq 'LWP::UserAgent';
    return $self->{ua};
}

sub limit {                     # limit of 0 is default (unlimited)
    my ($self, $limit) = @_;
    $self->{limit} = $limit if defined $limit && $limit =~ /^\d+$/;
    return $self->{limit};
}

sub dumplink {
    my ($self, $linkref) = @_;
    return unless $linkref;
    $linkref = $linkref->[7] if ref($linkref) eq 'ARRAY';
    return unless $linkref->{id};
    my $str = "   id = ".$linkref->{id}."\n";
    $str   .= "  url = ".$linkref->{url}."\n";
    $str   .= " desc = ".$linkref->{desc}."\n"  if $linkref->{desc};
    $str   .= "notes = ".$linkref->{notes}."\n" if $linkref->{notes};
    $str   .= "  pop = ".$linkref->{pop}."\n"   if $linkref->{pop};
    $str   .= " date = ".$linkref->{date}."\n"  if $linkref->{date};
    $str   .= " tags = ".join(', ', sort keys %{$linkref->{tag}})."\n"
                                                if $linkref->{tag};
    return $str;
}

sub _scrape {
    my ($ua, $url, $limit) = @_;
    my (%linkset, $page);
    my $num = 0;

    while (1) {

        my $url = "$url?setcount=100" . ($page ? "&page=$page" : '');
        my $rs = $ua->get($url);
        return unless $rs->is_success;
        my $html = $rs->content;

        my @tmp = split /
89
        for my $scrap (@tmp[1..$#tmp]) {
            $scrap =~ s/\s*<\/li>.*$//si;

            next unless (my ($id, $url, $desc) = $scrap =~
              /^(.*?)".*?a href="(.*?)".*?>(.*?)<\/a>/si) == 3;
            $linkset{$id} = { id => $id, url => $url, desc => $desc };

            my ($notes) = $scrap =~ /class="notes">(.*?)<\/p>/si;
            $linkset{$id}{notes} = $notes if $notes;

            for my $str (split /
                next unless my($tag) = $str =~ /^href=.*?>(.*?)<\/a>/;
                $linkset{$id}{tag}{$tag}++;
            }
            delete $linkset{$id}{tag} unless keys %{$linkset{$id}{tag}};

            my ($pop) = $scrap =~ /a class="pop".*?>.*?by (\d+) /si;
            $linkset{$id}{pop} = $pop if $pop;

            my ($date) = $scrap =~ / class="date" title="(.*?)"/si;
            $linkset{$id}{date} = $date if $date;

            last if ++$num == $limit;
        }

        last unless my($page0,$page1) = $html =~ / page (\d+) of (\d+)/si;
        last if $page0 == $page1;
        $page = $page ? $page + 1 : 2;
    }

    return %linkset;
}

1;
__END__