Extract type from the name.

[utils.git] / anndata

#!/usr/bin/perl -w

use LWP::UserAgent;
use Getopt::Long;

sub get
{
    my($uri, $ua, $res);
    ($uri) = @_;

    $ua = LWP::UserAgent->new;
    $ua->agent("ANNData/0.1");

    $res = $ua->request(HTTP::Request->new("GET", "$uri"));

    die "could not fetch $uri\n" unless $res->is_success;
    return $res->content;
}

sub getlist
{
    my($name, $il, $html, @ret);
    ($name) = @_;
    
    $il = uc(($name =~ /^(.)/)[0]);
    $il = "9" if (!($il =~ /[A-Z]/));
    $html = get "http://www.animenewsnetwork.com/encyclopedia/anime.php?list=$il";
    
    # The only way to recognize entries that seems sure is to look
    # after the "HOVERLINE" class.
    
    push @ret, $1 while $html =~ /<A\s.*CLASS=HOVERLINE\s.*>.*<FONT.*>([^<>]*$name[^<>]*)<\/FONT/ig;
    
    return(@ret);
}

sub getid
{
    my($name, $il, $html, $url);
    ($name) = @_;
    
    $il = uc(($name =~ /^(.)/)[0]);
    $il = "9" if (!($il =~ /[A-Z]/));
    $html = get "http://www.animenewsnetwork.com/encyclopedia/anime.php?list=$il";
    
    # The only way to recognize entries that seems sure is to look
    # after the "HOVERLINE" class.
    
    (($url) = ($html =~ /<A\s.*CLASS=HOVERLINE\s.*HREF=\"([^\"]+)\".*$name/i)) || return;
    
    return((($url =~ /\?id=(\d+)$/)[0]));
}

sub getnamefromid
{
    my($id, $html);
    ($id) = @_;
    
    $html = get "http://www.animenewsnetwork.com/encyclopedia/anime.php?id=$id";
    
    return(($html =~ /\<TITLE\>Anime News Network - ([^<]*)<\/TITLE>/)[0]);
}

GetOptions(\%options, ("l"));

if($options{"l"}) {
    @list = getlist $ARGV[0];
    foreach $name (@list) {
	print "$name\n";
    }
    exit 0;
}

unless($test = getid $ARGV[0]) {
    printf STDERR "could not find $ARGV[0]\n";
    exit 1;
}

print getnamefromid($test) . "\n";