#!/usr/bin/perl -w 

use strict;

use URI;

use XMLRPC::Transport::HTTP;


my $server = XMLRPC::Transport::HTTP::CGI
               -> dispatch_to( 'BlogCensus::getGenerator',
              				   'BlogCensus::getLanguage',
              				   'BlogCensus::getIncomingLinkCount',
              				   'BlogCensus::getOutgoingLinkCount',
              				   'BlogCensus::isBlog',
              				   'BlogCensus::crawledSince',
               	 )
               -> handle
             ;

sub handler {

	$server->handler( @_ );
	
	}


package BlogCensus;

use SOAP::Lite;
use CrawlDB;
sub getLanguage {

	my ( $self, $url ) = @_;

	my $id = check_id( $url );
	my $lang = single_result( "select language from metadata where id = '$id'");
	die "No language information available" unless defined $lang;
	return $lang;
}

sub isBlog {
	my ( $self, $url ) = @_;
	my $id = check_id( $url );
	return 0 unless $id;
	return 0 unless single_result( "select count(*) from blogs where id = '$id'");
	return 1;
}

sub crawledSince {

	my ( $self, $stamp ) = @_;

	die "Not a valid datestamp" unless $stamp > 0;
	die "Fishy number" unless $stamp == int( $stamp );
	
	my $result = do_sql( "select url, unix_timestamp(crawled) as crawled from blogs, urls where blogs.id = urls.id and crawled > from_unixtime( $stamp ) order by crawled desc;");
	
	my @list;
	while ( my $x = $result->fetchrow_arrayref() ){
		push @list, [ $x->[0], $x->[1] ];
	}
	return \@list;
}

sub getGenerator{

	my ( $self, $url ) = @_;

	my $id = check_id( $url );
	my $gen = single_result( "select flavor from metadata where id = '$id'");
	return $gen;
}

sub getIncomingLinkCount{

	my ( $self, $url ) = @_;
	my $id = check_id( $url );
	my $count = single_result( "select count(*) from links where sink = $id" );
	return $count;
}

sub getOutgoingLinkCount{

	my ( $self, $url ) = @_;
	my $id = check_id( $url );
	my $count = single_result( "select count(*) from links where source = $id" );
	return $count;
}


sub check_id {
	my $url = shift;
 	$url = normalize( $url );
	die "Improperly formed URL" unless $url =~ /^http:/;
	my $id = get_id( $url ) || die "This site does not appear in our URL list";
	die "This site is not listed as a weblog" unless single_result("select id from blogs where id = $id"); #get_status( $id ) eq 'c';
	return $id;
}
             
             



sub normalize {
	my $url = shift;
	my $c = URI->new( $url );
	my $canon = $c->canonical();
	return $canon->as_string;
}


sub die_bad_blog {
	die "This site is not in our blog list";
}