#!/usr/bin/perl -w

use lib qw(/home/blech/perllib/perl5
           /home/blech/perllib/share/perl/5.8.4
           /home/blech/perllib/lib/perl/5.8.4
           /home/blech/perllib/lib/home/blech/perllib/);
				 
use strict;
use CGI qw/:standard/;
use File::Spec;
use File::Find;
use File::Grep qw(fgrep fmap fdo);

use Time::Local;
use Data::Dumper;

use XML::XPath;
use XML::Simple;
use XML::LibXML;
use XML::LibXSLT;

# set up globals for nav stuff

my ($dates, $seen);

# decide mode and do despatch

my ($mode, $arg) = get_mode();
# print STDERR "to do '$mode' with '$arg'\n";

# um, why isn't this in (a renamed) get_mode ?
if ($mode eq 'index') {
	do_index($arg);
} elsif ($mode eq 'file') {
	do_file($arg);
} elsif ($mode eq 'month') {
	do_month($arg);
} elsif ($mode eq 'rss') {
	do_rss($arg);
} elsif ($mode eq 'search') {
	do_search($arg);
} else {
	error("Mode or argument not recognised.\n");
}

### subs

sub wanted {
	return if ($File::Find::dir =~ m/.AppleDouble/);
	return unless $_ =~ m/.xml$/;
	# return if ($_ =~ m/index.html/);
	
	my ($n, $y, $m, $d) = File::Spec->splitdir($File::Find::name);

	$dates->{'year'}{'name'} = $y;

	if (!exists $seen->{$y.$m}) {
		push (@{ $dates->{'year'}{'month'} }, { 'href' => "/bots/$y/$m/",
												'content' => "$y-$m",
											  });
	 	$seen->{$y.$m} = 'yes';
	}

	#$dates->{'year'}{'month'}{'href'} = "/bots/$y/$m/";
	#$dates->{'year'}{'month'}{'content'} = "$y-$m";

	$d =~ s/\.xml//;

	push @{ $dates->{'recent'} }, { 'content' => "$y-$m-$d",
									'href'    => "/bots/$y/$m/$d/" };
  my $offset = (scalar @{ $dates->{'recent'} })-6; 
	splice (@{ $dates->{'recent'} }, 0, $offset);
#	$dates->{$y}{$m}{$d} = $_;
}

sub get_nav {
	# this will need fixing at year end

	find(\&wanted, 'botd/2002', 'botd/2003');
	
	my $nav = XMLout($dates, rootname => 'nav');
}

sub get_root {
	return "/home/blech/web/husk.org/botd/";
}

sub get_mode {
	my $path_info = path_info;
	
	if (!$path_info || $path_info eq '/') {
		return ('index', 'index.xml');
	}
	
	if ($path_info =~ m/rss/) {
		return ('rss', 'index.xml');
	}
	
	if ($path_info =~ m/\.html?$/) {
		return ('file', $path_info);
	}
	
	if ($path_info =~ m!^[\d/]+$!) {
		# some kind of date
		my ($null, $year, $mon, $day) = split (/\//, $path_info);
		if ($day) {
			# three args, day
			my $file = sprintf("%04d/%02d/%02d/%04d-%02d-%02d.xml", 
								$year, $mon, $day, $year, $mon, $day);
			
			return ('index', $file);
		} elsif ($mon) {
			# two args, month
			my $month = sprintf("%04d/%02d", $year, $mon);
			return ('month', $month);
		}
	}
	
	if ($path_info =~ m!^/search/!) {
		(my $search = $path_info) =~ s!^/search/!!;
		$search =~ s/ .*$//;
		return ('search', $search) if $search;
	}

	my $q = new CGI;
	my $search = $q->param('search');
	$search =~ s/ .*$//;
	return ('search', $search) if $search;

	return ('error', 'whoopsie'); # error is handled back up there ^^
}

sub error {
	print CGI::header();
	print shift;
	exit;
}

sub do_index {
	my $date_file = shift;
	my $root = get_root();
	
	my $xml = $root.$date_file;
	my $xsl = $root.'scriborg.xsl';

	my $nav = get_nav();
	
	# read in data XML
	
	my $blog;
	open (SOURCE, $xml) or error("Can't open XML: $!\n");
	while (<SOURCE>) {
		s!(<relative-uri-stub value="\d+/\d+/\d+/)[\d-]+"/>!$1"/>!;
		$blog .= $_;
	}
	
	$blog =~ s!</churn>!$nav\n</churn>!;

	render ($blog, $xsl, 'html');
}

sub do_file {
	my $file = shift;
	my $nav  = get_nav();
	my $root = get_root();
	
	local $/ = undef;
	
	open (FILE, $root.'files'.$file) or error ("Can't open file '$file'.\n");
	my $xml = <FILE>;
	close (FILE);
	
	$xml = qq{<page title="page"><content><![CDATA[ $xml ]]></content>$nav</page>\n};

	my $xsl = $root.'files/files.xsl';
	render ($xml, $xsl, 'html');
}

sub do_search {
	my $search = shift;
	my $nav    = get_nav();
	my $root   = get_root();

	my @files = glob "botd/20*/*/*/*.xml";
	my ($xml, $matches);

	# debug- print files
	# print STDERR join ("\n", @files);
	
	# print CGI::header('text/xml');
	my @matches = fgrep { m/$search/i } @files;

	foreach my $matchref (@matches) {
		if ($matchref->{count}) {

			(my $link = $matchref->{filename}) =~ s!botd/!/bots/!;
			$link =~ s![\d-]*.xml!!;

			# initialise XML if necessary
			$xml = xml_header("search for '$search'") unless $xml;
			
			# add link item			
			$xml .= "<link>\n<url>$link</url>\n";
			foreach my $line (keys %{ $matchref->{'matches'} }) {
				$xml .= $matchref->{'matches'}->{$line};
			}
			$xml .= "</link>\n";

			# housekeeping
			$matches++;
		}			
	}

	unless ($xml) {
		$xml = xml_no_results("search for '$search'", $nav);
	} else {
		$xml .= "$nav\n</churn>\n";
		$xml =~ s/<matches>/$matches/;
	}

	#print $xml;	
	my $xsl = $root.'files/search.xsl';
	render ($xml, $xsl, 'html');
}

sub xml_header {
	my $title = shift;
	my $time  = localtime;
	my $date = scalar localtime;
	return <<EOX;
<!DOCTYPE churn>
<churn>
<last-updated value="$time.003962">$date</last-updated>
<itemcount value="<matches>" />
<topic>$title</topic>
EOX
}

sub xml_no_results {
	my $title = shift;
	my $nav   = shift;
	my $time  = localtime;
	my $date  = scalar localtime;
	return <<EOX;
<!DOCTYPE churn>
<churn>
<last-updated value="$time">$date</last-updated>
<itemcount value="1" />
<topic>$title</topic>
<link>
<time value="$time">$date</time>
<nick>scriborg</nick>
<title>No matches found. Sorry.</title>
</link>
$nav
</churn>
EOX
}

sub do_month {
	my $date = shift;
	my ($year, $mon) = split(/\//, $date);
	
	my $nav = get_nav();
	my $root = get_root();
	
	opendir(DIR, $root.$date) or error ("Can't make month index: $date\n");
	my @days = readdir(DIR);
	closedir(DIR);

	my $days;
	foreach my $day (reverse sort @days) {
		next if ($day !~ m/^\d+$/);

		# get topic (and as a sideeffect check we have an arch)		
		my $file = sprintf("%04d/%02d/%02d/%04d-%02d-%02d.xml", 
							$year, $mon, $day, $year, $mon, $day);
		my $xp = XML::XPath->new(filename => $root.$file) or next;

	    my $topic = $xp->find('/churn/topic');
	    
	    # this bit fixes both no topic and stringifies the XPath object
		$topic = ($topic !~ m/^\s*$/) ? $topic.'' : 'no topic set';		

		# build data structure
		my $href = sprintf("/bots/%04d/%02d/%02d", $year, $mon, $day);
		push @{ $days->{days} }, {	'content' => $topic,
								 	'day'	   => $day,
								 	'href'    => $href,
								};	
	}

	# build XML from DS
	my $month = XMLout($days, rootname => 'month');
	$month =~ s!<month>!<month name="$date">!;
	$month =~ s!</month>!$nav</month>!;

	# print STDERR "$month\n";
	
	my $xsl = $root."sb_month.xsl";
	render($month, $xsl, 'html');

}

sub do_rss {
	my $date_file = shift;
	my $root = get_root();
	
	my $xml = $root.$date_file;
	my $xsl = $root.'scrirss.xsl';

	# my $nav = get_nav();
	
	# read in data XML
	
	my $blog;
	open (SOURCE, $xml) or error("Can't open XML: $!\n");
	while (<SOURCE>) {
		s!(<relative-uri-stub value="\d+/\d+/\d+/)[\d-]+"/>!$1"/>!;
		$blog .= $_;
	}
	
	# $blog =~ s!</churn>!$nav\n</churn>!;

	render ($blog, $xsl, 'xml');
}

sub do_year {
	error("Not implemented\n");
}


sub render {
	my ($xml, $xsl, $type) = @_;

	my $parser = XML::LibXML->new();
	my $xslt = XML::LibXSLT->new();

	my $source = $parser->parse_string($xml);
	my $style  = $parser->parse_file($xsl);
	
	my $stylesheet  = $xslt->parse_stylesheet($style);
	
	my $results     = $stylesheet->transform($source);

	print CGI::header() if ($type eq 'html');
	print CGI::header('text/xml') if ($type eq 'xml');
	print $stylesheet->output_string($results);
}