#!/usr/local/bin/perl -w

use strict;

use WiKicker::File;
use WiKicker::URI;

use Storable qw(thaw);

my $db_home_dir   = '/home/www/var/WiKicker';
my $site          = 'http://www.somedomain.com/wiki.cgi';
my $sitemap_file  = '/home/www/public_html/sitemap_wiki.xml';

my $GZIP = '/usr/bin/gzip';

my $db_file_dir   = $db_home_dir . '/page';
my $db_cache_dir  = $db_home_dir . "/info/basic";
my $db_index_file = $db_file_dir . '/index';
my $db_touch_file = $db_home_dir . '/last-modified.txt';

open(FILE, "< $db_touch_file") || die "can't open $db_touch_file\n";
my $last_mtime = <FILE>;
close(FILE);

my $current_time = time;
my $last_mtime_string = get_gtime_string($last_mtime);

my @db_idx;
{;
 my @tmp_idx;
 open(INDEX, "< ${db_index_file}") || die "Can't open ${db_index_file}.\n";
 @tmp_idx = <INDEX>;
 close(INDEX) || die "Can't close ${db_index_file}\n";

 chomp(@tmp_idx);

 while (my $page_name = shift @tmp_idx) {
   push(@db_idx, $page_name) if ($page_name ne 'FrontPage');
 }
 @db_idx = ('FrontPage', sort @db_idx);
}
my $total = @db_idx;

my $sitemap =<<_EOS_;
<?xml version='1.0' encoding='UTF-8'?>
<urlset xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
         xsi:schemaLocation="http://www.sitemaps.org/schemas/sitemap/0.9
         http://www.sitemaps.org/schemas/sitemap/0.9/sitemap.xsd"
         xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
_EOS_

for (my $i = 0; $i < @db_idx; $i ++) {
  my $page_name = $db_idx[$i];
  my $db_path;
  my %url = qw(priority 0.5);
  my $mtime;

  my $key = WiKicker::File::key_to_file_name_base64($page_name);
  $db_path = sprintf("%s/%s", $db_file_dir, $key);
  die "Can't access to ${db_path} .\n" if (! -e $db_path);

  if ($page_name eq 'FrontPage') {
    $url{loc} = $site;
    $url{priority} += 0.2;
  } elsif ($page_name =~ /^IndexPage/) {
    ;
  } else {
    $url{loc} = sprintf("%s/%s.html", $site, WiKicker::URI::page_name_to_path_segments($page_name));
  }


  $mtime = get_last_modified_time_by_cache_file($key);
  $url{lastmod} = get_time_string($mtime);

  if ( ($current_time - $mtime) < (60 * 60 * 24 * 7) ) {
    $url{changefreq} = 'dairy';
    $url{priority} += 0.1;
  } else {
    $url{changefreq} = (($current_time - $mtime) < (60 * 60 * 24 * 30)) ? 'weekly' : 'monthly';
  }

  if ($i + 1 < @db_idx) {
    my $next_page_name = $db_idx[$i + 1];
    $url{priority} += 0.1 if ( $next_page_name =~ m#^${page_name}/# );
  }

  $sitemap .= get_url_tag(%url);
}

{;
 my %url;

 $url{lastmod} = get_time_string(int($last_mtime));
 $url{changefreq} = 'allways';
 $url{priority} = 0.4;

 for (my $i = 0; $i < $total; $i += 50) {
   $url{loc}  = "$site/IndexPage.html";
   $url{loc} .= sprintf("?offset=%d", $i) if ($i != 0);
   $sitemap .= get_url_tag(%url);
 }

# $url{loc} = "$site/RecentChanges.html";
# $url{priority} = 0.3;
# $sitemap .= get_url_tag(%url);

}

$sitemap .= "</urlset>\n";

open(FILE, "|$GZIP >${sitemap_file}.gz") || die "can't open ${sitemap_file}.\n";
print FILE $sitemap;
close(FILE) || die "can't close ${sitemap_file}\n";

print "Content-Type: text/html\n\n\n";
exit(0);

sub get_time_string {
  my $time = shift;
  # ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) = localtime($time);
  my ($sec, $min, $hour, $mday, $mon, $year) = gmtime($time);
  $year += 1900;  $mon ++;

  return sprintf("%0.4d-%0.2d-%0.2dT%0.2d:%0.2d:%0.2d+00:00",$year,$mon,$mday,$hour,$min,$sec);
}

sub get_gtime_string {
  my $time = shift;
  my($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) = gmtime($time);

  return sprintf("%s, %02d %s %d %02d:%02d:%02d GMT",
		 qw(Sun Mon Tue Wed Thu Fri Sat)[$wday],
		 $mday,
		 qw(Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec)[$mon],
		 $year + 1900,
		 $hour, $min, $sec
		);
}

sub get_last_modified_time_by_cache_file {
  my $key = shift;
  my $value;

  if (open(FILE, "$db_cache_dir/$key")) {
     binmode(FILE);
     local $/;
     $value = <FILE>;
     close(FILE);
   } else {
     die "can't read file $key.\n";
   }
   my  $record = \%{thaw($value)};

  return $record->{lastmodified};
}

sub get_url_tag {
  my %p = @_;
  my $t;

  $t  =         "  <url>\n";
  $t .= sprintf("    <loc>%s</loc>\n", $p{loc}) if (defined($p{loc}));
  $t .= sprintf("    <lastmod>%s</lastmod>\n", $p{lastmod}) if (defined($p{lastmod}));
  $t .= sprintf("    <changefreq>%s</changefreq>\n", $p{changefreq}) if (defined($p{changefreq}));
  $t .= sprintf("    <priority>%0.1f</priority>\n", $p{priority}) if (defined($p{priority}) && $p{priority} != 0.5);
  $t .=         "  </url>\n";

  return $t;
}

# Local Variables:
# mode: cperl
# coding: utf-8
# End: