sitemap-siteindex.php

<?php
/**
  *
  * This is a quick way to turn a simple text file
  * with a very long list of urls in a text file (sitemap-urls.txt)
  * Where "very long" is an expected url number greater than 10,000
  * If loaded without a valid query parameter "page" it will load a
  * Site Index site map, otherwise load the individual XML site map
  * 10,000 urls into a valid XML Sitemap:
  * http://en.wikipedia.org/wiki/Sitemaps
  * Put this file sitemap.xml.php and sitemap-urls.txt at
  * the webroot http://example.com/sitemap.xml.php
  * Then add the text in quotes below to your robots.txt file as a new line:
  * "Sitemap: http://example.com/sitemap.xml.php"
  * 
  * Questions? email [email protected]
  * 
  * Based on https://gist.github.com/artlung/210438
  */
$per_page = 10000;
$filename = 'sitemap-urls.txt';
$urls = file($filename);
$filectime = filectime($filename);
$urls = array_map('trim', $urls);
$page = (int)$_GET['page'];
$sitemap = array();
foreach($urls as $url) {
    if ($url != '') {
        $priority = '0.5';
        $sitemap[] = array(
          'loc' => $url,
          'lastmod' => date('Y-m-d',$filectime),
          'changefreq' => 'weekly',
          'priority' => $priority,
        );
    }
}
$pages = array_chunk($sitemap, $per_page);
$page_numbers = range(1, count($pages));

header('Content-Type: text/xml');
echo '<?xml version=\'1.0\' encoding=\'UTF-8\'?>';
echo "\n";

$path = explode('?', $_SERVER['REQUEST_URI']);
$path = array_shift($path);

$url = (isset($_SERVER['HTTPS']) && $_SERVER['HTTPS'] !== 'off' ? 'https' : 'http') . '://' . $_SERVER['HTTP_HOST'] . $path . "?page=";
$lastmod = date('Y-m-d',$filectime);


if (!in_array($page, $page_numbers)) {
    // Valid Page Number
    echo '<sitemapindex xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
      xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"
      xsi:schemaLocation="http://www.sitemaps.org/schemas/sitemap/0.9 http://www.sitemaps.org/schemas/sitemap/0.9/siteindex.xsd">';
    echo "\n";
    foreach ($page_numbers as $pg_num) {
        echo "\t<sitemap>\n";
        echo "\t\t<loc>" . htmlentities($url) . $pg_num . "</loc>\n";
        echo "\t\t<lastmod>{$lastmod}</lastmod>\n";
        echo "\t</sitemap>\n";
    }
    echo '</sitemapindex>';
} else {
    // Output the Site Map
    echo '<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"
         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">';
    echo "\n";
    foreach ($pages[$page-1] as $link) {
        echo "\t<url>\n";
        echo "\t\t<loc>" . htmlentities($link['loc']) . "</loc>\n";
        echo "\t\t<lastmod>{$link['lastmod']}</lastmod>\n";
        echo "\t\t<changefreq>{$link['changefreq']}</changefreq>\n";
        echo "\t\t<priority>{$link['priority']}</priority>\n";
        echo "\t</url>\n";
    }
    echo '</urlset>';
}