Last active
April 15, 2018 18:38
-
-
Save exileed/5f817c2cebe8e46d691778d2419c1b7a to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| <?php | |
| $sitemap = "sitemap.xml"; | |
| $startUrl = "http://php.net"; | |
| $extencions = [ | |
| ".html", | |
| ".php", | |
| "/", | |
| ]; | |
| $scanned = []; | |
| function checkUrl($url) | |
| { | |
| global $startUrl, $extencions, $scanned, $pf; | |
| if (in_array($url, $scanned) == true) { | |
| return; | |
| } | |
| array_push($scanned, $url); | |
| $html = getUrl($url); | |
| preg_match_all('/href="([^"]+)"/', $html, $links); | |
| foreach ($links[ 1 ] as $key => $link) { | |
| $baseUrl = preg_split('/[#|?]/', $link); | |
| $nextUrl = $baseUrl[ 0 ]; | |
| if ((substr($nextUrl, 0, 7) == "mailto:") || (substr($nextUrl, 0, 6) == "ftp://")) { | |
| return; | |
| } elseif ((substr($nextUrl, 0, 7) != "http://") && (substr($nextUrl, 0, 8) != "https://")) { | |
| $nextUrl = @filterUrls($nextUrl, $startUrl); | |
| } else { | |
| $nextUrl = @checkHost($nextUrl, $startUrl); | |
| } | |
| if (substr($nextUrl, 0, strlen($startUrl)) == $startUrl) { | |
| $ignore = false; | |
| if (in_array($nextUrl, $scanned)) { | |
| $ignore = true; | |
| } | |
| if ( ! $ignore) { | |
| foreach ($extencions as $ext) { | |
| if (strpos($nextUrl, $ext) != false) { | |
| fwrite($pf, " <url>\n" . | |
| " <loc>" . $nextUrl . "</loc>\n" . | |
| " </url>\n"); | |
| checkUrl($nextUrl); | |
| } | |
| } | |
| } | |
| } | |
| } | |
| } | |
| function getUrl($url) | |
| { | |
| $agent = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/60.0.3112.113 Safari/537.36'; | |
| $ch = curl_init(); | |
| curl_setopt($ch, CURLOPT_AUTOREFERER, true); | |
| curl_setopt($ch, CURLOPT_URL, $url); | |
| curl_setopt($ch, CURLOPT_USERAGENT, $agent); | |
| curl_setopt($ch, CURLOPT_VERBOSE, 1); | |
| curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, false); | |
| curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, false); | |
| curl_setopt($ch, CURLOPT_HEADER, 0); | |
| curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1); | |
| curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1); | |
| curl_setopt($ch, CURLOPT_CONNECTTIMEOUT, 5); | |
| $data = curl_exec($ch); | |
| curl_close($ch); | |
| return $data; | |
| } | |
| function checkHost($currentUrl, $startUrl) | |
| { | |
| $parsedUrl = parse_url($startUrl); | |
| $host = $parsedUrl[ 'host' ]; | |
| if (parse_url($currentUrl, PHP_URL_HOST) != $host) { | |
| return; | |
| } | |
| } | |
| function filterUrls($currentUrl, $startUrl) | |
| { | |
| $parsedUrl = parse_url($startUrl); | |
| $scheme = $parsedUrl[ 'scheme' ]; | |
| $host = $parsedUrl[ 'host' ]; | |
| if (strpos($currentUrl, "/") == 0) { | |
| return $scheme . '://' . $host . $currentUrl; | |
| } | |
| if ((parse_url($currentUrl, PHP_URL_SCHEME) == null) || ! strpos($currentUrl, "/")) { | |
| return $scheme . '://' . $host . '/' . $currentUrl; | |
| } | |
| } | |
| $pf = fopen($sitemap, "w"); | |
| fwrite($pf, "<urlset xmlns=\"http://www.sitemaps.org/schemas/sitemap/0.9\"\n" . | |
| "xmlns:xsi=\"http://www.w3.org/2001/XMLSchema-instance\"\n" . | |
| "xsi:schemaLocation=\"http://www.sitemaps.org/schemas/sitemap/0.9\n" . | |
| "http://www.sitemaps.org/schemas/sitemap/0.9/sitemap.xsd\">\n" . | |
| " <url>\n" . | |
| " <loc>" . $startUrl . "</loc>\n" . | |
| " </url>\n"); | |
| checkUrl($startUrl); | |
| fwrite($pf, "</urlset>\n"); | |
| fclose($pf); |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment