X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=src%2FUtil%2FParseUrl.php;h=79f427a654df4b5f672bc52f3470ebe01deb2b5b;hb=e1863951986ba5be173758324a00652bc5af870c;hp=0183d6b14ae5be6785f90eaf5c16af01e6974de9;hpb=46fdd9893cb0faef6982b113c85507297fac86d5;p=friendica.git

diff --git a/src/Util/ParseUrl.php b/src/Util/ParseUrl.php
index 0183d6b14a..79f427a654 100644
--- a/src/Util/ParseUrl.php
+++ b/src/Util/ParseUrl.php
@@ -1,6 +1,6 @@
 <?php
 /**
- * @copyright Copyright (C) 2010-2022, the Friendica project
+ * @copyright Copyright (C) 2010-2023, the Friendica project
  *
  * @license GNU AGPL version 3 or any later version
  *
@@ -24,6 +24,8 @@ namespace Friendica\Util;
 use DOMDocument;
 use DOMXPath;
 use Friendica\Content\OEmbed;
+use Friendica\Content\Text\HTML;
+use Friendica\Protocol\HTTP\MediaType;
 use Friendica\Core\Hook;
 use Friendica\Core\Logger;
 use Friendica\Database\Database;
@@ -68,7 +70,12 @@ class ParseUrl
 			$options = [];
 		}
 
-		$curlResult = DI::httpClient()->head($url, array_merge([HttpClientOptions::ACCEPT_CONTENT => $accept], $options));
+		try {
+			$curlResult = DI::httpClient()->head($url, array_merge([HttpClientOptions::ACCEPT_CONTENT => $accept], $options));
+		} catch (\Exception $e) {
+			DI::logger()->debug('Got exception', ['url' => $url, 'message' => $e->getMessage()]);
+			return [];
+		}
 
 		// Workaround for systems that can't handle a HEAD request. Don't retry on timeouts.
 		if (!$curlResult->isSuccess() && ($curlResult->getReturnCode() >= 400) && !in_array($curlResult->getReturnCode(), [408, 504])) {
@@ -89,7 +96,7 @@ class ParseUrl
 	}
 
 	/**
-	 * Search for chached embeddable data of an url otherwise fetch it
+	 * Search for cached embeddable data of an url otherwise fetch it
 	 *
 	 * @param string $url         The url of the page which should be scraped
 	 * @param bool   $do_oembed   The false option is used by the function fetch_oembed()
@@ -187,7 +194,7 @@ class ParseUrl
 	 * </body>
 	 * @endverbatim
 	 */
-	public static function getSiteinfo(string $url, bool $do_oembed = true, int $count = 1)
+	public static function getSiteinfo(string $url, bool $do_oembed = true, int $count = 1): array
 	{
 		if (empty($url)) {
 			return [
@@ -214,7 +221,7 @@ class ParseUrl
 		];
 
 		if ($count > 10) {
-			Logger::notice('Endless loop detected', ['url' => $url]);
+			Logger::warning('Endless loop detected', ['url' => $url]);
 			return $siteinfo;
 		}
 
@@ -283,25 +290,13 @@ class ParseUrl
 		}
 
 		$charset = '';
-		// Look for a charset, first in headers
-		// Expected form: Content-Type: text/html; charset=ISO-8859-4
-		if (preg_match('/charset=([a-z0-9-_.\/]+)/i', $curlResult->getContentType(), $matches)) {
-			$charset = trim(trim(trim(array_pop($matches)), ';,'));
-		}
-
-		// Then in body that gets precedence
-		// Expected forms:
-		// - <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
-		// - <meta charset="utf-8">
-		// - <meta charset=utf-8>
-		// - <meta charSet="utf-8">
-		// We escape <style> and <script> tags since they can contain irrelevant charset information
-		// (see https://github.com/friendica/friendica/issues/9251#issuecomment-698636806)
-		Strings::performWithEscapedBlocks($body, '#<(?:style|script).*?</(?:style|script)>#ism', function ($body) use (&$charset) {
-			if (preg_match('/charset=["\']?([a-z0-9-_.\/]+)/i', $body, $matches)) {
-				$charset = trim(trim(trim(array_pop($matches)), ';,'));
+		try {
+			// Look for a charset, first in headers
+			$mediaType = MediaType::fromContentType($curlResult->getContentType());
+			if (isset($mediaType->parameters['charset'])) {
+				$charset = $mediaType->parameters['charset'];
 			}
-		});
+		} catch(\InvalidArgumentException $e) {}
 
 		$siteinfo['charset'] = $charset;
 
@@ -315,9 +310,15 @@ class ParseUrl
 
 		$body = mb_convert_encoding($body, 'HTML-ENTITIES', 'UTF-8');
 
+		if (empty($body)) {
+			return $siteinfo;
+		}
+
 		$doc = new DOMDocument();
 		@$doc->loadHTML($body);
 
+		$siteinfo['charset'] = HTML::extractCharset($doc) ?? $siteinfo['charset'];
+
 		XML::deleteNode($doc, 'style');
 		XML::deleteNode($doc, 'option');
 		XML::deleteNode($doc, 'h1');
@@ -563,6 +564,7 @@ class ParseUrl
 						$image['width'] = $photodata[0];
 						$image['height'] = $photodata[1];
 						$image['contenttype'] = $photodata['mime'];
+						$image['blurhash'] = $photodata['blurhash'] ?? null;
 						unset($image['url']);
 						ksort($image);
 					} else {
@@ -641,6 +643,7 @@ class ParseUrl
 	 * Convert tags from CSV to an array
 	 *
 	 * @param string $string Tags
+	 *
 	 * @return array with formatted Hashtags
 	 */
 	public static function convertTagsToArray(string $string): array
@@ -648,7 +651,7 @@ class ParseUrl
 		$arr_tags = str_getcsv($string);
 		if (count($arr_tags)) {
 			// add the # sign to every tag
-			array_walk($arr_tags, ['self', 'arrAddHashes']);
+			array_walk($arr_tags, [self::class, 'arrAddHashes']);
 
 			return $arr_tags;
 		}
@@ -662,6 +665,7 @@ class ParseUrl
 	 *
 	 * @param string $tag The pure tag name
 	 * @param int    $k   Counter for internal use
+	 *
 	 * @return void
 	 */
 	private static function arrAddHashes(string &$tag, int $k)
@@ -687,7 +691,7 @@ class ParseUrl
 	{
 		$urlarr = parse_url($url);
 
-		// If the url does allready have an scheme
+		// If the url does already have an scheme
 		// we can stop the process here
 		if (isset($urlarr['scheme'])) {
 			return $url;
@@ -725,6 +729,7 @@ class ParseUrl
 	 *
 	 * @param array $siteinfo
 	 * @param array $jsonld
+	 *
 	 * @return array siteinfo
 	 */
 	private static function parseParts(array $siteinfo, array $jsonld): array
@@ -770,6 +775,7 @@ class ParseUrl
 	 *
 	 * @param array $siteinfo
 	 * @param array $jsonld
+	 *
 	 * @return array siteinfo
 	 */
 	private static function parseJsonLd(array $siteinfo, array $jsonld): array
@@ -863,6 +869,7 @@ class ParseUrl
 	 *
 	 * @param array $siteinfo
 	 * @param array $jsonld
+	 *
 	 * @return array siteinfo
 	 */
 	private static function parseJsonLdAuthor(array $siteinfo, array $jsonld): array
@@ -947,6 +954,7 @@ class ParseUrl
 	 *
 	 * @param array $siteinfo
 	 * @param array $jsonld
+	 *
 	 * @return array siteinfo
 	 */
 	private static function parseJsonLdArticle(array $siteinfo, array $jsonld): array
@@ -1017,6 +1025,7 @@ class ParseUrl
 	 *
 	 * @param array $siteinfo
 	 * @param array $jsonld
+	 *
 	 * @return array siteinfo
 	 */
 	private static function parseJsonLdWebPage(array $siteinfo, array $jsonld): array
@@ -1029,17 +1038,17 @@ class ParseUrl
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'description');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['text'] = trim($content);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'image');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['image'] = trim($content);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['image'] = trim($content);
 		}
 
@@ -1056,6 +1065,7 @@ class ParseUrl
 	 *
 	 * @param array $siteinfo
 	 * @param array $jsonld
+	 *
 	 * @return array siteinfo
 	 */
 	private static function parseJsonLdWebSite(array $siteinfo, array $jsonld): array
@@ -1063,22 +1073,22 @@ class ParseUrl
 		$jsonldinfo = [];
 
 		$content = JsonLD::fetchElement($jsonld, 'name');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['publisher_name'] = trim($content);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'description');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['publisher_description'] = trim($content);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'url');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['publisher_url'] = trim($content);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['image'] = trim($content);
 		}
 
@@ -1094,6 +1104,7 @@ class ParseUrl
 	 *
 	 * @param array $siteinfo
 	 * @param array $jsonld
+	 *
 	 * @return array siteinfo
 	 */
 	private static function parseJsonLdWebOrganization(array $siteinfo, array $jsonld): array
@@ -1101,32 +1112,34 @@ class ParseUrl
 		$jsonldinfo = [];
 
 		$content = JsonLD::fetchElement($jsonld, 'name');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['publisher_name'] = trim($content);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'description');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['publisher_description'] = trim($content);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'url');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['publisher_url'] = trim($content);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'logo', 'url', '@type', 'ImageObject');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['publisher_img'] = trim($content);
+		} elseif (!empty($content) && is_array($content)) {
+			$jsonldinfo['publisher_img'] = trim($content[0]);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'brand', 'name', '@type', 'Organization');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['publisher_name'] = trim($content);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'brand', 'url', '@type', 'Organization');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['publisher_url'] = trim($content);
 		}
 
@@ -1140,6 +1153,7 @@ class ParseUrl
 	 *
 	 * @param array $siteinfo
 	 * @param array $jsonld
+	 *
 	 * @return array siteinfo
 	 */
 	private static function parseJsonLdWebPerson(array $siteinfo, array $jsonld): array
@@ -1147,12 +1161,12 @@ class ParseUrl
 		$jsonldinfo = [];
 
 		$content = JsonLD::fetchElement($jsonld, 'name');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['author_name'] = trim($content);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'description');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['author_description'] = trim($content);
 		}
 
@@ -1162,7 +1176,7 @@ class ParseUrl
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'url');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['author_url'] = trim($content);
 		}
 
@@ -1185,6 +1199,7 @@ class ParseUrl
 	 *
 	 * @param array $siteinfo
 	 * @param array $jsonld
+	 *
 	 * @return array siteinfo
 	 */
 	private static function parseJsonLdMediaObject(array $siteinfo, array $jsonld, string $name): array
@@ -1192,22 +1207,22 @@ class ParseUrl
 		$media = [];
 
 		$content = JsonLD::fetchElement($jsonld, 'caption');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$media['caption'] = trim($content);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'url');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$media['url'] = trim($content);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'mainEntityOfPage');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$media['main'] = Strings::compareLink($content, $siteinfo['url']);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'description');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$media['description'] = trim($content);
 		}
 
@@ -1217,27 +1232,27 @@ class ParseUrl
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'contentUrl');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$media['content'] = trim($content);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'embedUrl');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$media['embed'] = trim($content);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'height');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$media['height'] = trim($content);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'width');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$media['width'] = trim($content);
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'image');
-		if (!empty($content)) {
+		if (!empty($content) && is_string($content)) {
 			$media['image'] = trim($content);
 		}