X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=src%2FUtil%2FParseUrl.php;h=04afc927ba6cec4229c8167a6bcafe66e4c317d9;hb=720a43461d67ab229de0aecfc5008f22cc4c1c54;hp=915a143a0473b87cb2f0de059ed0d84b13e2ebc2;hpb=0651ace4e4b3668a61ba076dfb7df8b93b4485e6;p=friendica.git

diff --git a/src/Util/ParseUrl.php b/src/Util/ParseUrl.php
index 915a143a04..04afc927ba 100644
--- a/src/Util/ParseUrl.php
+++ b/src/Util/ParseUrl.php
@@ -1,6 +1,6 @@
 <?php
 /**
- * @copyright Copyright (C) 2010-2021, the Friendica project
+ * @copyright Copyright (C) 2010-2022, the Friendica project
  *
  * @license GNU AGPL version 3 or any later version
  *
@@ -30,6 +30,7 @@ use Friendica\Database\Database;
 use Friendica\Database\DBA;
 use Friendica\DI;
 use Friendica\Network\HTTPException;
+use Friendica\Network\HTTPClient\Client\HttpClientOptions;
 
 /**
  * Get information about a given URL
@@ -58,7 +59,7 @@ class ParseUrl
 	 */
 	public static function getContentType(string $url)
 	{
-		$curlResult = DI::httpRequest()->head($url);
+		$curlResult = DI::httpClient()->head($url);
 		if (!$curlResult->isSuccess()) {
 			return [];
 		}
@@ -213,8 +214,9 @@ class ParseUrl
 			return $siteinfo;
 		}
 
-		$curlResult = DI::httpRequest()->get($url, ['content_length' => 1000000]);
+		$curlResult = DI::httpClient()->get($url, [HttpClientOptions::CONTENT_LENGTH => 1000000]);
 		if (!$curlResult->isSuccess() || empty($curlResult->getBody())) {
+			Logger::info('Empty body or error when fetching', ['url' => $url, 'success' => $curlResult->isSuccess(), 'code' => $curlResult->getReturnCode()]);
 			return $siteinfo;
 		}
 
@@ -343,9 +345,6 @@ class ParseUrl
 			$siteinfo['title'] = trim($list->item(0)->nodeValue);
 		}
 
-		$twitter_card = false;
-		$twitter_image = false;
-
 		$list = $xpath->query('//meta[@name]');
 		foreach ($list as $node) {
 			$meta_tag = [];
@@ -373,23 +372,28 @@ class ParseUrl
 					break;
 				case 'twitter:image':
 					$siteinfo['image'] = $meta_tag['content'];
-					$twitter_image = true;
 					break;
 				case 'twitter:image:src':
 					$siteinfo['image'] = $meta_tag['content'];
 					break;
-				case 'twitter:card':
-					// Detect photo pages
-					if ($meta_tag['content'] == 'summary_large_image') {
-						$twitter_card = true;
-					}
-					break;
 				case 'twitter:description':
 					$siteinfo['text'] = trim($meta_tag['content']);
 					break;
 				case 'twitter:title':
 					$siteinfo['title'] = trim($meta_tag['content']);
 					break;
+				case 'twitter:player':
+					$siteinfo['player']['embed'] = trim($meta_tag['content']);
+					break;
+				case 'twitter:player:stream':
+					$siteinfo['player']['stream'] = trim($meta_tag['content']);
+					break;
+				case 'twitter:player:width':
+					$siteinfo['player']['width'] = intval($meta_tag['content']);
+					break;
+				case 'twitter:player:height':
+					$siteinfo['player']['height'] = intval($meta_tag['content']);
+					break;
 				case 'dc.title':
 					$siteinfo['title'] = trim($meta_tag['content']);
 					break;
@@ -448,6 +452,12 @@ class ParseUrl
 					case 'og:site_name':
 						$siteinfo['publisher_name'] = trim($meta_tag['content']);
 						break;
+					case 'og:locale':
+						$siteinfo['language'] = trim($meta_tag['content']);
+						break;
+					case 'og:type':
+						$siteinfo['pagetype'] = trim($meta_tag['content']);
+						break;
 					case 'twitter:description':
 						$siteinfo['text'] = trim($meta_tag['content']);
 						break;
@@ -456,7 +466,6 @@ class ParseUrl
 						break;
 					case 'twitter:image':
 						$siteinfo['image'] = $meta_tag['content'];
-						$twitter_image = true;
 						break;
 				}
 			}
@@ -471,11 +480,22 @@ class ParseUrl
 			}
 		}
 
-// Currently deactivated, see https://github.com/friendica/friendica/pull/10148#issuecomment-821512658
-		// Prevent to have a photo type without an image
-//		if ($twitter_card && $twitter_image && !empty($siteinfo['image'])) {
-//			$siteinfo['type'] = 'photo';
-//		}
+		if (!empty($siteinfo['player']['stream'])) {
+			// Only add player data to media arrays if there is no duplicate
+			$content_urls = array_merge(array_column($siteinfo['audio'] ?? [], 'content'), array_column($siteinfo['video'] ?? [], 'content'));
+			if (!in_array($siteinfo['player']['stream'], $content_urls)) {
+				$contenttype = self::getContentType($siteinfo['player']['stream']);
+				if (!empty($contenttype[0]) && in_array($contenttype[0], ['audio', 'video'])) {
+					$media = ['content' => $siteinfo['player']['stream']];
+
+					if (!empty($siteinfo['player']['embed'])) {
+						$media['embed'] = $siteinfo['player']['embed'];
+					}
+
+					$siteinfo[$contenttype[0]][] = $media;
+				}
+			}
+		}
 
 		if (!empty($siteinfo['image'])) {
 			$siteinfo['images'] = $siteinfo['images'] ?? [];
@@ -497,6 +517,8 @@ class ParseUrl
 
 		Hook::callAll('getsiteinfo', $siteinfo);
 
+		ksort($siteinfo);
+
 		return $siteinfo;
 	}
 
@@ -506,9 +528,9 @@ class ParseUrl
 	 *
 	 * @param string $page_url
 	 * @param array $siteinfo
-	 * @return void
+	 * @return array
 	 */
-	private static function checkMedia(string $page_url, array $siteinfo)
+	private static function checkMedia(string $page_url, array $siteinfo) : array
 	{
 		if (!empty($siteinfo['images'])) {
 			array_walk($siteinfo['images'], function (&$image) use ($page_url) {
@@ -571,8 +593,8 @@ class ParseUrl
 					}
 					if (!empty($embed)) {
 						$media['embed'] = $embed;
-						if (!empty($media['main'])) {
-							$siteinfo['embed'] = $embed;
+						if (empty($siteinfo['player']['embed'])) {
+							$siteinfo['player']['embed'] = $embed;
 						}
 					}
 					if (!empty($content)) {
@@ -952,6 +974,16 @@ class ParseUrl
 			}
 		}
 
+		$content = JsonLD::fetchElement($jsonld, 'datePublished');
+		if (!empty($content) && is_string($content)) {
+			$jsonldinfo['published'] = DateTimeFormat::utc($content);
+		}
+
+		$content = JsonLD::fetchElement($jsonld, 'dateModified');
+		if (!empty($content) && is_string($content)) {
+			$jsonldinfo['modified'] = DateTimeFormat::utc($content);
+		}
+
 		$jsonldinfo = self::parseJsonLdAuthor($jsonldinfo, $jsonld);
 
 		Logger::info('Fetched article information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
@@ -1115,7 +1147,11 @@ class ParseUrl
 		}
 
 		$content = JsonLD::fetchElement($jsonld, 'image', 'url', '@type', 'ImageObject');
-		if (!empty($content)) {
+		if (!empty($content) && !is_string($content)) {
+			Logger::notice('Unexpected return value for the author image', ['content' => $content]);
+		}
+
+		if (!empty($content) && is_string($content)) {
 			$jsonldinfo['author_img'] = trim($content);
 		}