]> git.mxchange.org Git - friendica.git/blobdiff - src/Protocol/Feed.php
Fix capitalization of table definition key in DBStructure
[friendica.git] / src / Protocol / Feed.php
index 816820bc4a56236613a918491f80c34d2814dd0b..705a094c38149b433b454fc2b5626181ebb9b598 100644 (file)
@@ -14,7 +14,8 @@ use Friendica\Core\Protocol;
 use Friendica\Core\System;
 use Friendica\Database\DBA;
 use Friendica\Model\Item;
-use Friendica\Protocol\Activity\ANamespace;
+use Friendica\Protocol\ActivityNamespace;
+use Friendica\Util\ParseUrl;
 use Friendica\Util\Network;
 use Friendica\Util\XML;
 
@@ -60,13 +61,13 @@ class Feed {
                $doc = new DOMDocument();
                @$doc->loadXML(trim($xml));
                $xpath = new DOMXPath($doc);
-               $xpath->registerNamespace('atom', ANamespace::ATOM1);
+               $xpath->registerNamespace('atom', ActivityNamespace::ATOM1);
                $xpath->registerNamespace('dc', "http://purl.org/dc/elements/1.1/");
                $xpath->registerNamespace('content', "http://purl.org/rss/1.0/modules/content/");
                $xpath->registerNamespace('rdf', "http://www.w3.org/1999/02/22-rdf-syntax-ns#");
                $xpath->registerNamespace('rss', "http://purl.org/rss/1.0/");
                $xpath->registerNamespace('media', "http://search.yahoo.com/mrss/");
-               $xpath->registerNamespace('poco', ANamespace::POCO);
+               $xpath->registerNamespace('poco', ActivityNamespace::POCO);
 
                $author = [];
                $entries = null;
@@ -399,7 +400,7 @@ class Feed {
 
                                // Remove a possible link to the item itself
                                $item["body"] = str_replace($item["plink"], '', $item["body"]);
-                               $item["body"] = preg_replace('/\[url\=\](\w+.*?)\[\/url\]/i', '', $item["body"]);
+                               $item["body"] = trim(preg_replace('/\[url\=\](\w+.*?)\[\/url\]/i', '', $item["body"]));
 
                                // Replace the content when the title is longer than the body
                                $replace = (strlen($item["title"]) > strlen($item["body"]));
@@ -415,8 +416,21 @@ class Feed {
                                }
 
                                if ($replace) {
-                                       $item["body"] = $item["title"];
+                                       $item["body"] = trim($item["title"]);
                                }
+
+                               $data = ParseUrl::getSiteinfoCached($item['plink'], true);
+                               if (!empty($data['text']) && !empty($data['title']) && (mb_strlen($item['body']) < mb_strlen($data['text']))) {
+                                       // When the fetched page info text is longer than the body, we do try to enhance the body
+                                       if (!empty($item['body']) && (strpos($data['title'], $item['body']) === false) && (strpos($data['text'], $item['body']) === false)) {
+                                               // The body is not part of the fetched page info title or page info text. So we add the text to the body
+                                               $item['body'] .= "\n\n" . $data['text'];
+                                       } else {
+                                               // Else we replace the body with the page info text
+                                               $item['body'] = $data['text'];
+                                       }
+                               }
+
                                // We always strip the title since it will be added in the page information
                                $item["title"] = "";
                                $item["body"] = $item["body"].add_page_info($item["plink"], false, $preview, ($contact["fetch_further_information"] == 2), $contact["ffi_keyword_blacklist"]);