X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=src%2FProtocol%2FFeed.php;h=35e2cd5778487d8611de09b4f389da0e86d1898d;hb=cc8491223ec17bd12b7f37d5b2cc16b58184291e;hp=65f2e7b863715ce06c5582e9dca41a74481167c7;hpb=1fd18496416b94a8b4477955e380b19eaedbb05a;p=friendica.git diff --git a/src/Protocol/Feed.php b/src/Protocol/Feed.php index 65f2e7b863..35e2cd5778 100644 --- a/src/Protocol/Feed.php +++ b/src/Protocol/Feed.php @@ -1,9 +1,24 @@ . * */ + namespace Friendica\Protocol; use DOMDocument; @@ -11,41 +26,41 @@ use DOMXPath; use Friendica\Content\Text\HTML; use Friendica\Core\Logger; use Friendica\Core\Protocol; -use Friendica\Core\System; use Friendica\Database\DBA; +use Friendica\DI; use Friendica\Model\Item; +use Friendica\Model\Tag; use Friendica\Util\Network; +use Friendica\Util\ParseUrl; use Friendica\Util\XML; /** - * @brief This class contain functions to import feeds - * + * This class contain functions to import feeds (RSS/RDF/Atom) */ class Feed { /** - * @brief Read a RSS/RDF/Atom feed and create an item entry for it + * Read a RSS/RDF/Atom feed and create an item entry for it * * @param string $xml The feed data * @param array $importer The user record of the importer * @param array $contact The contact record of the feed - * @param string $hub Unused dummy value for compatibility reasons - * @param bool $simulate If enabled, no data is imported * - * @return array In simulation mode it returns the header and the first item + * @return array Returns the header and the first item in dry run mode * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ - public static function import($xml, $importer, &$contact, &$hub, $simulate = false) { - - $a = \get_app(); + public static function import($xml, array $importer = [], array $contact = []) + { + $dryRun = empty($importer) && empty($contact); - if (!$simulate) { - Logger::log("Import Atom/RSS feed '".$contact["name"]."' (Contact ".$contact["id"].") for user ".$importer["uid"], Logger::DEBUG); + if ($dryRun) { + Logger::info("Test Atom/RSS feed"); } else { - Logger::log("Test Atom/RSS feed", Logger::DEBUG); + Logger::info("Import Atom/RSS feed '" . $contact["name"] . "' (Contact " . $contact["id"] . ") for user " . $importer["uid"]); } + if (empty($xml)) { - Logger::log('XML is empty.', Logger::DEBUG); - return; + Logger::info('XML is empty.'); + return []; } if (!empty($contact['poll'])) { @@ -59,13 +74,13 @@ class Feed { $doc = new DOMDocument(); @$doc->loadXML(trim($xml)); $xpath = new DOMXPath($doc); - $xpath->registerNamespace('atom', NAMESPACE_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', NAMESPACE_POCO); + $xpath->registerNamespace('poco', ActivityNamespace::POCO); $author = []; $entries = null; @@ -113,14 +128,17 @@ class Feed { if (empty($author["author-name"])) { $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:subtitle/text()'); } + if (empty($author["author-name"])) { $author["author-name"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:author/atom:name/text()'); } + $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:displayName/text()'); if ($value != "") { $author["author-name"] = $value; } - if ($simulate) { + + if ($dryRun) { $author["author-id"] = XML::getFirstNodeValue($xpath, '/atom:feed/atom:author/atom:id/text()'); // See https://tools.ietf.org/html/rfc4287#section-3.2.2 @@ -133,14 +151,17 @@ class Feed { if ($value != "") { $author["author-nick"] = $value; } + $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:address/poco:formatted/text()'); if ($value != "") { $author["author-location"] = $value; } + $value = XML::getFirstNodeValue($xpath, 'atom:author/poco:note/text()'); if ($value != "") { $author["author-about"] = $value; } + $avatar = XML::getFirstAttributes($xpath, "atom:author/atom:link[@rel='avatar']"); if (is_object($avatar)) { foreach ($avatar AS $attribute) { @@ -168,9 +189,11 @@ class Feed { if (empty($author["author-name"])) { $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/copyright/text()'); } + if (empty($author["author-name"])) { $author["author-name"] = XML::getFirstNodeValue($xpath, '/rss/channel/description/text()'); } + $author["edited"] = $author["created"] = XML::getFirstNodeValue($xpath, '/rss/channel/pubDate/text()'); $author["app"] = XML::getFirstNodeValue($xpath, '/rss/channel/generator/text()'); @@ -178,12 +201,13 @@ class Feed { $entries = $xpath->query('/rss/channel/item'); } - if (!$simulate) { + if (!$dryRun) { $author["author-link"] = $contact["url"]; if (empty($author["author-name"])) { $author["author-name"] = $contact["name"]; } + $author["author-avatar"] = $contact["thumb"]; $author["owner-link"] = $contact["url"]; @@ -192,25 +216,33 @@ class Feed { } $header = []; - $header["uid"] = $importer["uid"]; + $header["uid"] = $importer["uid"] ?? 0; $header["network"] = Protocol::FEED; $header["wall"] = 0; $header["origin"] = 0; $header["gravity"] = GRAVITY_PARENT; - $header["private"] = 2; - $header["verb"] = ACTIVITY_POST; - $header["object-type"] = ACTIVITY_OBJ_NOTE; + $header["private"] = Item::PUBLIC; + $header["verb"] = Activity::POST; + $header["object-type"] = Activity\ObjectType::NOTE; - $header["contact-id"] = $contact["id"]; + $header["contact-id"] = $contact["id"] ?? 0; if (!is_object($entries)) { - Logger::log("There are no entries in this feed.", Logger::DEBUG); - return; + Logger::info("There are no entries in this feed."); + return []; } $items = []; + + // Limit the number of items that are about to be fetched + $total_items = ($entries->length - 1); + $max_items = DI::config()->get('system', 'max_feed_items'); + if (($max_items > 0) && ($total_items > $max_items)) { + $total_items = $max_items; + } + // Importing older entries first - for($i = $entries->length - 1; $i >= 0;--$i) { + for ($i = $total_items; $i >= 0; --$i) { $entry = $entries->item($i); $item = array_merge($header, $author); @@ -226,9 +258,11 @@ class Feed { } } } + if (empty($item["plink"])) { $item["plink"] = XML::getFirstNodeValue($xpath, 'link/text()', $entry); } + if (empty($item["plink"])) { $item["plink"] = XML::getFirstNodeValue($xpath, 'rss:link/text()', $entry); } @@ -238,6 +272,7 @@ class Feed { if (empty($item["uri"])) { $item["uri"] = XML::getFirstNodeValue($xpath, 'guid/text()', $entry); } + if (empty($item["uri"])) { $item["uri"] = $item["plink"]; } @@ -248,12 +283,12 @@ class Feed { $item["parent-uri"] = $item["uri"]; - if (!$simulate) { + if (!$dryRun) { $condition = ["`uid` = ? AND `uri` = ? AND `network` IN (?, ?)", $importer["uid"], $item["uri"], Protocol::FEED, Protocol::DFRN]; $previous = Item::selectFirst(['id'], $condition); if (DBA::isResult($previous)) { - Logger::log("Item with uri ".$item["uri"]." for user ".$importer["uid"]." already existed under id ".$previous["id"], Logger::DEBUG); + Logger::info("Item with uri " . $item["uri"] . " for user " . $importer["uid"] . " already existed under id " . $previous["id"]); continue; } } @@ -267,16 +302,18 @@ class Feed { $item["title"] = XML::getFirstNodeValue($xpath, 'rss:title/text()', $entry); } - $item["title"] = mb_convert_encoding($item["title"], 'HTML-ENTITIES', "UTF-8"); + $item["title"] = html_entity_decode($item["title"], ENT_QUOTES, 'UTF-8'); $published = XML::getFirstNodeValue($xpath, 'atom:published/text()', $entry); if (empty($published)) { $published = XML::getFirstNodeValue($xpath, 'pubDate/text()', $entry); } + if (empty($published)) { $published = XML::getFirstNodeValue($xpath, 'dc:date/text()', $entry); } + $updated = XML::getFirstNodeValue($xpath, 'atom:updated/text()', $entry); if (empty($updated) && !empty($published)) { @@ -290,20 +327,25 @@ class Feed { if ($published != "") { $item["created"] = $published; } + if ($updated != "") { $item["edited"] = $updated; } + $creator = XML::getFirstNodeValue($xpath, 'author/text()', $entry); if (empty($creator)) { $creator = XML::getFirstNodeValue($xpath, 'atom:author/atom:name/text()', $entry); } + if (empty($creator)) { $creator = XML::getFirstNodeValue($xpath, 'dc:creator/text()', $entry); } + if ($creator != "") { $item["author-name"] = $creator; } + $creator = XML::getFirstNodeValue($xpath, 'dc:creator/text()', $entry); if ($creator != "") { @@ -331,6 +373,7 @@ class Feed { $type = $attribute->textContent; } } + if (!empty($item["attach"])) { $item["attach"] .= ','; } else { @@ -339,19 +382,13 @@ class Feed { $attachments[] = ["link" => $href, "type" => $type, "length" => $length]; - $item["attach"] .= '[attach]href="'.$href.'" length="'.$length.'" type="'.$type.'"[/attach]'; + $item["attach"] .= '[attach]href="' . $href . '" length="' . $length . '" type="' . $type . '"[/attach]'; } - $tags = ''; + $taglist = []; $categories = $xpath->query("category", $entry); foreach ($categories AS $category) { - $hashtag = $category->nodeValue; - if ($tags != '') { - $tags .= ', '; - } - - $taglink = "#[url=" . System::baseUrl() . "/search?tag=" . $hashtag . "]" . $hashtag . "[/url]"; - $tags .= $taglink; + $taglist[] = $category->nodeValue; } $body = trim(XML::getFirstNodeValue($xpath, 'atom:content/text()', $entry)); @@ -398,7 +435,7 @@ 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($item["body"])); @@ -414,43 +451,59 @@ class Feed { } if ($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"]); - $item["tag"] = add_page_keywords($item["plink"], $preview, ($contact["fetch_further_information"] == 2), $contact["ffi_keyword_blacklist"]); - $item["object-type"] = ACTIVITY_OBJ_BOOKMARK; + $item["body"] = $item["body"] . add_page_info($item["plink"], false, $preview, ($contact["fetch_further_information"] == 2), $contact["ffi_keyword_blacklist"]); + $taglist = get_page_keywords($item["plink"], $preview, ($contact["fetch_further_information"] == 2), $contact["ffi_keyword_blacklist"]); + $item["object-type"] = Activity\ObjectType::BOOKMARK; unset($item["attach"]); } else { if (!empty($summary)) { $item["body"] = '[abstract]' . HTML::toBBCode($summary, $basepath) . "[/abstract]\n" . $item["body"]; } - if ($contact["fetch_further_information"] == 3) { - if (!empty($tags)) { - $item["tag"] = $tags; - } else { - // @todo $preview is never set in this case, is it intended? - @MrPetovan 2018-02-13 - $item["tag"] = add_page_keywords($item["plink"], $preview, true, $contact["ffi_keyword_blacklist"]); + if (!empty($contact["fetch_further_information"]) && ($contact["fetch_further_information"] == 3)) { + if (empty($taglist)) { + $taglist = get_page_keywords($item["plink"], $preview, true, $contact["ffi_keyword_blacklist"]); } - $item["body"] .= "\n".$item['tag']; + $item["body"] .= "\n" . self::tagToString($taglist); + } else { + $taglist = []; } + // Add the link to the original feed entry if not present in feed if (($item['plink'] != '') && !strstr($item["body"], $item['plink'])) { - $item["body"] .= "[hr][url]".$item['plink']."[/url]"; + $item["body"] .= "[hr][url]" . $item['plink'] . "[/url]"; } } - if (!$simulate) { - Logger::log("Stored feed: ".print_r($item, true), Logger::DEBUG); + if ($dryRun) { + $items[] = $item; + break; + } else { + Logger::info("Stored feed: " . print_r($item, true)); $notify = Item::isRemoteSelf($contact, $item); // Distributed items should have a well formatted URI. // Additionally we have to avoid conflicts with identical URI between imported feeds and these items. if ($notify) { - $item['guid'] = Item::guidFromUri($orig_plink, $a->getHostName()); + $item['guid'] = Item::guidFromUri($orig_plink, DI::baseUrl()->getHostname()); unset($item['uri']); unset($item['parent-uri']); @@ -458,20 +511,41 @@ class Feed { $notify = PRIORITY_MEDIUM; } - $id = Item::insert($item, false, $notify); + $id = Item::insert($item, $notify); - Logger::log("Feed for contact ".$contact["url"]." stored under id ".$id); - } else { - $items[] = $item; - } - if ($simulate) { - break; + Logger::info("Feed for contact " . $contact["url"] . " stored under id " . $id); + + if (!empty($id) && !empty($taglist)) { + $feeditem = Item::selectFirst(['uri-id'], ['id' => $id]); + foreach ($taglist as $tag) { + Tag::store($feeditem['uri-id'], Tag::HASHTAG, $tag); + } + } } } - if ($simulate) { - return ["header" => $author, "items" => $items]; + return ["header" => $author, "items" => $items]; + } + + /** + * Convert a tag array to a tag string + * + * @param array $tags + * @return string tag string + */ + private static function tagToString(array $tags) + { + $tagstr = ''; + + foreach ($tags as $tag) { + if ($tagstr != "") { + $tagstr .= ", "; + } + + $tagstr .= "#[url=" . DI::baseUrl() . "/search?tag=" . urlencode($tag) . "]" . $tag . "[/url]"; } + + return $tagstr; } private static function titleIsBody($title, $body)