X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;ds=sidebyside;f=src%2FUtil%2FParseUrl.php;h=abd9154da1cb83ba4adcbf3bd50ec58f00d25d6a;hb=c33611c484e2cf5a3c626566588e26a1d14e12c6;hp=683544f44860d42b7cf2c008b199835ada1a762a;hpb=2ed98bd04056a341fd71ac23ed28d5ab5dd6ac48;p=friendica.git diff --git a/src/Util/ParseUrl.php b/src/Util/ParseUrl.php index 683544f448..abd9154da1 100644 --- a/src/Util/ParseUrl.php +++ b/src/Util/ParseUrl.php @@ -1,6 +1,6 @@ head($url); + $curlResult = DI::httpClient()->head($url, [HttpClientOptions::ACCEPT_CONTENT => $accept]); + + // Workaround for systems that can't handle a HEAD request + if (!$curlResult->isSuccess() && ($curlResult->getReturnCode() == 405)) { + $curlResult = DI::httpClient()->get($url, $accept, [HttpClientOptions::CONTENT_LENGTH => 1000000]); + } + if (!$curlResult->isSuccess()) { + Logger::debug('Got HTTP Error', ['http error' => $curlResult->getReturnCode(), 'url' => $url]); return []; } - $contenttype = $curlResult->getHeader('Content-Type'); + $contenttype = $curlResult->getHeader('Content-Type')[0] ?? ''; if (empty($contenttype)) { - return []; + return ['application', 'octet-stream']; } return explode('/', current(explode(';', $contenttype))); @@ -197,7 +207,7 @@ class ParseUrl ]; if ($count > 10) { - Logger::log('Endless loop detected for ' . $url, Logger::DEBUG); + Logger::notice('Endless loop detected', ['url' => $url]); return $siteinfo; } @@ -213,26 +223,21 @@ class ParseUrl return $siteinfo; } - $curlResult = DI::httpRequest()->get($url); + $curlResult = DI::httpClient()->get($url, HttpClientAccept::HTML, [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; } $siteinfo['expires'] = DateTimeFormat::utc(self::DEFAULT_EXPIRATION_SUCCESS); - // If the file is too large then exit - if (($curlResult->getInfo()['download_content_length'] ?? 0) > 1000000) { - return $siteinfo; - } - - if ($cacheControlHeader = $curlResult->getHeader('Cache-Control')) { + if ($cacheControlHeader = $curlResult->getHeader('Cache-Control')[0] ?? '') { if (preg_match('/max-age=([0-9]+)/i', $cacheControlHeader, $matches)) { $maxAge = max(86400, (int)array_pop($matches)); $siteinfo['expires'] = DateTimeFormat::utc("now + $maxAge seconds"); } } - $header = $curlResult->getHeader(); $body = $curlResult->getBody(); if ($do_oembed) { @@ -273,7 +278,7 @@ 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', $header, $matches)) { + if (preg_match('/charset=([a-z0-9-_.\/]+)/i', $curlResult->getContentType(), $matches)) { $charset = trim(trim(trim(array_pop($matches)), ';,')); } @@ -297,7 +302,7 @@ class ParseUrl // See https://github.com/friendica/friendica/issues/5470#issuecomment-418351211 $charset = str_ireplace('latin-1', 'latin1', $charset); - Logger::log('detected charset ' . $charset, Logger::DEBUG); + Logger::info('detected charset', ['charset' => $charset]); $body = iconv($charset, 'UTF-8//TRANSLIT', $body); } @@ -349,9 +354,6 @@ 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 = []; @@ -379,23 +381,28 @@ 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; @@ -454,6 +461,12 @@ 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; @@ -462,7 +475,6 @@ class ParseUrl break; case 'twitter:image': $siteinfo['image'] = $meta_tag['content']; - $twitter_image = true; break; } } @@ -477,9 +489,21 @@ class ParseUrl } } - // Prevent to have a photo type without an image - if ($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'])) { @@ -502,6 +526,8 @@ class ParseUrl Hook::callAll('getsiteinfo', $siteinfo); + ksort($siteinfo); + return $siteinfo; } @@ -511,9 +537,9 @@ 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 ($page_url) { @@ -576,8 +602,8 @@ class ParseUrl } if (!empty($embed)) { $media['embed'] = $embed; - if (!empty($media['main'])) { - $siteinfo['embed'] = $embed; + if (empty($siteinfo['player']['embed'])) { + $siteinfo['player']['embed'] = $embed; } } if (!empty($content)) { @@ -694,7 +720,7 @@ class ParseUrl { if (!empty($jsonld['@graph']) && is_array($jsonld['@graph'])) { foreach ($jsonld['@graph'] as $part) { - if (!empty($part)) { + if (!empty($part) && is_array($part)) { $siteinfo = self::parseParts($siteinfo, $part); } } @@ -710,7 +736,7 @@ class ParseUrl } if ($numeric_keys) { foreach ($jsonld as $part) { - if (!empty($part)) { + if (!empty($part) && is_array($part)) { $siteinfo = self::parseParts($siteinfo, $part); } } @@ -777,7 +803,7 @@ class ParseUrl case 'QAPage': case 'RealEstateListing': case 'SearchResultsPage': - case 'MediaGallery': + case 'MediaGallery': case 'ImageGallery': case 'VideoGallery': case 'RadioEpisode': @@ -806,7 +832,7 @@ class ParseUrl case 'PerformingGroup': case 'DanceGroup'; case 'MusicGroup': - case 'TheaterGroup': + case 'TheaterGroup': return self::parseJsonLdWebPerson($siteinfo, $jsonld); case 'AudioObject': case 'Audio': @@ -957,6 +983,16 @@ 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]); @@ -1120,7 +1156,11 @@ 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); }