<?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
*
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;
$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])) {
}
/**
- * 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()
* </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 [
];
if ($count > 10) {
- Logger::notice('Endless loop detected', ['url' => $url]);
+ Logger::warning('Endless loop detected', ['url' => $url]);
return $siteinfo;
}
}
$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;
$body = mb_convert_encoding($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');
case 'news_keywords':
$keywords = explode(',', $meta_tag['content']);
break;
- default:
- Logger::debug('Unsupported meta-tag found', ['meta-tag' => $meta_tag]);
}
}
$image['width'] = $photodata[0];
$image['height'] = $photodata[1];
$image['contenttype'] = $photodata['mime'];
+ $image['blurhash'] = $photodata['blurhash'] ?? null;
unset($image['url']);
ksort($image);
} else {
* Convert tags from CSV to an array
*
* @param string $string Tags
+ *
* @return array with formatted Hashtags
*/
public static function convertTagsToArray(string $string): array
$arr_tags = str_getcsv($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;
}
*
* @param string $tag The pure tag name
* @param int $k Counter for internal use
+ *
* @return void
*/
private static function arrAddHashes(string &$tag, int $k)
{
$urlarr = parse_url($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;
*
* @param array $siteinfo
* @param array $jsonld
+ *
* @return array siteinfo
*/
private static function parseParts(array $siteinfo, array $jsonld): array
*
* @param array $siteinfo
* @param array $jsonld
+ *
* @return array siteinfo
*/
private static function parseJsonLd(array $siteinfo, array $jsonld): array
*
* @param array $siteinfo
* @param array $jsonld
+ *
* @return array siteinfo
*/
private static function parseJsonLdAuthor(array $siteinfo, array $jsonld): array
*
* @param array $siteinfo
* @param array $jsonld
+ *
* @return array siteinfo
*/
private static function parseJsonLdArticle(array $siteinfo, array $jsonld): array
*
* @param array $siteinfo
* @param array $jsonld
+ *
* @return array siteinfo
*/
private static function parseJsonLdWebPage(array $siteinfo, array $jsonld): array
}
$content = JsonLD::fetchElement($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);
}
*
* @param array $siteinfo
* @param array $jsonld
+ *
* @return array siteinfo
*/
private static function parseJsonLdWebSite(array $siteinfo, array $jsonld): array
$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);
}
*
* @param array $siteinfo
* @param array $jsonld
+ *
* @return array siteinfo
*/
private static function parseJsonLdWebOrganization(array $siteinfo, array $jsonld): array
$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);
}
*
* @param array $siteinfo
* @param array $jsonld
+ *
* @return array siteinfo
*/
private static function parseJsonLdWebPerson(array $siteinfo, array $jsonld): array
$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);
}
}
$content = JsonLD::fetchElement($jsonld, 'url');
- if (!empty($content)) {
+ if (!empty($content) && is_string($content)) {
$jsonldinfo['author_url'] = trim($content);
}
*
* @param array $siteinfo
* @param array $jsonld
+ *
* @return array siteinfo
*/
private static function parseJsonLdMediaObject(array $siteinfo, array $jsonld, string $name): array
$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);
}
}
$content = JsonLD::fetchElement($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);
}
projects / friendica.git / blobdiff