2016-11-24 01:11:22 +01:00
|
|
|
<?php
|
|
|
|
/**
|
2020-02-09 16:18:46 +01:00
|
|
|
* @copyright Copyright (C) 2020, Friendica
|
|
|
|
*
|
|
|
|
* @license GNU AGPL version 3 or any later version
|
|
|
|
*
|
|
|
|
* This program is free software: you can redistribute it and/or modify
|
|
|
|
* it under the terms of the GNU Affero General Public License as
|
|
|
|
* published by the Free Software Foundation, either version 3 of the
|
|
|
|
* License, or (at your option) any later version.
|
|
|
|
*
|
|
|
|
* This program is distributed in the hope that it will be useful,
|
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
* GNU Affero General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU Affero General Public License
|
|
|
|
* along with this program. If not, see <https://www.gnu.org/licenses/>.
|
|
|
|
*
|
2016-11-24 01:11:22 +01:00
|
|
|
*/
|
2020-02-09 16:18:46 +01:00
|
|
|
|
2018-01-04 18:03:15 +01:00
|
|
|
namespace Friendica\Util;
|
2016-11-24 01:11:22 +01:00
|
|
|
|
2018-07-20 04:15:21 +02:00
|
|
|
use DOMDocument;
|
|
|
|
use DOMXPath;
|
2018-01-01 02:58:09 +01:00
|
|
|
use Friendica\Content\OEmbed;
|
2018-12-26 07:06:24 +01:00
|
|
|
use Friendica\Core\Hook;
|
2018-10-29 22:20:46 +01:00
|
|
|
use Friendica\Core\Logger;
|
2020-11-19 20:34:48 +01:00
|
|
|
use Friendica\Database\Database;
|
2018-07-20 14:19:26 +02:00
|
|
|
use Friendica\Database\DBA;
|
2020-03-04 22:35:09 +01:00
|
|
|
use Friendica\DI;
|
2021-02-16 16:16:04 +01:00
|
|
|
use Friendica\Network\HTTPException;
|
2017-05-11 17:53:04 +02:00
|
|
|
|
2016-11-24 01:11:22 +01:00
|
|
|
/**
|
2020-02-09 16:18:46 +01:00
|
|
|
* Get information about a given URL
|
|
|
|
*
|
2020-01-19 07:05:23 +01:00
|
|
|
* Class with methods for extracting certain content from an url
|
2016-11-24 01:11:22 +01:00
|
|
|
*/
|
2017-11-08 23:02:50 +01:00
|
|
|
class ParseUrl
|
|
|
|
{
|
2021-02-16 16:16:04 +01:00
|
|
|
const DEFAULT_EXPIRATION_FAILURE = 'now + 1 day';
|
|
|
|
const DEFAULT_EXPIRATION_SUCCESS = 'now + 3 months';
|
|
|
|
|
2019-11-15 14:28:42 +01:00
|
|
|
/**
|
|
|
|
* Maximum number of characters for the description
|
|
|
|
*/
|
|
|
|
const MAX_DESC_COUNT = 250;
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Minimum number of characters for the description
|
|
|
|
*/
|
|
|
|
const MIN_DESC_COUNT = 100;
|
|
|
|
|
2021-03-13 00:04:51 +01:00
|
|
|
/**
|
|
|
|
* Fetch the content type of the given url
|
|
|
|
* @param string $url URL of the page
|
|
|
|
* @return string content type
|
|
|
|
*/
|
|
|
|
public static function getContentType(string $url)
|
|
|
|
{
|
|
|
|
$curlResult = DI::httpRequest()->head($url);
|
|
|
|
if (!$curlResult->isSuccess()) {
|
|
|
|
return '';
|
|
|
|
}
|
|
|
|
|
|
|
|
$contenttype = $curlResult->getHeader('Content-Type');
|
|
|
|
if (empty($contenttype)) {
|
|
|
|
return '';
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!preg_match('#(image|video|audio)/#i', $contenttype, $matches)) {
|
|
|
|
return '';
|
|
|
|
}
|
|
|
|
|
|
|
|
return array_pop($matches);
|
|
|
|
}
|
|
|
|
|
2016-11-27 23:41:55 +01:00
|
|
|
/**
|
2020-01-19 07:05:23 +01:00
|
|
|
* Search for chached embeddable data of an url otherwise fetch it
|
2017-02-18 04:32:33 +01:00
|
|
|
*
|
2017-12-17 21:27:50 +01:00
|
|
|
* @param string $url The url of the page which should be scraped
|
2019-01-06 22:06:53 +01:00
|
|
|
* @param bool $no_guessing If true the parse doens't search for
|
|
|
|
* preview pictures
|
|
|
|
* @param bool $do_oembed The false option is used by the function fetch_oembed()
|
|
|
|
* to avoid endless loops
|
2017-02-18 04:32:33 +01:00
|
|
|
*
|
2016-11-27 23:41:55 +01:00
|
|
|
* @return array which contains needed data for embedding
|
2020-07-18 01:18:27 +02:00
|
|
|
* string 'url' => The url of the parsed page
|
|
|
|
* string 'type' => Content type
|
|
|
|
* string 'title' => (optional) The title of the content
|
|
|
|
* string 'text' => (optional) The description for the content
|
|
|
|
* string 'image' => (optional) A preview image of the content (only available if $no_geuessing = false)
|
|
|
|
* array 'images' => (optional) Array of preview pictures
|
|
|
|
* string 'keywords' => (optional) The tags which belong to the content
|
2017-02-18 04:32:33 +01:00
|
|
|
*
|
2021-02-16 16:16:04 +01:00
|
|
|
* @throws HTTPException\InternalServerErrorException
|
2019-01-06 22:06:53 +01:00
|
|
|
* @see ParseUrl::getSiteinfo() for more information about scraping
|
2017-02-18 04:32:33 +01:00
|
|
|
* embeddable content
|
2016-11-27 23:41:55 +01:00
|
|
|
*/
|
2021-02-16 16:16:04 +01:00
|
|
|
public static function getSiteinfoCached($url, $no_guessing = false, $do_oembed = true): array
|
2017-11-08 23:02:50 +01:00
|
|
|
{
|
2021-02-16 16:16:04 +01:00
|
|
|
if (empty($url)) {
|
|
|
|
return [
|
|
|
|
'url' => '',
|
|
|
|
'type' => 'error',
|
|
|
|
];
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
|
2021-02-16 16:16:04 +01:00
|
|
|
$urlHash = hash('sha256', $url);
|
|
|
|
|
2018-07-20 14:19:26 +02:00
|
|
|
$parsed_url = DBA::selectFirst('parsed_url', ['content'],
|
2021-02-16 16:16:04 +01:00
|
|
|
['url_hash' => $urlHash, 'guessing' => !$no_guessing, 'oembed' => $do_oembed]
|
2017-11-08 23:02:50 +01:00
|
|
|
);
|
2018-02-14 06:05:00 +01:00
|
|
|
if (!empty($parsed_url['content'])) {
|
|
|
|
$data = unserialize($parsed_url['content']);
|
2016-11-24 01:11:22 +01:00
|
|
|
return $data;
|
|
|
|
}
|
|
|
|
|
|
|
|
$data = self::getSiteinfo($url, $no_guessing, $do_oembed);
|
|
|
|
|
2021-02-16 16:16:04 +01:00
|
|
|
$expires = $data['expires'];
|
|
|
|
|
|
|
|
unset($data['expires']);
|
|
|
|
|
|
|
|
DI::dba()->insert(
|
2017-11-08 23:02:50 +01:00
|
|
|
'parsed_url',
|
2018-01-15 14:05:12 +01:00
|
|
|
[
|
2021-02-16 16:16:04 +01:00
|
|
|
'url_hash' => $urlHash,
|
|
|
|
'guessing' => !$no_guessing,
|
|
|
|
'oembed' => $do_oembed,
|
|
|
|
'url' => $url,
|
|
|
|
'content' => serialize($data),
|
|
|
|
'created' => DateTimeFormat::utcNow(),
|
|
|
|
'expires' => $expires,
|
2018-01-25 03:08:45 +01:00
|
|
|
],
|
2020-11-19 20:34:48 +01:00
|
|
|
Database::INSERT_UPDATE
|
2017-11-08 23:02:50 +01:00
|
|
|
);
|
2016-11-24 01:11:22 +01:00
|
|
|
|
|
|
|
return $data;
|
|
|
|
}
|
2019-01-06 22:06:53 +01:00
|
|
|
|
2016-11-27 23:41:55 +01:00
|
|
|
/**
|
2020-01-19 07:05:23 +01:00
|
|
|
* Parse a page for embeddable content information
|
2017-02-18 04:32:33 +01:00
|
|
|
*
|
2016-11-27 23:41:55 +01:00
|
|
|
* This method parses to url for meta data which can be used to embed
|
|
|
|
* the content. If available it prioritizes Open Graph meta tags.
|
|
|
|
* If this is not available it uses the twitter cards meta tags.
|
|
|
|
* As fallback it uses standard html elements with meta informations
|
|
|
|
* like \<title\>Awesome Title\</title\> or
|
|
|
|
* \<meta name="description" content="An awesome description"\>
|
2017-02-18 04:32:33 +01:00
|
|
|
*
|
2017-12-17 21:27:50 +01:00
|
|
|
* @param string $url The url of the page which should be scraped
|
2019-01-06 22:06:53 +01:00
|
|
|
* @param bool $no_guessing If true the parse doens't search for
|
|
|
|
* preview pictures
|
|
|
|
* @param bool $do_oembed The false option is used by the function fetch_oembed()
|
|
|
|
* to avoid endless loops
|
|
|
|
* @param int $count Internal counter to avoid endless loops
|
2017-02-18 04:32:33 +01:00
|
|
|
*
|
2016-11-27 23:41:55 +01:00
|
|
|
* @return array which contains needed data for embedding
|
2020-07-18 01:18:27 +02:00
|
|
|
* string 'url' => The url of the parsed page
|
2021-02-16 16:16:04 +01:00
|
|
|
* string 'type' => Content type (error, link, photo, image, audio, video)
|
2020-07-18 01:18:27 +02:00
|
|
|
* string 'title' => (optional) The title of the content
|
|
|
|
* string 'text' => (optional) The description for the content
|
|
|
|
* string 'image' => (optional) A preview image of the content (only available if $no_guessing = false)
|
|
|
|
* array 'images' => (optional) Array of preview pictures
|
|
|
|
* string 'keywords' => (optional) The tags which belong to the content
|
2017-02-18 04:32:33 +01:00
|
|
|
*
|
2019-01-06 22:06:53 +01:00
|
|
|
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
|
|
|
|
* @todo https://developers.google.com/+/plugins/snippet/
|
2016-11-27 23:41:55 +01:00
|
|
|
* @verbatim
|
|
|
|
* <meta itemprop="name" content="Awesome title">
|
|
|
|
* <meta itemprop="description" content="An awesome description">
|
|
|
|
* <meta itemprop="image" content="http://maple.libertreeproject.org/images/tree-icon.png">
|
2017-02-18 04:32:33 +01:00
|
|
|
*
|
2016-11-27 23:41:55 +01:00
|
|
|
* <body itemscope itemtype="http://schema.org/Product">
|
|
|
|
* <h1 itemprop="name">Shiny Trinket</h1>
|
|
|
|
* <img itemprop="image" src="{image-url}" />
|
|
|
|
* <p itemprop="description">Shiny trinkets are shiny.</p>
|
|
|
|
* </body>
|
|
|
|
* @endverbatim
|
|
|
|
*/
|
2017-11-08 23:02:50 +01:00
|
|
|
public static function getSiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1)
|
|
|
|
{
|
2021-02-16 16:16:04 +01:00
|
|
|
if (empty($url)) {
|
|
|
|
return [
|
|
|
|
'url' => '',
|
|
|
|
'type' => 'error',
|
|
|
|
];
|
|
|
|
}
|
|
|
|
|
2016-11-24 01:11:22 +01:00
|
|
|
// Check if the URL does contain a scheme
|
|
|
|
$scheme = parse_url($url, PHP_URL_SCHEME);
|
|
|
|
|
2018-09-02 23:35:55 +02:00
|
|
|
if ($scheme == '') {
|
2020-07-18 01:18:27 +02:00
|
|
|
$url = 'http://' . ltrim($url, '/');
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
|
2020-07-18 01:18:27 +02:00
|
|
|
$url = trim($url, "'\"");
|
|
|
|
|
|
|
|
$url = Network::stripTrackingQueryParams($url);
|
|
|
|
|
|
|
|
$siteinfo = [
|
|
|
|
'url' => $url,
|
|
|
|
'type' => 'link',
|
2021-02-16 16:16:04 +01:00
|
|
|
'expires' => DateTimeFormat::utc(self::DEFAULT_EXPIRATION_FAILURE),
|
2020-07-18 01:18:27 +02:00
|
|
|
];
|
|
|
|
|
2016-11-24 01:11:22 +01:00
|
|
|
if ($count > 10) {
|
2018-10-30 14:58:45 +01:00
|
|
|
Logger::log('Endless loop detected for ' . $url, Logger::DEBUG);
|
2018-09-02 23:35:55 +02:00
|
|
|
return $siteinfo;
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
|
2020-10-11 23:25:47 +02:00
|
|
|
$curlResult = DI::httpRequest()->get($url);
|
2018-10-10 21:08:43 +02:00
|
|
|
if (!$curlResult->isSuccess()) {
|
2018-09-02 23:35:55 +02:00
|
|
|
return $siteinfo;
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
|
2021-02-16 16:16:04 +01:00
|
|
|
$siteinfo['expires'] = DateTimeFormat::utc(self::DEFAULT_EXPIRATION_SUCCESS);
|
|
|
|
|
2020-10-11 23:25:47 +02:00
|
|
|
// If the file is too large then exit
|
|
|
|
if (($curlResult->getInfo()['download_content_length'] ?? 0) > 1000000) {
|
|
|
|
return $siteinfo;
|
|
|
|
}
|
|
|
|
|
2021-02-16 16:16:04 +01:00
|
|
|
// Native media type, no need for HTML parsing
|
|
|
|
$type = $curlResult->getHeader('Content-Type');
|
|
|
|
if ($type) {
|
|
|
|
preg_match('#(image|video|audio)/#i', $type, $matches);
|
|
|
|
if ($matches) {
|
|
|
|
$siteinfo['type'] = array_pop($matches);
|
|
|
|
return $siteinfo;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2016-11-24 01:11:22 +01:00
|
|
|
// If it isn't a HTML file then exit
|
2020-10-11 23:25:40 +02:00
|
|
|
if (($curlResult->getContentType() != '') && !strstr(strtolower($curlResult->getContentType()), 'html')) {
|
2018-09-02 23:35:55 +02:00
|
|
|
return $siteinfo;
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
|
2021-02-16 16:16:04 +01:00
|
|
|
if ($cacheControlHeader = $curlResult->getHeader('Cache-Control')) {
|
|
|
|
if (preg_match('/max-age=([0-9]+)/i', $cacheControlHeader, $matches)) {
|
|
|
|
$maxAge = max(86400, (int)array_pop($matches));
|
|
|
|
$siteinfo['expires'] = DateTimeFormat::utc("now + $maxAge seconds");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-10-11 23:26:03 +02:00
|
|
|
$header = $curlResult->getHeader();
|
2018-10-10 21:08:43 +02:00
|
|
|
$body = $curlResult->getBody();
|
2017-10-15 21:29:58 +02:00
|
|
|
|
2016-11-24 01:11:22 +01:00
|
|
|
if ($do_oembed) {
|
2018-01-01 02:58:09 +01:00
|
|
|
$oembed_data = OEmbed::fetchURL($url);
|
2016-11-24 01:11:22 +01:00
|
|
|
|
2018-07-10 14:27:56 +02:00
|
|
|
if (!empty($oembed_data->type)) {
|
2018-09-02 23:35:55 +02:00
|
|
|
if (!in_array($oembed_data->type, ['error', 'rich', ''])) {
|
|
|
|
$siteinfo['type'] = $oembed_data->type;
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
2018-07-10 14:27:56 +02:00
|
|
|
|
2018-09-16 15:18:31 +02:00
|
|
|
// See https://github.com/friendica/friendica/pull/5763#discussion_r217913178
|
|
|
|
if ($siteinfo['type'] != 'photo') {
|
2018-07-10 14:27:56 +02:00
|
|
|
if (isset($oembed_data->title)) {
|
2018-09-02 23:35:55 +02:00
|
|
|
$siteinfo['title'] = trim($oembed_data->title);
|
2018-07-10 14:27:56 +02:00
|
|
|
}
|
|
|
|
if (isset($oembed_data->description)) {
|
2018-09-02 23:35:55 +02:00
|
|
|
$siteinfo['text'] = trim($oembed_data->description);
|
2018-07-10 14:27:56 +02:00
|
|
|
}
|
|
|
|
if (isset($oembed_data->thumbnail_url)) {
|
2018-09-02 23:35:55 +02:00
|
|
|
$siteinfo['image'] = $oembed_data->thumbnail_url;
|
2018-07-10 14:27:56 +02:00
|
|
|
}
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-09-02 23:35:55 +02:00
|
|
|
$charset = '';
|
2020-09-23 02:35:08 +02:00
|
|
|
// Look for a charset, first in headers
|
|
|
|
// Expected form: Content-Type: text/html; charset=ISO-8859-4
|
2020-10-11 23:26:03 +02:00
|
|
|
if (preg_match('/charset=([a-z0-9-_.\/]+)/i', $header, $matches)) {
|
2020-10-11 23:25:40 +02:00
|
|
|
$charset = trim(trim(trim(array_pop($matches)), ';,'));
|
2020-09-23 02:35:08 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
// Then in body that gets precedence
|
|
|
|
// Expected forms:
|
|
|
|
// - <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
|
|
|
|
// - <meta charset="utf-8">
|
2020-09-25 17:36:04 +02:00
|
|
|
// - <meta charset=utf-8>
|
|
|
|
// - <meta charSet="utf-8">
|
2020-09-25 12:55:52 +02:00
|
|
|
// 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) {
|
2020-09-25 17:36:04 +02:00
|
|
|
if (preg_match('/charset=["\']?([a-z0-9-_.\/]+)/i', $body, $matches)) {
|
2020-09-25 12:55:52 +02:00
|
|
|
$charset = trim(trim(trim(array_pop($matches)), ';,'));
|
|
|
|
}
|
|
|
|
});
|
2016-11-24 01:11:22 +01:00
|
|
|
|
2020-09-25 17:36:04 +02:00
|
|
|
$siteinfo['charset'] = $charset;
|
|
|
|
|
2018-09-04 16:52:17 +02:00
|
|
|
if ($charset && strtoupper($charset) != 'UTF-8') {
|
|
|
|
// See https://github.com/friendica/friendica/issues/5470#issuecomment-418351211
|
|
|
|
$charset = str_ireplace('latin-1', 'latin1', $charset);
|
2016-11-24 01:11:22 +01:00
|
|
|
|
2018-10-30 14:58:45 +01:00
|
|
|
Logger::log('detected charset ' . $charset, Logger::DEBUG);
|
2018-09-02 23:35:55 +02:00
|
|
|
$body = iconv($charset, 'UTF-8//TRANSLIT', $body);
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
|
2018-09-02 23:35:55 +02:00
|
|
|
$body = mb_convert_encoding($body, 'HTML-ENTITIES', 'UTF-8');
|
2016-11-24 01:11:22 +01:00
|
|
|
|
2017-05-11 17:53:04 +02:00
|
|
|
$doc = new DOMDocument();
|
2016-11-24 01:11:22 +01:00
|
|
|
@$doc->loadHTML($body);
|
|
|
|
|
2018-09-02 23:35:55 +02:00
|
|
|
XML::deleteNode($doc, 'style');
|
|
|
|
XML::deleteNode($doc, 'script');
|
|
|
|
XML::deleteNode($doc, 'option');
|
|
|
|
XML::deleteNode($doc, 'h1');
|
|
|
|
XML::deleteNode($doc, 'h2');
|
|
|
|
XML::deleteNode($doc, 'h3');
|
|
|
|
XML::deleteNode($doc, 'h4');
|
|
|
|
XML::deleteNode($doc, 'h5');
|
|
|
|
XML::deleteNode($doc, 'h6');
|
|
|
|
XML::deleteNode($doc, 'ol');
|
|
|
|
XML::deleteNode($doc, 'ul');
|
2017-05-11 17:53:04 +02:00
|
|
|
|
2017-12-17 21:24:57 +01:00
|
|
|
$xpath = new DOMXPath($doc);
|
2016-11-24 01:11:22 +01:00
|
|
|
|
2018-09-02 23:35:55 +02:00
|
|
|
$list = $xpath->query('//meta[@content]');
|
2016-11-24 01:11:22 +01:00
|
|
|
foreach ($list as $node) {
|
2018-09-02 23:35:55 +02:00
|
|
|
$meta_tag = [];
|
2016-11-24 01:11:22 +01:00
|
|
|
if ($node->attributes->length) {
|
|
|
|
foreach ($node->attributes as $attribute) {
|
2018-09-02 23:35:55 +02:00
|
|
|
$meta_tag[$attribute->name] = $attribute->value;
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-09-02 23:35:55 +02:00
|
|
|
if (@$meta_tag['http-equiv'] == 'refresh') {
|
|
|
|
$path = $meta_tag['content'];
|
|
|
|
$pathinfo = explode(';', $path);
|
|
|
|
$content = '';
|
2016-11-24 01:11:22 +01:00
|
|
|
foreach ($pathinfo as $value) {
|
2018-09-02 23:35:55 +02:00
|
|
|
if (substr(strtolower($value), 0, 4) == 'url=') {
|
2016-11-24 01:11:22 +01:00
|
|
|
$content = substr($value, 4);
|
|
|
|
}
|
|
|
|
}
|
2018-09-02 23:35:55 +02:00
|
|
|
if ($content != '') {
|
2016-11-24 01:11:22 +01:00
|
|
|
$siteinfo = self::getSiteinfo($content, $no_guessing, $do_oembed, ++$count);
|
2018-09-02 23:35:55 +02:00
|
|
|
return $siteinfo;
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-09-02 23:35:55 +02:00
|
|
|
$list = $xpath->query('//title');
|
2016-11-24 01:11:22 +01:00
|
|
|
if ($list->length > 0) {
|
2018-09-02 23:35:55 +02:00
|
|
|
$siteinfo['title'] = trim($list->item(0)->nodeValue);
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
|
2018-09-02 23:35:55 +02:00
|
|
|
$list = $xpath->query('//meta[@name]');
|
2016-11-24 01:11:22 +01:00
|
|
|
foreach ($list as $node) {
|
2018-09-02 23:35:55 +02:00
|
|
|
$meta_tag = [];
|
2016-11-24 01:11:22 +01:00
|
|
|
if ($node->attributes->length) {
|
|
|
|
foreach ($node->attributes as $attribute) {
|
2018-09-02 23:35:55 +02:00
|
|
|
$meta_tag[$attribute->name] = $attribute->value;
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-09-02 23:50:27 +02:00
|
|
|
if (empty($meta_tag['content'])) {
|
|
|
|
continue;
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
2018-09-02 23:50:27 +02:00
|
|
|
|
|
|
|
$meta_tag['content'] = trim(html_entity_decode($meta_tag['content'], ENT_QUOTES, 'UTF-8'));
|
|
|
|
|
|
|
|
switch (strtolower($meta_tag['name'])) {
|
|
|
|
case 'fulltitle':
|
|
|
|
$siteinfo['title'] = trim($meta_tag['content']);
|
|
|
|
break;
|
|
|
|
case 'description':
|
|
|
|
$siteinfo['text'] = trim($meta_tag['content']);
|
|
|
|
break;
|
|
|
|
case 'thumbnail':
|
|
|
|
$siteinfo['image'] = $meta_tag['content'];
|
|
|
|
break;
|
|
|
|
case 'twitter:image':
|
|
|
|
$siteinfo['image'] = $meta_tag['content'];
|
|
|
|
break;
|
|
|
|
case 'twitter:image:src':
|
|
|
|
$siteinfo['image'] = $meta_tag['content'];
|
|
|
|
break;
|
|
|
|
case 'twitter:card':
|
2018-09-06 22:30:16 +02:00
|
|
|
// Detect photo pages
|
2018-09-06 06:24:34 +02:00
|
|
|
if ($meta_tag['content'] == 'summary_large_image') {
|
|
|
|
$siteinfo['type'] = 'photo';
|
2018-09-02 23:50:27 +02:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
case 'twitter:description':
|
|
|
|
$siteinfo['text'] = trim($meta_tag['content']);
|
|
|
|
break;
|
|
|
|
case 'twitter:title':
|
|
|
|
$siteinfo['title'] = trim($meta_tag['content']);
|
|
|
|
break;
|
|
|
|
case 'dc.title':
|
|
|
|
$siteinfo['title'] = trim($meta_tag['content']);
|
|
|
|
break;
|
|
|
|
case 'dc.description':
|
|
|
|
$siteinfo['text'] = trim($meta_tag['content']);
|
|
|
|
break;
|
|
|
|
case 'keywords':
|
|
|
|
$keywords = explode(',', $meta_tag['content']);
|
|
|
|
break;
|
|
|
|
case 'news_keywords':
|
|
|
|
$keywords = explode(',', $meta_tag['content']);
|
|
|
|
break;
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (isset($keywords)) {
|
2018-09-02 23:35:55 +02:00
|
|
|
$siteinfo['keywords'] = [];
|
2016-11-24 01:11:22 +01:00
|
|
|
foreach ($keywords as $keyword) {
|
2018-09-02 23:35:55 +02:00
|
|
|
if (!in_array(trim($keyword), $siteinfo['keywords'])) {
|
|
|
|
$siteinfo['keywords'][] = trim($keyword);
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-09-02 23:35:55 +02:00
|
|
|
$list = $xpath->query('//meta[@property]');
|
2016-11-24 01:11:22 +01:00
|
|
|
foreach ($list as $node) {
|
2018-09-02 23:35:55 +02:00
|
|
|
$meta_tag = [];
|
2016-11-24 01:11:22 +01:00
|
|
|
if ($node->attributes->length) {
|
|
|
|
foreach ($node->attributes as $attribute) {
|
2018-09-02 23:35:55 +02:00
|
|
|
$meta_tag[$attribute->name] = $attribute->value;
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-09-02 23:35:55 +02:00
|
|
|
if (!empty($meta_tag['content'])) {
|
|
|
|
$meta_tag['content'] = trim(html_entity_decode($meta_tag['content'], ENT_QUOTES, 'UTF-8'));
|
2016-11-24 01:11:22 +01:00
|
|
|
|
2018-09-02 23:35:55 +02:00
|
|
|
switch (strtolower($meta_tag['property'])) {
|
|
|
|
case 'og:image':
|
|
|
|
$siteinfo['image'] = $meta_tag['content'];
|
2016-11-24 01:11:22 +01:00
|
|
|
break;
|
2018-09-02 23:35:55 +02:00
|
|
|
case 'og:title':
|
|
|
|
$siteinfo['title'] = trim($meta_tag['content']);
|
2016-11-24 01:11:22 +01:00
|
|
|
break;
|
2018-09-02 23:35:55 +02:00
|
|
|
case 'og:description':
|
|
|
|
$siteinfo['text'] = trim($meta_tag['content']);
|
2016-11-24 01:11:22 +01:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-09-06 22:30:16 +02:00
|
|
|
// Prevent to have a photo type without an image
|
2018-09-07 08:10:33 +02:00
|
|
|
if ((empty($siteinfo['image']) || !empty($siteinfo['text'])) && ($siteinfo['type'] == 'photo')) {
|
2018-09-06 22:30:16 +02:00
|
|
|
$siteinfo['type'] = 'link';
|
|
|
|
}
|
|
|
|
|
2019-11-15 14:28:42 +01:00
|
|
|
if (!empty($siteinfo['image'])) {
|
2018-09-02 23:35:55 +02:00
|
|
|
$src = self::completeUrl($siteinfo['image'], $url);
|
2016-11-24 01:11:22 +01:00
|
|
|
|
2018-09-02 23:35:55 +02:00
|
|
|
unset($siteinfo['image']);
|
2016-11-24 01:11:22 +01:00
|
|
|
|
2019-10-18 03:26:15 +02:00
|
|
|
$photodata = Images::getInfoFromURLCached($src);
|
2016-11-24 01:11:22 +01:00
|
|
|
|
|
|
|
if (($photodata) && ($photodata[0] > 10) && ($photodata[1] > 10)) {
|
2018-09-02 23:35:55 +02:00
|
|
|
$siteinfo['images'][] = ['src' => $src,
|
|
|
|
'width' => $photodata[0],
|
|
|
|
'height' => $photodata[1]];
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2019-11-15 14:28:42 +01:00
|
|
|
if (!empty($siteinfo['text']) && mb_strlen($siteinfo['text']) > self::MAX_DESC_COUNT) {
|
|
|
|
$siteinfo['text'] = mb_substr($siteinfo['text'], 0, self::MAX_DESC_COUNT) . '…';
|
|
|
|
$pos = mb_strrpos($siteinfo['text'], '.');
|
|
|
|
if ($pos > self::MIN_DESC_COUNT) {
|
|
|
|
$siteinfo['text'] = mb_substr($siteinfo['text'], 0, $pos + 1);
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2019-11-15 14:28:42 +01:00
|
|
|
Logger::info('Siteinfo fetched', ['url' => $url, 'siteinfo' => $siteinfo]);
|
2016-11-24 01:11:22 +01:00
|
|
|
|
2018-12-26 07:06:24 +01:00
|
|
|
Hook::callAll('getsiteinfo', $siteinfo);
|
2016-11-24 01:11:22 +01:00
|
|
|
|
2018-09-02 23:35:55 +02:00
|
|
|
return $siteinfo;
|
2016-11-24 01:11:22 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2020-01-19 07:05:23 +01:00
|
|
|
* Convert tags from CSV to an array
|
2017-02-18 04:32:33 +01:00
|
|
|
*
|
2016-11-24 01:11:22 +01:00
|
|
|
* @param string $string Tags
|
|
|
|
* @return array with formatted Hashtags
|
|
|
|
*/
|
2017-11-08 23:02:50 +01:00
|
|
|
public static function convertTagsToArray($string)
|
|
|
|
{
|
2016-11-24 01:11:22 +01:00
|
|
|
$arr_tags = str_getcsv($string);
|
|
|
|
if (count($arr_tags)) {
|
|
|
|
// add the # sign to every tag
|
2018-01-15 14:05:12 +01:00
|
|
|
array_walk($arr_tags, ["self", "arrAddHashes"]);
|
2016-11-24 01:11:22 +01:00
|
|
|
|
|
|
|
return $arr_tags;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2020-01-19 07:05:23 +01:00
|
|
|
* Add a hasht sign to a string
|
2017-02-18 04:32:33 +01:00
|
|
|
*
|
2020-01-19 07:05:23 +01:00
|
|
|
* This method is used as callback function
|
2017-02-18 04:32:33 +01:00
|
|
|
*
|
2016-11-24 01:11:22 +01:00
|
|
|
* @param string $tag The pure tag name
|
2017-11-08 23:02:50 +01:00
|
|
|
* @param int $k Counter for internal use
|
2017-11-19 23:04:40 +01:00
|
|
|
* @return void
|
2016-11-24 01:11:22 +01:00
|
|
|
*/
|
2017-11-08 23:02:50 +01:00
|
|
|
private static function arrAddHashes(&$tag, $k)
|
|
|
|
{
|
2016-11-24 01:11:22 +01:00
|
|
|
$tag = "#" . $tag;
|
|
|
|
}
|
|
|
|
|
2016-11-27 23:41:55 +01:00
|
|
|
/**
|
2020-01-19 07:05:23 +01:00
|
|
|
* Add a scheme to an url
|
2017-02-18 04:32:33 +01:00
|
|
|
*
|
2016-11-27 23:41:55 +01:00
|
|
|
* The src attribute of some html elements (e.g. images)
|
|
|
|
* can miss the scheme so we need to add the correct
|
|
|
|
* scheme
|
2017-02-18 04:32:33 +01:00
|
|
|
*
|
2017-11-08 23:02:50 +01:00
|
|
|
* @param string $url The url which possibly does have
|
|
|
|
* a missing scheme (a link to an image)
|
2016-11-27 23:41:55 +01:00
|
|
|
* @param string $scheme The url with a correct scheme
|
2017-11-08 23:02:50 +01:00
|
|
|
* (e.g. the url from the webpage which does contain the image)
|
2017-02-18 04:32:33 +01:00
|
|
|
*
|
2016-11-27 23:41:55 +01:00
|
|
|
* @return string The url with a scheme
|
|
|
|
*/
|
2017-11-08 23:02:50 +01:00
|
|
|
private static function completeUrl($url, $scheme)
|
|
|
|
{
|
2016-11-24 01:11:22 +01:00
|
|
|
$urlarr = parse_url($url);
|
|
|
|
|
2016-11-27 23:41:55 +01:00
|
|
|
// If the url does allready have an scheme
|
|
|
|
// we can stop the process here
|
2016-11-24 01:11:22 +01:00
|
|
|
if (isset($urlarr["scheme"])) {
|
|
|
|
return($url);
|
|
|
|
}
|
|
|
|
|
|
|
|
$schemearr = parse_url($scheme);
|
|
|
|
|
|
|
|
$complete = $schemearr["scheme"]."://".$schemearr["host"];
|
|
|
|
|
2018-09-13 23:11:52 +02:00
|
|
|
if (!empty($schemearr["port"])) {
|
2016-11-24 01:11:22 +01:00
|
|
|
$complete .= ":".$schemearr["port"];
|
|
|
|
}
|
|
|
|
|
2018-09-13 23:11:52 +02:00
|
|
|
if (!empty($urlarr["path"])) {
|
|
|
|
if (strpos($urlarr["path"], "/") !== 0) {
|
|
|
|
$complete .= "/";
|
|
|
|
}
|
2016-11-24 01:11:22 +01:00
|
|
|
|
2018-09-13 23:11:52 +02:00
|
|
|
$complete .= $urlarr["path"];
|
|
|
|
}
|
2016-11-24 01:11:22 +01:00
|
|
|
|
2018-09-13 23:11:52 +02:00
|
|
|
if (!empty($urlarr["query"])) {
|
2016-11-24 01:11:22 +01:00
|
|
|
$complete .= "?".$urlarr["query"];
|
|
|
|
}
|
|
|
|
|
2018-09-13 23:11:52 +02:00
|
|
|
if (!empty($urlarr["fragment"])) {
|
2016-11-24 01:11:22 +01:00
|
|
|
$complete .= "#".$urlarr["fragment"];
|
|
|
|
}
|
|
|
|
|
|
|
|
return($complete);
|
|
|
|
}
|
|
|
|
}
|