friendica/src/Util/ParseUrl.php

1247 lines
37 KiB
PHP
Raw Normal View History

<?php
/**
2022-01-02 07:27:47 +00:00
* @copyright Copyright (C) 2010-2022, the Friendica project
*
* @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/>.
*
*/
2018-01-04 17:03:15 +00:00
namespace Friendica\Util;
use DOMDocument;
use DOMXPath;
2018-01-01 01:58:09 +00:00
use Friendica\Content\OEmbed;
use Friendica\Core\Hook;
2018-10-29 21:20:46 +00:00
use Friendica\Core\Logger;
2020-11-19 19:34:48 +00:00
use Friendica\Database\Database;
use Friendica\Database\DBA;
use Friendica\DI;
use Friendica\Network\HTTPClient\Client\HttpClientAccept;
use Friendica\Network\HTTPException;
use Friendica\Network\HTTPClient\Client\HttpClientOptions;
2017-05-11 15:53:04 +00:00
/**
* Get information about a given URL
*
2020-01-19 06:05:23 +00:00
* Class with methods for extracting certain content from an url
*/
class ParseUrl
{
const DEFAULT_EXPIRATION_FAILURE = 'now + 1 day';
const DEFAULT_EXPIRATION_SUCCESS = 'now + 3 months';
/**
* Maximum number of characters for the description
*/
const MAX_DESC_COUNT = 250;
/**
* Minimum number of characters for the description
*/
const MIN_DESC_COUNT = 100;
/**
* Fetch the content type of the given url
* @param string $url URL of the page
* @param string $accept content-type to accept
2021-04-26 06:50:12 +00:00
* @return array content type
*/
public static function getContentType(string $url, string $accept = HttpClientAccept::DEFAULT)
{
2022-04-08 18:48:57 +00:00
$curlResult = DI::httpClient()->head($url, [HttpClientOptions::ACCEPT_CONTENT => $accept]);
2022-01-22 15:24:51 +00:00
// Workaround for systems that can't handle a HEAD request
if (!$curlResult->isSuccess() && ($curlResult->getReturnCode() == 405)) {
$curlResult = DI::httpClient()->get($url, $accept, [HttpClientOptions::CONTENT_LENGTH => 1000000]);
2022-01-22 15:24:51 +00:00
}
if (!$curlResult->isSuccess()) {
Logger::debug('Got HTTP Error', ['http error' => $curlResult->getReturnCode(), 'url' => $url]);
2021-03-13 13:17:42 +00:00
return [];
}
2021-08-20 18:05:41 +00:00
$contenttype = $curlResult->getHeader('Content-Type')[0] ?? '';
if (empty($contenttype)) {
return ['application', 'octet-stream'];
}
2021-03-13 13:17:42 +00:00
return explode('/', current(explode(';', $contenttype)));
}
2016-11-27 22:41:55 +00:00
/**
2020-01-19 06:05:23 +00:00
* Search for chached embeddable data of an url otherwise fetch it
2017-02-18 03:32:33 +00:00
*
2017-12-17 20:27:50 +00:00
* @param string $url The url of the page which should be scraped
2019-01-06 21:06:53 +00:00
* @param bool $do_oembed The false option is used by the function fetch_oembed()
* to avoid endless loops
2017-02-18 03:32:33 +00:00
*
2016-11-27 22:41:55 +00:00
* @return array which contains needed data for embedding
* string 'url' => The url of the parsed page
* string 'type' => Content type
* string 'title' => (optional) The title of the content
* string 'text' => (optional) The description for the content
2021-03-16 07:15:20 +00:00
* string 'image' => (optional) A preview image of the content
* array 'images' => (optional) Array of preview pictures
* string 'keywords' => (optional) The tags which belong to the content
2017-02-18 03:32:33 +00:00
*
* @throws HTTPException\InternalServerErrorException
2019-01-06 21:06:53 +00:00
* @see ParseUrl::getSiteinfo() for more information about scraping
2017-02-18 03:32:33 +00:00
* embeddable content
2016-11-27 22:41:55 +00:00
*/
2021-03-16 07:15:20 +00:00
public static function getSiteinfoCached($url, $do_oembed = true): array
{
if (empty($url)) {
return [
'url' => '',
'type' => 'error',
];
}
$urlHash = hash('sha256', $url);
$parsed_url = DBA::selectFirst('parsed_url', ['content'],
2021-03-16 07:15:20 +00:00
['url_hash' => $urlHash, 'oembed' => $do_oembed]
);
if (!empty($parsed_url['content'])) {
$data = unserialize($parsed_url['content']);
return $data;
}
2021-03-16 07:15:20 +00:00
$data = self::getSiteinfo($url, $do_oembed);
$expires = $data['expires'];
unset($data['expires']);
DI::dba()->insert(
'parsed_url',
[
'url_hash' => $urlHash,
'oembed' => $do_oembed,
'url' => $url,
'content' => serialize($data),
'created' => DateTimeFormat::utcNow(),
'expires' => $expires,
],
2020-11-19 19:34:48 +00:00
Database::INSERT_UPDATE
);
return $data;
}
2019-01-06 21:06:53 +00:00
2016-11-27 22:41:55 +00:00
/**
2020-01-19 06:05:23 +00:00
* Parse a page for embeddable content information
2017-02-18 03:32:33 +00:00
*
2016-11-27 22:41:55 +00:00
* 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 03:32:33 +00:00
*
2017-12-17 20:27:50 +00:00
* @param string $url The url of the page which should be scraped
2019-01-06 21:06:53 +00:00
* @param bool $do_oembed The false option is used by the function fetch_oembed()
* to avoid endless loops
* @param int $count Internal counter to avoid endless loops
2017-02-18 03:32:33 +00:00
*
2016-11-27 22:41:55 +00:00
* @return array which contains needed data for embedding
* string 'url' => The url of the parsed page
* string 'type' => Content type (error, link, photo, image, audio, video)
* string 'title' => (optional) The title of the content
* string 'text' => (optional) The description for the content
2021-03-16 07:15:20 +00:00
* string 'image' => (optional) A preview image of the content
* array 'images' => (optional) Array of preview pictures
* string 'keywords' => (optional) The tags which belong to the content
2017-02-18 03:32:33 +00:00
*
2019-01-06 21:06:53 +00:00
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
* @todo https://developers.google.com/+/plugins/snippet/
2016-11-27 22:41:55 +00:00
* @verbatim
* <meta itemprop="name" content="Awesome title">
* <meta itemprop="description" content="An awesome description">
* <meta itemprop="image" content="http://maple.libertreeproject.org/images/tree-icon.png">
2017-02-18 03:32:33 +00:00
*
2016-11-27 22:41:55 +00:00
* <body itemscope itemtype="http://schema.org/Product">
* <h1 itemprop="name">Shiny Trinket</h1>
* <img itemprop="image" src="{image-url}" />
* <p itemprop="description">Shiny trinkets are shiny.</p>
* </body>
* @endverbatim
*/
2021-03-16 07:15:20 +00:00
public static function getSiteinfo($url, $do_oembed = true, $count = 1)
{
if (empty($url)) {
return [
'url' => '',
'type' => 'error',
];
}
// Check if the URL does contain a scheme
$scheme = parse_url($url, PHP_URL_SCHEME);
if ($scheme == '') {
$url = 'http://' . ltrim($url, '/');
}
$url = trim($url, "'\"");
$url = Network::stripTrackingQueryParams($url);
$siteinfo = [
'url' => $url,
'type' => 'link',
'expires' => DateTimeFormat::utc(self::DEFAULT_EXPIRATION_FAILURE),
];
if ($count > 10) {
2021-04-26 06:50:12 +00:00
Logger::notice('Endless loop detected', ['url' => $url]);
return $siteinfo;
}
$type = self::getContentType($url);
2021-03-13 13:17:42 +00:00
Logger::info('Got content-type', ['content-type' => $type, 'url' => $url]);
if (!empty($type) && in_array($type[0], ['image', 'video', 'audio'])) {
$siteinfo['type'] = $type[0];
return $siteinfo;
}
if ((count($type) >= 2) && (($type[0] != 'text') || ($type[1] != 'html'))) {
Logger::info('Unparseable content-type, quitting here, ', ['content-type' => $type, 'url' => $url]);
return $siteinfo;
}
$curlResult = DI::httpClient()->get($url, HttpClientAccept::HTML, [HttpClientOptions::CONTENT_LENGTH => 1000000]);
2021-04-10 05:46:19 +00:00
if (!$curlResult->isSuccess() || empty($curlResult->getBody())) {
2022-01-15 21:38:19 +00:00
Logger::info('Empty body or error when fetching', ['url' => $url, 'success' => $curlResult->isSuccess(), 'code' => $curlResult->getReturnCode()]);
return $siteinfo;
}
$siteinfo['expires'] = DateTimeFormat::utc(self::DEFAULT_EXPIRATION_SUCCESS);
2021-08-20 18:05:41 +00:00
if ($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");
}
}
$body = $curlResult->getBody();
if ($do_oembed) {
2021-03-16 07:04:16 +00:00
$oembed_data = OEmbed::fetchURL($url, false, false);
if (!empty($oembed_data->type)) {
if (!in_array($oembed_data->type, ['error', 'rich', 'image', 'video', 'audio', ''])) {
$siteinfo['type'] = $oembed_data->type;
}
// See https://github.com/friendica/friendica/pull/5763#discussion_r217913178
if ($siteinfo['type'] != 'photo') {
2021-03-16 07:04:16 +00:00
if (!empty($oembed_data->title)) {
$siteinfo['title'] = trim($oembed_data->title);
}
2021-03-16 07:04:16 +00:00
if (!empty($oembed_data->description)) {
$siteinfo['text'] = trim($oembed_data->description);
}
2021-03-16 07:04:16 +00:00
if (!empty($oembed_data->author_name)) {
$siteinfo['author_name'] = trim($oembed_data->author_name);
}
if (!empty($oembed_data->author_url)) {
$siteinfo['author_url'] = trim($oembed_data->author_url);
}
if (!empty($oembed_data->provider_name)) {
$siteinfo['publisher_name'] = trim($oembed_data->provider_name);
}
if (!empty($oembed_data->provider_url)) {
$siteinfo['publisher_url'] = trim($oembed_data->provider_url);
}
if (!empty($oembed_data->thumbnail_url)) {
$siteinfo['image'] = $oembed_data->thumbnail_url;
}
}
}
}
$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)), ';,'));
}
});
$siteinfo['charset'] = $charset;
if ($charset && strtoupper($charset) != 'UTF-8') {
// See https://github.com/friendica/friendica/issues/5470#issuecomment-418351211
$charset = str_ireplace('latin-1', 'latin1', $charset);
2021-04-26 06:50:12 +00:00
Logger::info('detected charset', ['charset' => $charset]);
$body = iconv($charset, 'UTF-8//TRANSLIT', $body);
}
$body = mb_convert_encoding($body, 'HTML-ENTITIES', 'UTF-8');
2017-05-11 15:53:04 +00:00
$doc = new DOMDocument();
@$doc->loadHTML($body);
XML::deleteNode($doc, 'style');
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 15:53:04 +00:00
$xpath = new DOMXPath($doc);
$list = $xpath->query('//meta[@content]');
foreach ($list as $node) {
$meta_tag = [];
if ($node->attributes->length) {
foreach ($node->attributes as $attribute) {
$meta_tag[$attribute->name] = $attribute->value;
}
}
if (@$meta_tag['http-equiv'] == 'refresh') {
$path = $meta_tag['content'];
$pathinfo = explode(';', $path);
$content = '';
foreach ($pathinfo as $value) {
if (substr(strtolower($value), 0, 4) == 'url=') {
$content = substr($value, 4);
}
}
if ($content != '') {
2021-03-16 07:15:20 +00:00
$siteinfo = self::getSiteinfo($content, $do_oembed, ++$count);
return $siteinfo;
}
}
}
$list = $xpath->query('//title');
if ($list->length > 0) {
$siteinfo['title'] = trim($list->item(0)->nodeValue);
}
$list = $xpath->query('//meta[@name]');
foreach ($list as $node) {
$meta_tag = [];
if ($node->attributes->length) {
foreach ($node->attributes as $attribute) {
$meta_tag[$attribute->name] = $attribute->value;
}
}
if (empty($meta_tag['content'])) {
continue;
}
$meta_tag['content'] = trim(html_entity_decode($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:description':
$siteinfo['text'] = trim($meta_tag['content']);
break;
case 'twitter:title':
$siteinfo['title'] = trim($meta_tag['content']);
break;
2021-12-04 19:54:33 +00:00
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;
case 'dc.description':
$siteinfo['text'] = trim($meta_tag['content']);
break;
case 'dc.creator':
2021-03-16 06:37:43 +00:00
$siteinfo['publisher_name'] = trim($meta_tag['content']);
break;
case 'keywords':
$keywords = explode(',', $meta_tag['content']);
break;
case 'news_keywords':
$keywords = explode(',', $meta_tag['content']);
break;
}
}
if (isset($keywords)) {
$siteinfo['keywords'] = [];
foreach ($keywords as $keyword) {
if (!in_array(trim($keyword), $siteinfo['keywords'])) {
$siteinfo['keywords'][] = trim($keyword);
}
}
}
$list = $xpath->query('//meta[@property]');
foreach ($list as $node) {
$meta_tag = [];
if ($node->attributes->length) {
foreach ($node->attributes as $attribute) {
$meta_tag[$attribute->name] = $attribute->value;
}
}
if (!empty($meta_tag['content'])) {
$meta_tag['content'] = trim(html_entity_decode($meta_tag['content'], ENT_QUOTES, 'UTF-8'));
switch (strtolower($meta_tag['property'])) {
case 'og:image':
$siteinfo['image'] = $meta_tag['content'];
break;
case 'og:image:url':
$siteinfo['image'] = $meta_tag['content'];
break;
case 'og:image:secure_url':
$siteinfo['image'] = $meta_tag['content'];
break;
case 'og:title':
$siteinfo['title'] = trim($meta_tag['content']);
break;
case 'og:description':
$siteinfo['text'] = trim($meta_tag['content']);
break;
case 'og:site_name':
2021-03-16 06:37:43 +00:00
$siteinfo['publisher_name'] = trim($meta_tag['content']);
break;
2021-12-08 13:32:20 +00:00
case 'og:locale':
$siteinfo['language'] = trim($meta_tag['content']);
break;
case 'og:type':
$siteinfo['pagetype'] = trim($meta_tag['content']);
2021-12-08 13:32:20 +00:00
break;
case 'twitter:description':
$siteinfo['text'] = trim($meta_tag['content']);
break;
case 'twitter:title':
$siteinfo['title'] = trim($meta_tag['content']);
break;
case 'twitter:image':
$siteinfo['image'] = $meta_tag['content'];
break;
}
}
}
$list = $xpath->query("//script[@type='application/ld+json']");
foreach ($list as $node) {
if (!empty($node->nodeValue)) {
if ($jsonld = json_decode($node->nodeValue, true)) {
2021-03-16 22:57:24 +00:00
$siteinfo = self::parseParts($siteinfo, $jsonld);
}
}
}
2021-12-04 19:54:33 +00:00
if (!empty($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'])) {
$siteinfo['images'] = $siteinfo['images'] ?? [];
array_unshift($siteinfo['images'], ['url' => $siteinfo['image']]);
unset($siteinfo['image']);
}
$siteinfo = self::checkMedia($url, $siteinfo);
if (!empty($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);
}
}
Logger::info('Siteinfo fetched', ['url' => $url, 'siteinfo' => $siteinfo]);
Hook::callAll('getsiteinfo', $siteinfo);
2021-12-04 19:54:33 +00:00
ksort($siteinfo);
return $siteinfo;
}
/**
* Check the attached media elements.
* Fix existing data and add missing data.
*
* @param string $page_url
* @param array $siteinfo
2021-12-08 13:32:20 +00:00
* @return array
*/
2021-12-04 19:54:33 +00:00
private static function checkMedia(string $page_url, array $siteinfo) : array
{
if (!empty($siteinfo['images'])) {
array_walk($siteinfo['images'], function (&$image) use ($page_url) {
// According to the specifications someone could place a picture url into the content field as well.
// But this doesn't seem to happen in the wild, so we don't cover it here.
2021-03-23 20:03:08 +00:00
if (!empty($image['url'])) {
$image['url'] = self::completeUrl($image['url'], $page_url);
$photodata = Images::getInfoFromURLCached($image['url']);
if (!empty($photodata) && ($photodata[0] > 50) && ($photodata[1] > 50)) {
$image['src'] = $image['url'];
$image['width'] = $photodata[0];
$image['height'] = $photodata[1];
$image['contenttype'] = $photodata['mime'];
unset($image['url']);
ksort($image);
} else {
$image = [];
2021-03-23 20:03:08 +00:00
}
} else {
$image = [];
}
});
$siteinfo['images'] = array_values(array_filter($siteinfo['images']));
}
foreach (['audio', 'video'] as $element) {
if (!empty($siteinfo[$element])) {
array_walk($siteinfo[$element], function (&$media) use ($page_url, &$siteinfo) {
$url = '';
$embed = '';
$content = '';
$contenttype = '';
foreach (['embed', 'content', 'url'] as $field) {
if (!empty($media[$field])) {
$media[$field] = self::completeUrl($media[$field], $page_url);
$type = self::getContentType($media[$field]);
2021-03-23 20:03:08 +00:00
if (($type[0] ?? '') == 'text') {
if ($field == 'embed') {
$embed = $media[$field];
} else {
$url = $media[$field];
}
} elseif (!empty($type[0])) {
$content = $media[$field];
$contenttype = implode('/', $type);
}
}
unset($media[$field]);
}
foreach (['image', 'preview'] as $field) {
if (!empty($media[$field])) {
$media[$field] = self::completeUrl($media[$field], $page_url);
}
}
if (!empty($url)) {
$media['url'] = $url;
}
if (!empty($embed)) {
$media['embed'] = $embed;
2021-12-04 19:54:33 +00:00
if (empty($siteinfo['player']['embed'])) {
$siteinfo['player']['embed'] = $embed;
}
}
if (!empty($content)) {
$media['src'] = $content;
}
if (!empty($contenttype)) {
$media['contenttype'] = $contenttype;
}
if (empty($url) && empty($content) && empty($embed)) {
$media = [];
}
ksort($media);
});
$siteinfo[$element] = array_values(array_filter($siteinfo[$element]));
}
if (empty($siteinfo[$element])) {
unset($siteinfo[$element]);
}
}
return $siteinfo;
}
2021-03-16 22:57:24 +00:00
/**
2021-03-17 22:29:12 +00:00
* Convert tags from CSV to an array
*
* @param string $string Tags
* @return array with formatted Hashtags
*/
public static function convertTagsToArray($string)
{
$arr_tags = str_getcsv($string);
if (count($arr_tags)) {
// add the # sign to every tag
array_walk($arr_tags, ["self", "arrAddHashes"]);
return $arr_tags;
}
}
/**
* Add a hasht sign to a string
*
* This method is used as callback function
*
* @param string $tag The pure tag name
* @param int $k Counter for internal use
* @return void
*/
private static function arrAddHashes(&$tag, $k)
{
$tag = "#" . $tag;
}
/**
* Add a scheme to an url
2021-03-16 22:57:24 +00:00
*
2021-03-17 22:29:12 +00:00
* The src attribute of some html elements (e.g. images)
* can miss the scheme so we need to add the correct
* scheme
*
* @param string $url The url which possibly does have
* a missing scheme (a link to an image)
* @param string $scheme The url with a correct scheme
* (e.g. the url from the webpage which does contain the image)
*
* @return string The url with a scheme
*/
private static function completeUrl($url, $scheme)
{
$urlarr = parse_url($url);
// If the url does allready have an scheme
// we can stop the process here
if (isset($urlarr["scheme"])) {
return($url);
}
$schemearr = parse_url($scheme);
$complete = $schemearr["scheme"]."://".$schemearr["host"];
if (!empty($schemearr["port"])) {
$complete .= ":".$schemearr["port"];
}
if (!empty($urlarr["path"])) {
if (strpos($urlarr["path"], "/") !== 0) {
$complete .= "/";
}
$complete .= $urlarr["path"];
}
if (!empty($urlarr["query"])) {
$complete .= "?".$urlarr["query"];
}
if (!empty($urlarr["fragment"])) {
$complete .= "#".$urlarr["fragment"];
}
return($complete);
}
/**
* Parse the Json-Ld parts of a web page
*
* @param array $siteinfo
* @param array $jsonld
2021-03-16 22:57:24 +00:00
* @return array siteinfo
*/
private static function parseParts(array $siteinfo, array $jsonld)
{
if (!empty($jsonld['@graph']) && is_array($jsonld['@graph'])) {
foreach ($jsonld['@graph'] as $part) {
if (!empty($part) && is_array($part)) {
2021-03-23 20:03:08 +00:00
$siteinfo = self::parseParts($siteinfo, $part);
}
2021-03-16 22:57:24 +00:00
}
} elseif (!empty($jsonld['@type'])) {
$siteinfo = self::parseJsonLd($siteinfo, $jsonld);
} elseif (!empty($jsonld)) {
$keys = array_keys($jsonld);
$numeric_keys = true;
foreach ($keys as $key) {
if (!is_int($key)) {
$numeric_keys = false;
}
}
if ($numeric_keys) {
foreach ($jsonld as $part) {
if (!empty($part) && is_array($part)) {
2021-03-23 20:03:08 +00:00
$siteinfo = self::parseParts($siteinfo, $part);
}
}
2021-03-16 22:57:24 +00:00
}
}
array_walk_recursive($siteinfo, function (&$element) {
if (is_string($element)) {
2021-03-28 11:32:26 +00:00
$element = trim(strip_tags(html_entity_decode($element, ENT_COMPAT, 'UTF-8')));
}
});
2021-03-16 22:57:24 +00:00
return $siteinfo;
}
/**
* Improve the siteinfo with information from the provided JSON-LD information
* @see https://jsonld.com/
2021-03-17 22:29:12 +00:00
* @see https://schema.org/
*
* @param array $siteinfo
* @param array $jsonld
* @return array siteinfo
*/
private static function parseJsonLd(array $siteinfo, array $jsonld)
{
$type = JsonLD::fetchElement($jsonld, '@type');
2021-03-16 22:57:24 +00:00
if (empty($type)) {
Logger::info('Empty type', ['url' => $siteinfo['url']]);
return $siteinfo;
}
2021-03-17 17:11:50 +00:00
// Silently ignore some types that aren't processed
2021-03-17 22:29:12 +00:00
if (in_array($type, ['SiteNavigationElement', 'JobPosting', 'CreativeWork', 'MusicAlbum',
2021-03-18 08:04:52 +00:00
'WPHeader', 'WPSideBar', 'WPFooter', 'LegalService', 'MusicRecording',
2021-03-17 17:11:50 +00:00
'ItemList', 'BreadcrumbList', 'Blog', 'Dataset', 'Product'])) {
return $siteinfo;
}
switch ($type) {
case 'Article':
2021-03-17 17:11:50 +00:00
case 'AdvertiserContentArticle':
case 'NewsArticle':
2021-03-17 17:11:50 +00:00
case 'Report':
case 'SatiricalArticle':
2021-03-16 22:57:24 +00:00
case 'ScholarlyArticle':
2021-03-17 17:11:50 +00:00
case 'SocialMediaPosting':
case 'TechArticle':
2021-03-16 22:57:24 +00:00
case 'ReportageNewsArticle':
2021-03-17 07:36:16 +00:00
case 'SocialMediaPosting':
case 'BlogPosting':
2021-03-17 17:11:50 +00:00
case 'LiveBlogPosting':
2021-03-17 07:36:16 +00:00
case 'DiscussionForumPosting':
return self::parseJsonLdArticle($siteinfo, $jsonld);
case 'WebPage':
2021-03-17 17:11:50 +00:00
case 'AboutPage':
case 'CheckoutPage':
2021-03-17 07:36:16 +00:00
case 'CollectionPage':
2021-03-17 17:11:50 +00:00
case 'ContactPage':
case 'FAQPage':
case 'ItemPage':
case 'MedicalWebPage':
case 'ProfilePage':
case 'QAPage':
case 'RealEstateListing':
case 'SearchResultsPage':
2021-04-26 06:50:12 +00:00
case 'MediaGallery':
2021-03-17 07:36:16 +00:00
case 'ImageGallery':
2021-03-17 17:11:50 +00:00
case 'VideoGallery':
2021-03-16 22:57:24 +00:00
case 'RadioEpisode':
case 'Event':
return self::parseJsonLdWebPage($siteinfo, $jsonld);
case 'WebSite':
return self::parseJsonLdWebSite($siteinfo, $jsonld);
case 'Organization':
2021-03-17 17:11:50 +00:00
case 'Airline':
case 'Consortium':
case 'Corporation':
case 'EducationalOrganization':
case 'FundingScheme':
case 'GovernmentOrganization':
case 'LibrarySystem':
2021-03-16 22:57:24 +00:00
case 'LocalBusiness':
2021-03-17 17:11:50 +00:00
case 'MedicalOrganization':
case 'NGO':
case 'NewsMediaOrganization':
case 'Project':
case 'SportsOrganization':
case 'WorkersUnion':
2021-03-17 07:36:16 +00:00
return self::parseJsonLdWebOrganization($siteinfo, $jsonld);
case 'Person':
2021-03-17 17:11:50 +00:00
case 'Patient':
2021-03-17 22:29:12 +00:00
case 'PerformingGroup':
2021-03-17 17:11:50 +00:00
case 'DanceGroup';
case 'MusicGroup':
2021-04-26 06:50:12 +00:00
case 'TheaterGroup':
return self::parseJsonLdWebPerson($siteinfo, $jsonld);
2021-03-16 22:57:24 +00:00
case 'AudioObject':
2021-03-17 07:36:16 +00:00
case 'Audio':
return self::parseJsonLdMediaObject($siteinfo, $jsonld, 'audio');
case 'VideoObject':
2021-03-17 07:36:16 +00:00
return self::parseJsonLdMediaObject($siteinfo, $jsonld, 'video');
case 'ImageObject':
2021-03-17 07:36:16 +00:00
return self::parseJsonLdMediaObject($siteinfo, $jsonld, 'images');
default:
2021-03-17 17:11:50 +00:00
Logger::info('Unknown type', ['type' => $type, 'url' => $siteinfo['url']]);
return $siteinfo;
}
}
/**
2021-03-17 22:29:12 +00:00
* Fetch author and publisher data
*
* @param array $siteinfo
* @param array $jsonld
* @return array siteinfo
*/
private static function parseJsonLdAuthor(array $siteinfo, array $jsonld)
{
$jsonldinfo = [];
2021-03-16 06:37:43 +00:00
if (!empty($jsonld['publisher']) && is_array($jsonld['publisher'])) {
2021-03-16 22:57:24 +00:00
$content = JsonLD::fetchElement($jsonld, 'publisher', 'name');
if (!empty($content) && is_string($content)) {
$jsonldinfo['publisher_name'] = trim($content);
}
2021-03-16 22:57:24 +00:00
$content = JsonLD::fetchElement($jsonld, 'publisher', 'url');
if (!empty($content) && is_string($content)) {
$jsonldinfo['publisher_url'] = trim($content);
}
$brand = JsonLD::fetchElement($jsonld, 'publisher', 'brand', '@type', 'Organization');
if (!empty($brand) && is_array($brand)) {
2021-03-17 23:31:16 +00:00
$content = JsonLD::fetchElement($brand, 'name');
if (!empty($content) && is_string($content)) {
2021-03-16 06:37:43 +00:00
$jsonldinfo['publisher_name'] = trim($content);
}
2021-03-17 23:31:16 +00:00
$content = JsonLD::fetchElement($brand, 'url');
2021-03-17 17:11:50 +00:00
if (!empty($content) && is_string($content)) {
$jsonldinfo['publisher_url'] = trim($content);
}
2021-03-17 23:31:16 +00:00
$content = JsonLD::fetchElement($brand, 'logo', 'url');
if (!empty($content) && is_string($content)) {
$jsonldinfo['publisher_img'] = trim($content);
}
}
$logo = JsonLD::fetchElement($jsonld, 'publisher', 'logo');
if (!empty($logo) && is_array($logo)) {
$content = JsonLD::fetchElement($logo, 'url');
if (!empty($content) && is_string($content)) {
$jsonldinfo['publisher_img'] = trim($content);
}
}
2021-03-16 22:57:24 +00:00
} elseif (!empty($jsonld['publisher']) && is_string($jsonld['publisher'])) {
$jsonldinfo['publisher_name'] = trim($jsonld['publisher']);
}
2021-03-16 06:37:43 +00:00
if (!empty($jsonld['author']) && is_array($jsonld['author'])) {
2021-03-16 22:57:24 +00:00
$content = JsonLD::fetchElement($jsonld, 'author', 'name');
if (!empty($content) && is_string($content)) {
$jsonldinfo['author_name'] = trim($content);
}
2021-03-16 22:57:24 +00:00
$content = JsonLD::fetchElement($jsonld, 'author', 'sameAs');
if (!empty($content) && is_string($content)) {
$jsonldinfo['author_url'] = trim($content);
}
2021-03-16 22:57:24 +00:00
$content = JsonLD::fetchElement($jsonld, 'author', 'url');
if (!empty($content) && is_string($content)) {
$jsonldinfo['author_url'] = trim($content);
}
2021-03-17 23:31:16 +00:00
$logo = JsonLD::fetchElement($jsonld, 'author', 'logo');
if (!empty($logo) && is_array($logo)) {
$content = JsonLD::fetchElement($logo, 'url');
if (!empty($content) && is_string($content)) {
$jsonldinfo['author_img'] = trim($content);
}
}
2021-03-16 22:57:24 +00:00
} elseif (!empty($jsonld['author']) && is_string($jsonld['author'])) {
$jsonldinfo['author_name'] = trim($jsonld['author']);
}
2021-03-17 22:29:12 +00:00
Logger::info('Fetched Author information', ['fetched' => $jsonldinfo]);
return array_merge($siteinfo, $jsonldinfo);
}
/**
2021-03-17 22:29:12 +00:00
* Fetch data from the provided JSON-LD Article type
2021-03-17 07:36:16 +00:00
* @see https://schema.org/Article
*
* @param array $siteinfo
* @param array $jsonld
* @return array siteinfo
*/
private static function parseJsonLdArticle(array $siteinfo, array $jsonld)
{
$jsonldinfo = [];
$content = JsonLD::fetchElement($jsonld, 'headline');
if (!empty($content) && is_string($content)) {
$jsonldinfo['title'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'alternativeHeadline');
2021-03-16 22:57:24 +00:00
if (!empty($content) && is_string($content) && (($jsonldinfo['title'] ?? '') != trim($content))) {
$jsonldinfo['alternative_title'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'description');
if (!empty($content) && is_string($content)) {
$jsonldinfo['text'] = trim($content);
}
2021-03-16 22:57:24 +00:00
$content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
if (!empty($content)) {
$jsonldinfo['image'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'image', 'url', '@type', 'ImageObject');
if (!empty($content) && is_string($content)) {
$jsonldinfo['image'] = trim($content);
}
2021-03-16 22:57:24 +00:00
if (!empty($jsonld['keywords']) && !is_array($jsonld['keywords'])) {
$content = JsonLD::fetchElement($jsonld, 'keywords');
if (!empty($content)) {
$siteinfo['keywords'] = [];
$keywords = explode(',', $content);
foreach ($keywords as $keyword) {
$siteinfo['keywords'][] = trim($keyword);
}
}
} elseif (!empty($jsonld['keywords'])) {
2021-03-16 22:57:24 +00:00
$content = JsonLD::fetchElementArray($jsonld, 'keywords');
if (!empty($content) && is_array($content)) {
$jsonldinfo['keywords'] = $content;
}
}
2021-12-08 13:32:20 +00:00
$content = JsonLD::fetchElement($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' => $jsonldinfo]);
return array_merge($siteinfo, $jsonldinfo);
}
/**
2021-03-17 22:29:12 +00:00
* Fetch data from the provided JSON-LD WebPage type
2021-03-17 07:36:16 +00:00
* @see https://schema.org/WebPage
*
* @param array $siteinfo
* @param array $jsonld
* @return array siteinfo
*/
private static function parseJsonLdWebPage(array $siteinfo, array $jsonld)
{
$jsonldinfo = [];
$content = JsonLD::fetchElement($jsonld, 'name');
if (!empty($content)) {
$jsonldinfo['title'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'description');
if (!empty($content)) {
$jsonldinfo['text'] = trim($content);
}
2021-03-16 22:57:24 +00:00
$content = JsonLD::fetchElement($jsonld, 'image');
if (!empty($content)) {
$jsonldinfo['image'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
if (!empty($content)) {
$jsonldinfo['image'] = trim($content);
}
$jsonldinfo = self::parseJsonLdAuthor($jsonldinfo, $jsonld);
2021-03-17 22:29:12 +00:00
Logger::info('Fetched WebPage information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
return array_merge($siteinfo, $jsonldinfo);
}
/**
2021-03-17 22:29:12 +00:00
* Fetch data from the provided JSON-LD WebSite type
2021-03-17 07:36:16 +00:00
* @see https://schema.org/WebSite
*
* @param array $siteinfo
* @param array $jsonld
* @return array siteinfo
*/
private static function parseJsonLdWebSite(array $siteinfo, array $jsonld)
{
$jsonldinfo = [];
$content = JsonLD::fetchElement($jsonld, 'name');
if (!empty($content)) {
2021-03-16 06:37:43 +00:00
$jsonldinfo['publisher_name'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'description');
if (!empty($content)) {
$jsonldinfo['publisher_description'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'url');
if (!empty($content)) {
$jsonldinfo['publisher_url'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
if (!empty($content)) {
$jsonldinfo['image'] = trim($content);
}
$jsonldinfo = self::parseJsonLdAuthor($jsonldinfo, $jsonld);
Logger::info('Fetched WebSite information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
return array_merge($siteinfo, $jsonldinfo);
}
/**
2021-03-17 22:29:12 +00:00
* Fetch data from the provided JSON-LD Organization type
2021-03-17 07:36:16 +00:00
* @see https://schema.org/Organization
*
* @param array $siteinfo
* @param array $jsonld
* @return array siteinfo
*/
private static function parseJsonLdWebOrganization(array $siteinfo, array $jsonld)
{
$jsonldinfo = [];
$content = JsonLD::fetchElement($jsonld, 'name');
if (!empty($content)) {
2021-03-16 06:37:43 +00:00
$jsonldinfo['publisher_name'] = trim($content);
}
2021-03-16 22:57:24 +00:00
$content = JsonLD::fetchElement($jsonld, 'description');
if (!empty($content)) {
$jsonldinfo['publisher_description'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'url');
if (!empty($content)) {
$jsonldinfo['publisher_url'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'logo', 'url', '@type', 'ImageObject');
if (!empty($content)) {
2021-03-16 06:37:43 +00:00
$jsonldinfo['publisher_img'] = trim($content);
}
2021-03-17 17:11:50 +00:00
$content = JsonLD::fetchElement($jsonld, 'brand', 'name', '@type', 'Organization');
if (!empty($content)) {
$jsonldinfo['publisher_name'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'brand', 'url', '@type', 'Organization');
if (!empty($content)) {
$jsonldinfo['publisher_url'] = trim($content);
}
Logger::info('Fetched Organization information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
return array_merge($siteinfo, $jsonldinfo);
}
/**
2021-03-17 22:29:12 +00:00
* Fetch data from the provided JSON-LD Person type
2021-03-17 07:36:16 +00:00
* @see https://schema.org/Person
*
* @param array $siteinfo
* @param array $jsonld
* @return array siteinfo
*/
private static function parseJsonLdWebPerson(array $siteinfo, array $jsonld)
{
$jsonldinfo = [];
$content = JsonLD::fetchElement($jsonld, 'name');
if (!empty($content)) {
2021-03-16 06:37:43 +00:00
$jsonldinfo['author_name'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'description');
if (!empty($content)) {
$jsonldinfo['author_description'] = trim($content);
}
2021-03-17 17:11:50 +00:00
$content = JsonLD::fetchElement($jsonld, 'sameAs');
if (!empty($content) && is_string($content)) {
$jsonldinfo['author_url'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'url');
if (!empty($content)) {
$jsonldinfo['author_url'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'image', 'url', '@type', 'ImageObject');
2021-09-22 03:25:01 +00:00
if (!empty($content) && !is_string($content)) {
Logger::notice('Unexpected return value for the author image', ['content' => $content]);
}
2021-09-22 02:41:24 +00:00
if (!empty($content) && is_string($content)) {
2021-03-16 06:37:43 +00:00
$jsonldinfo['author_img'] = trim($content);
}
Logger::info('Fetched Person information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
return array_merge($siteinfo, $jsonldinfo);
}
2021-03-16 22:57:24 +00:00
/**
2021-03-17 22:29:12 +00:00
* Fetch data from the provided JSON-LD MediaObject type
2021-03-17 07:36:16 +00:00
* @see https://schema.org/MediaObject
2021-03-16 22:57:24 +00:00
*
* @param array $siteinfo
* @param array $jsonld
* @return array siteinfo
*/
2021-03-17 07:36:16 +00:00
private static function parseJsonLdMediaObject(array $siteinfo, array $jsonld, string $name)
2021-03-16 22:57:24 +00:00
{
2021-03-17 07:36:16 +00:00
$media = [];
$content = JsonLD::fetchElement($jsonld, 'caption');
if (!empty($content)) {
$media['caption'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'url');
if (!empty($content)) {
$media['url'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'mainEntityOfPage');
if (!empty($content)) {
$media['main'] = Strings::compareLink($content, $siteinfo['url']);
2021-03-17 07:36:16 +00:00
}
2021-03-16 22:57:24 +00:00
$content = JsonLD::fetchElement($jsonld, 'description');
if (!empty($content)) {
2021-03-17 07:36:16 +00:00
$media['description'] = trim($content);
2021-03-16 22:57:24 +00:00
}
$content = JsonLD::fetchElement($jsonld, 'name');
2021-03-17 07:36:16 +00:00
if (!empty($content) && (($media['description'] ?? '') != trim($content))) {
$media['name'] = trim($content);
2021-03-16 22:57:24 +00:00
}
$content = JsonLD::fetchElement($jsonld, 'contentUrl');
if (!empty($content)) {
2021-03-17 07:36:16 +00:00
$media['content'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'embedUrl');
if (!empty($content)) {
$media['embed'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'height');
if (!empty($content)) {
$media['height'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'width');
if (!empty($content)) {
$media['width'] = trim($content);
2021-03-16 22:57:24 +00:00
}
2021-03-17 07:36:16 +00:00
$content = JsonLD::fetchElement($jsonld, 'image');
if (!empty($content)) {
$media['image'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
if (!empty($content) && (($media['image'] ?? '') != trim($content))) {
if (!empty($media['image'])) {
$media['preview'] = trim($content);
} else {
$media['image'] = trim($content);
}
}
2021-03-17 07:36:16 +00:00
Logger::info('Fetched Media information', ['url' => $siteinfo['url'], 'fetched' => $media]);
$siteinfo[$name][] = $media;
return $siteinfo;
2021-03-16 22:57:24 +00:00
}
}