2016-11-24 00:11:22 +00:00
|
|
|
<?php
|
|
|
|
|
/**
|
2018-01-04 17:03:15 +00:00
|
|
|
* @file src/Util/ParseUrl.php
|
2016-11-24 00:11:22 +00:00
|
|
|
* @brief Get informations about a given URL
|
|
|
|
|
*/
|
2018-01-04 17:03:15 +00:00
|
|
|
namespace Friendica\Util;
|
2016-11-24 00:11:22 +00:00
|
|
|
|
2018-07-20 02:15:21 +00:00
|
|
|
use DOMDocument;
|
|
|
|
|
use DOMXPath;
|
2018-01-01 01:58:09 +00:00
|
|
|
use Friendica\Content\OEmbed;
|
2018-01-17 18:42:40 +00:00
|
|
|
use Friendica\Core\Addon;
|
2018-10-29 21:20:46 +00:00
|
|
|
use Friendica\Core\Logger;
|
2018-07-20 12:19:26 +00:00
|
|
|
use Friendica\Database\DBA;
|
2017-12-07 13:56:11 +00:00
|
|
|
use Friendica\Object\Image;
|
2017-05-11 15:53:04 +00:00
|
|
|
|
2017-12-17 20:24:57 +00:00
|
|
|
require_once 'include/dba.php';
|
2016-11-24 00:11:22 +00:00
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* @brief Class with methods for extracting certain content from an url
|
|
|
|
|
*/
|
2017-11-08 22:02:50 +00:00
|
|
|
class ParseUrl
|
|
|
|
|
{
|
2016-11-27 22:41:55 +00:00
|
|
|
/**
|
|
|
|
|
* @brief 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
|
|
|
|
|
* @param bool $no_guessing If true the parse doens't search for
|
2017-11-08 22:02:50 +00:00
|
|
|
* preview pictures
|
2017-12-17 20:27:50 +00:00
|
|
|
* @param bool $do_oembed The false option is used by the function fetch_oembed()
|
2017-11-08 22:02:50 +00:00
|
|
|
* 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' => The title of the content
|
|
|
|
|
* string 'text' => The description for the content
|
|
|
|
|
* string 'image' => A preview image of the content (only available
|
|
|
|
|
* if $no_geuessing = false
|
|
|
|
|
* array'images' = Array of preview pictures
|
|
|
|
|
* string 'keywords' => The tags which belong to the content
|
2017-02-18 03:32:33 +00:00
|
|
|
*
|
2016-11-27 22:41:55 +00:00
|
|
|
* @see ParseUrl::getSiteinfo() for more information about scraping
|
2017-02-18 03:32:33 +00:00
|
|
|
* embeddable content
|
2016-11-27 22:41:55 +00:00
|
|
|
*/
|
2017-11-08 22:02:50 +00:00
|
|
|
public static function getSiteinfoCached($url, $no_guessing = false, $do_oembed = true)
|
|
|
|
|
{
|
2016-11-24 00:11:22 +00:00
|
|
|
if ($url == "") {
|
|
|
|
|
return false;
|
|
|
|
|
}
|
|
|
|
|
|
2018-07-20 12:19:26 +00:00
|
|
|
$parsed_url = DBA::selectFirst('parsed_url', ['content'],
|
2018-02-14 05:05:00 +00:00
|
|
|
['url' => normalise_link($url), 'guessing' => !$no_guessing, 'oembed' => $do_oembed]
|
2017-11-08 22:02:50 +00:00
|
|
|
);
|
2018-02-14 05:05:00 +00:00
|
|
|
if (!empty($parsed_url['content'])) {
|
|
|
|
|
$data = unserialize($parsed_url['content']);
|
2016-11-24 00:11:22 +00:00
|
|
|
return $data;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
$data = self::getSiteinfo($url, $no_guessing, $do_oembed);
|
|
|
|
|
|
2018-07-20 12:19:26 +00:00
|
|
|
DBA::insert(
|
2017-11-08 22:02:50 +00:00
|
|
|
'parsed_url',
|
2018-01-15 13:05:12 +00:00
|
|
|
[
|
2017-11-08 22:02:50 +00:00
|
|
|
'url' => normalise_link($url), 'guessing' => !$no_guessing,
|
2017-08-09 23:02:57 +00:00
|
|
|
'oembed' => $do_oembed, 'content' => serialize($data),
|
2018-01-27 02:38:34 +00:00
|
|
|
'created' => DateTimeFormat::utcNow()
|
2018-01-25 02:08:45 +00:00
|
|
|
],
|
2017-11-08 22:02:50 +00:00
|
|
|
true
|
|
|
|
|
);
|
2016-11-24 00:11:22 +00:00
|
|
|
|
|
|
|
|
return $data;
|
|
|
|
|
}
|
2016-11-27 22:41:55 +00:00
|
|
|
/**
|
|
|
|
|
* @brief 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
|
|
|
|
|
* @param bool $no_guessing If true the parse doens't search for
|
2017-11-08 22:02:50 +00:00
|
|
|
* preview pictures
|
2017-12-17 20:27:50 +00:00
|
|
|
* @param bool $do_oembed The false option is used by the function fetch_oembed()
|
2017-11-08 22:02:50 +00:00
|
|
|
* to avoid endless loops
|
2017-12-17 20:27:50 +00:00
|
|
|
* @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
|
|
|
|
|
* string 'title' => The title of the content
|
|
|
|
|
* string 'text' => The description for the content
|
|
|
|
|
* string 'image' => A preview image of the content (only available
|
|
|
|
|
* if $no_geuessing = false
|
|
|
|
|
* array'images' = Array of preview pictures
|
|
|
|
|
* string 'keywords' => The tags which belong to the content
|
2017-02-18 03:32:33 +00:00
|
|
|
*
|
2016-11-27 22:41:55 +00:00
|
|
|
* @todo https://developers.google.com/+/plugins/snippet/
|
|
|
|
|
* @verbatim
|
|
|
|
|
* <meta itemprop="name" content="Awesome title">
|
|
|
|
|
* <meta itemprop="description" content="An awesome description">
|
|
|
|
|
* <meta itemprop="image" content="http://maple.libertreeproject.org/images/tree-icon.png">
|
2017-02-18 03:32:33 +00:00
|
|
|
*
|
2016-11-27 22:41:55 +00:00
|
|
|
* <body itemscope itemtype="http://schema.org/Product">
|
|
|
|
|
* <h1 itemprop="name">Shiny Trinket</h1>
|
|
|
|
|
* <img itemprop="image" src="{image-url}" />
|
|
|
|
|
* <p itemprop="description">Shiny trinkets are shiny.</p>
|
|
|
|
|
* </body>
|
|
|
|
|
* @endverbatim
|
|
|
|
|
*/
|
2017-11-08 22:02:50 +00:00
|
|
|
public static function getSiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1)
|
|
|
|
|
{
|
2016-11-24 00:11:22 +00:00
|
|
|
$a = get_app();
|
|
|
|
|
|
2018-01-15 13:05:12 +00:00
|
|
|
$siteinfo = [];
|
2016-11-24 00:11:22 +00:00
|
|
|
|
|
|
|
|
// Check if the URL does contain a scheme
|
|
|
|
|
$scheme = parse_url($url, PHP_URL_SCHEME);
|
|
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
if ($scheme == '') {
|
|
|
|
|
$url = 'http://' . trim($url, '/');
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if ($count > 10) {
|
2018-10-29 21:20:46 +00:00
|
|
|
Logger::log('Endless loop detected for ' . $url, LOGGER_DEBUG);
|
2018-09-02 21:35:55 +00:00
|
|
|
return $siteinfo;
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
$url = trim($url, "'");
|
|
|
|
|
$url = trim($url, '"');
|
|
|
|
|
|
2018-01-27 13:25:54 +00:00
|
|
|
$url = Network::stripTrackingQueryParams($url);
|
2016-11-24 00:11:22 +00:00
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
$siteinfo['url'] = $url;
|
|
|
|
|
$siteinfo['type'] = 'link';
|
2016-11-24 00:11:22 +00:00
|
|
|
|
2018-10-10 19:08:43 +00:00
|
|
|
$curlResult = Network::curl($url);
|
|
|
|
|
if (!$curlResult->isSuccess()) {
|
2018-09-02 21:35:55 +00:00
|
|
|
return $siteinfo;
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
// If the file is too large then exit
|
2018-10-16 22:27:13 +00:00
|
|
|
if (defaults($curlResult->getInfo(), 'download_content_length', 0) > 1000000) {
|
2018-09-02 21:35:55 +00:00
|
|
|
return $siteinfo;
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
// If it isn't a HTML file then exit
|
2018-10-10 19:08:43 +00:00
|
|
|
if (($curlResult->getContentType() != '') && !strstr(strtolower($curlResult->getContentType()), 'html')) {
|
2018-09-02 21:35:55 +00:00
|
|
|
return $siteinfo;
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
|
2018-10-10 19:08:43 +00:00
|
|
|
$header = $curlResult->getHeader();
|
|
|
|
|
$body = $curlResult->getBody();
|
2017-10-15 19:29:58 +00:00
|
|
|
|
2016-11-24 00:11:22 +00:00
|
|
|
if ($do_oembed) {
|
2018-01-01 01:58:09 +00:00
|
|
|
$oembed_data = OEmbed::fetchURL($url);
|
2016-11-24 00:11:22 +00:00
|
|
|
|
2018-07-10 12:27:56 +00:00
|
|
|
if (!empty($oembed_data->type)) {
|
2018-09-02 21:35:55 +00:00
|
|
|
if (!in_array($oembed_data->type, ['error', 'rich', ''])) {
|
|
|
|
|
$siteinfo['type'] = $oembed_data->type;
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
2018-07-10 12:27:56 +00:00
|
|
|
|
2018-09-16 13:18:31 +00:00
|
|
|
// See https://github.com/friendica/friendica/pull/5763#discussion_r217913178
|
|
|
|
|
if ($siteinfo['type'] != 'photo') {
|
2018-07-10 12:27:56 +00:00
|
|
|
if (isset($oembed_data->title)) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$siteinfo['title'] = trim($oembed_data->title);
|
2018-07-10 12:27:56 +00:00
|
|
|
}
|
|
|
|
|
if (isset($oembed_data->description)) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$siteinfo['text'] = trim($oembed_data->description);
|
2018-07-10 12:27:56 +00:00
|
|
|
}
|
|
|
|
|
if (isset($oembed_data->thumbnail_url)) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$siteinfo['image'] = $oembed_data->thumbnail_url;
|
2018-07-10 12:27:56 +00:00
|
|
|
}
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
// Fetch the first mentioned charset. Can be in body or header
|
2018-09-02 21:35:55 +00:00
|
|
|
$charset = '';
|
|
|
|
|
if (preg_match('/charset=(.*?)[\'"\s\n]/', $header, $matches)) {
|
2016-11-24 00:11:22 +00:00
|
|
|
$charset = trim(trim(trim(array_pop($matches)), ';,'));
|
|
|
|
|
}
|
|
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
if ($charset == '') {
|
|
|
|
|
$charset = 'utf-8';
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
if (($charset != '') && (strtoupper($charset) != 'UTF-8')) {
|
2018-10-29 21:20:46 +00:00
|
|
|
Logger::log('detected charset ' . $charset, LOGGER_DEBUG);
|
2018-09-02 21:35:55 +00:00
|
|
|
$body = iconv($charset, 'UTF-8//TRANSLIT', $body);
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
$body = mb_convert_encoding($body, 'HTML-ENTITIES', 'UTF-8');
|
2016-11-24 00:11:22 +00:00
|
|
|
|
2017-05-11 15:53:04 +00:00
|
|
|
$doc = new DOMDocument();
|
2016-11-24 00:11:22 +00:00
|
|
|
@$doc->loadHTML($body);
|
|
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
XML::deleteNode($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 15:53:04 +00:00
|
|
|
|
2017-12-17 20:24:57 +00:00
|
|
|
$xpath = new DOMXPath($doc);
|
2016-11-24 00:11:22 +00:00
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
$list = $xpath->query('//meta[@content]');
|
2016-11-24 00:11:22 +00:00
|
|
|
foreach ($list as $node) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$meta_tag = [];
|
2016-11-24 00:11:22 +00:00
|
|
|
if ($node->attributes->length) {
|
|
|
|
|
foreach ($node->attributes as $attribute) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$meta_tag[$attribute->name] = $attribute->value;
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
if (@$meta_tag['http-equiv'] == 'refresh') {
|
|
|
|
|
$path = $meta_tag['content'];
|
|
|
|
|
$pathinfo = explode(';', $path);
|
|
|
|
|
$content = '';
|
2016-11-24 00:11:22 +00:00
|
|
|
foreach ($pathinfo as $value) {
|
2018-09-02 21:35:55 +00:00
|
|
|
if (substr(strtolower($value), 0, 4) == 'url=') {
|
2016-11-24 00:11:22 +00:00
|
|
|
$content = substr($value, 4);
|
|
|
|
|
}
|
|
|
|
|
}
|
2018-09-02 21:35:55 +00:00
|
|
|
if ($content != '') {
|
2016-11-24 00:11:22 +00:00
|
|
|
$siteinfo = self::getSiteinfo($content, $no_guessing, $do_oembed, ++$count);
|
2018-09-02 21:35:55 +00:00
|
|
|
return $siteinfo;
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
$list = $xpath->query('//title');
|
2016-11-24 00:11:22 +00:00
|
|
|
if ($list->length > 0) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$siteinfo['title'] = trim($list->item(0)->nodeValue);
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
$list = $xpath->query('//meta[@name]');
|
2016-11-24 00:11:22 +00:00
|
|
|
foreach ($list as $node) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$meta_tag = [];
|
2016-11-24 00:11:22 +00:00
|
|
|
if ($node->attributes->length) {
|
|
|
|
|
foreach ($node->attributes as $attribute) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$meta_tag[$attribute->name] = $attribute->value;
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-09-02 21:50:27 +00:00
|
|
|
if (empty($meta_tag['content'])) {
|
|
|
|
|
continue;
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
2018-09-02 21:50:27 +00:00
|
|
|
|
|
|
|
|
$meta_tag['content'] = trim(html_entity_decode($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'] = $meta_tag['content'];
|
|
|
|
|
break;
|
|
|
|
|
case 'twitter:card':
|
2018-09-06 20:30:16 +00:00
|
|
|
// Detect photo pages
|
2018-09-06 04:24:34 +00:00
|
|
|
if ($meta_tag['content'] == 'summary_large_image') {
|
|
|
|
|
$siteinfo['type'] = 'photo';
|
2018-09-02 21:50:27 +00:00
|
|
|
}
|
|
|
|
|
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(',', $meta_tag['content']);
|
|
|
|
|
break;
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (isset($keywords)) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$siteinfo['keywords'] = [];
|
2016-11-24 00:11:22 +00:00
|
|
|
foreach ($keywords as $keyword) {
|
2018-09-02 21:35:55 +00:00
|
|
|
if (!in_array(trim($keyword), $siteinfo['keywords'])) {
|
|
|
|
|
$siteinfo['keywords'][] = trim($keyword);
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
$list = $xpath->query('//meta[@property]');
|
2016-11-24 00:11:22 +00:00
|
|
|
foreach ($list as $node) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$meta_tag = [];
|
2016-11-24 00:11:22 +00:00
|
|
|
if ($node->attributes->length) {
|
|
|
|
|
foreach ($node->attributes as $attribute) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$meta_tag[$attribute->name] = $attribute->value;
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
if (!empty($meta_tag['content'])) {
|
|
|
|
|
$meta_tag['content'] = trim(html_entity_decode($meta_tag['content'], ENT_QUOTES, 'UTF-8'));
|
2016-11-24 00:11:22 +00:00
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
switch (strtolower($meta_tag['property'])) {
|
|
|
|
|
case 'og:image':
|
|
|
|
|
$siteinfo['image'] = $meta_tag['content'];
|
2016-11-24 00:11:22 +00:00
|
|
|
break;
|
2018-09-02 21:35:55 +00:00
|
|
|
case 'og:title':
|
|
|
|
|
$siteinfo['title'] = trim($meta_tag['content']);
|
2016-11-24 00:11:22 +00:00
|
|
|
break;
|
2018-09-02 21:35:55 +00:00
|
|
|
case 'og:description':
|
|
|
|
|
$siteinfo['text'] = trim($meta_tag['content']);
|
2016-11-24 00:11:22 +00:00
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-09-06 20:30:16 +00:00
|
|
|
// Prevent to have a photo type without an image
|
2018-09-07 06:10:33 +00:00
|
|
|
if ((empty($siteinfo['image']) || !empty($siteinfo['text'])) && ($siteinfo['type'] == 'photo')) {
|
2018-09-06 20:30:16 +00:00
|
|
|
$siteinfo['type'] = 'link';
|
|
|
|
|
}
|
|
|
|
|
|
2018-09-14 01:50:52 +00:00
|
|
|
if (empty($siteinfo['image']) && !$no_guessing) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$list = $xpath->query('//img[@src]');
|
2016-11-24 00:11:22 +00:00
|
|
|
foreach ($list as $node) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$img_tag = [];
|
2016-11-24 00:11:22 +00:00
|
|
|
if ($node->attributes->length) {
|
|
|
|
|
foreach ($node->attributes as $attribute) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$img_tag[$attribute->name] = $attribute->value;
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
$src = self::completeUrl($img_tag['src'], $url);
|
2017-12-07 13:56:11 +00:00
|
|
|
$photodata = Image::getInfoFromURL($src);
|
2016-11-24 00:11:22 +00:00
|
|
|
|
|
|
|
|
if (($photodata) && ($photodata[0] > 150) && ($photodata[1] > 150)) {
|
|
|
|
|
if ($photodata[0] > 300) {
|
|
|
|
|
$photodata[1] = round($photodata[1] * (300 / $photodata[0]));
|
|
|
|
|
$photodata[0] = 300;
|
|
|
|
|
}
|
|
|
|
|
if ($photodata[1] > 300) {
|
|
|
|
|
$photodata[0] = round($photodata[0] * (300 / $photodata[1]));
|
|
|
|
|
$photodata[1] = 300;
|
|
|
|
|
}
|
2018-09-02 21:35:55 +00:00
|
|
|
$siteinfo['images'][] = [
|
|
|
|
|
'src' => $src,
|
|
|
|
|
'width' => $photodata[0],
|
|
|
|
|
'height' => $photodata[1]
|
|
|
|
|
];
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
2017-11-08 22:02:50 +00:00
|
|
|
}
|
2018-09-02 21:35:55 +00:00
|
|
|
} elseif (!empty($siteinfo['image'])) {
|
|
|
|
|
$src = self::completeUrl($siteinfo['image'], $url);
|
2016-11-24 00:11:22 +00:00
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
unset($siteinfo['image']);
|
2016-11-24 00:11:22 +00:00
|
|
|
|
2017-12-07 13:56:11 +00:00
|
|
|
$photodata = Image::getInfoFromURL($src);
|
2016-11-24 00:11:22 +00:00
|
|
|
|
|
|
|
|
if (($photodata) && ($photodata[0] > 10) && ($photodata[1] > 10)) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$siteinfo['images'][] = ['src' => $src,
|
|
|
|
|
'width' => $photodata[0],
|
|
|
|
|
'height' => $photodata[1]];
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
if ((@$siteinfo['text'] == '') && (@$siteinfo['title'] != '') && !$no_guessing) {
|
|
|
|
|
$text = '';
|
2016-11-24 00:11:22 +00:00
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
$list = $xpath->query('//div[@class="article"]');
|
2016-11-24 00:11:22 +00:00
|
|
|
foreach ($list as $node) {
|
|
|
|
|
if (strlen($node->nodeValue) > 40) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$text .= ' ' . trim($node->nodeValue);
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
if ($text == '') {
|
|
|
|
|
$list = $xpath->query('//div[@class="content"]');
|
2016-11-24 00:11:22 +00:00
|
|
|
foreach ($list as $node) {
|
|
|
|
|
if (strlen($node->nodeValue) > 40) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$text .= ' ' . trim($node->nodeValue);
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
// If none text was found then take the paragraph content
|
2018-09-02 21:35:55 +00:00
|
|
|
if ($text == '') {
|
|
|
|
|
$list = $xpath->query('//p');
|
2016-11-24 00:11:22 +00:00
|
|
|
foreach ($list as $node) {
|
|
|
|
|
if (strlen($node->nodeValue) > 40) {
|
2018-09-02 21:35:55 +00:00
|
|
|
$text .= ' ' . trim($node->nodeValue);
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
if ($text != '') {
|
|
|
|
|
$text = trim(str_replace(["\n", "\r"], [' ', ' '], $text));
|
2016-11-24 00:11:22 +00:00
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
while (strpos($text, ' ')) {
|
|
|
|
|
$text = trim(str_replace(' ', ' ', $text));
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
$siteinfo['text'] = trim(html_entity_decode(substr($text, 0, 350), ENT_QUOTES, 'UTF-8') . '...');
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-10-29 21:20:46 +00:00
|
|
|
Logger::log('Siteinfo for ' . $url . ' ' . print_r($siteinfo, true), LOGGER_DEBUG);
|
2016-11-24 00:11:22 +00:00
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
Addon::callHooks('getsiteinfo', $siteinfo);
|
2016-11-24 00:11:22 +00:00
|
|
|
|
2018-09-02 21:35:55 +00:00
|
|
|
return $siteinfo;
|
2016-11-24 00:11:22 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* @brief Convert tags from CSV to an array
|
2017-02-18 03:32:33 +00:00
|
|
|
*
|
2016-11-24 00:11:22 +00:00
|
|
|
* @param string $string Tags
|
|
|
|
|
* @return array with formatted Hashtags
|
|
|
|
|
*/
|
2017-11-08 22:02:50 +00:00
|
|
|
public static function convertTagsToArray($string)
|
|
|
|
|
{
|
2016-11-24 00:11:22 +00:00
|
|
|
$arr_tags = str_getcsv($string);
|
|
|
|
|
if (count($arr_tags)) {
|
|
|
|
|
// add the # sign to every tag
|
2018-01-15 13:05:12 +00:00
|
|
|
array_walk($arr_tags, ["self", "arrAddHashes"]);
|
2016-11-24 00:11:22 +00:00
|
|
|
|
|
|
|
|
return $arr_tags;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* @brief Add a hasht sign to a string
|
2017-02-18 03:32:33 +00:00
|
|
|
*
|
2016-11-24 00:11:22 +00:00
|
|
|
* This method is used as callback function
|
2017-02-18 03:32:33 +00:00
|
|
|
*
|
2016-11-24 00:11:22 +00:00
|
|
|
* @param string $tag The pure tag name
|
2017-11-08 22:02:50 +00:00
|
|
|
* @param int $k Counter for internal use
|
2017-11-19 22:04:40 +00:00
|
|
|
* @return void
|
2016-11-24 00:11:22 +00:00
|
|
|
*/
|
2017-11-08 22:02:50 +00:00
|
|
|
private static function arrAddHashes(&$tag, $k)
|
|
|
|
|
{
|
2016-11-24 00:11:22 +00:00
|
|
|
$tag = "#" . $tag;
|
|
|
|
|
}
|
|
|
|
|
|
2016-11-27 22:41:55 +00:00
|
|
|
/**
|
|
|
|
|
* @brief Add a scheme to an url
|
2017-02-18 03:32:33 +00:00
|
|
|
*
|
2016-11-27 22:41:55 +00:00
|
|
|
* 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 03:32:33 +00:00
|
|
|
*
|
2017-11-08 22:02:50 +00:00
|
|
|
* @param string $url The url which possibly does have
|
|
|
|
|
* a missing scheme (a link to an image)
|
2016-11-27 22:41:55 +00:00
|
|
|
* @param string $scheme The url with a correct scheme
|
2017-11-08 22:02:50 +00:00
|
|
|
* (e.g. the url from the webpage which does contain the image)
|
2017-02-18 03:32:33 +00:00
|
|
|
*
|
2016-11-27 22:41:55 +00:00
|
|
|
* @return string The url with a scheme
|
|
|
|
|
*/
|
2017-11-08 22:02:50 +00:00
|
|
|
private static function completeUrl($url, $scheme)
|
|
|
|
|
{
|
2016-11-24 00:11:22 +00:00
|
|
|
$urlarr = parse_url($url);
|
|
|
|
|
|
2016-11-27 22:41:55 +00:00
|
|
|
// If the url does allready have an scheme
|
|
|
|
|
// we can stop the process here
|
2016-11-24 00:11:22 +00:00
|
|
|
if (isset($urlarr["scheme"])) {
|
|
|
|
|
return($url);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
$schemearr = parse_url($scheme);
|
|
|
|
|
|
|
|
|
|
$complete = $schemearr["scheme"]."://".$schemearr["host"];
|
|
|
|
|
|
2018-09-13 21:11:52 +00:00
|
|
|
if (!empty($schemearr["port"])) {
|
2016-11-24 00:11:22 +00:00
|
|
|
$complete .= ":".$schemearr["port"];
|
|
|
|
|
}
|
|
|
|
|
|
2018-09-13 21:11:52 +00:00
|
|
|
if (!empty($urlarr["path"])) {
|
|
|
|
|
if (strpos($urlarr["path"], "/") !== 0) {
|
|
|
|
|
$complete .= "/";
|
|
|
|
|
}
|
2016-11-24 00:11:22 +00:00
|
|
|
|
2018-09-13 21:11:52 +00:00
|
|
|
$complete .= $urlarr["path"];
|
|
|
|
|
}
|
2016-11-24 00:11:22 +00:00
|
|
|
|
2018-09-13 21:11:52 +00:00
|
|
|
if (!empty($urlarr["query"])) {
|
2016-11-24 00:11:22 +00:00
|
|
|
$complete .= "?".$urlarr["query"];
|
|
|
|
|
}
|
|
|
|
|
|
2018-09-13 21:11:52 +00:00
|
|
|
if (!empty($urlarr["fragment"])) {
|
2016-11-24 00:11:22 +00:00
|
|
|
$complete .= "#".$urlarr["fragment"];
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return($complete);
|
|
|
|
|
}
|
|
|
|
|
}
|
