<?php
/**
- * @file src/Util/ParseUrl.php
- * @brief Get informations about a given URL
+ * @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/>.
+ *
*/
+
namespace Friendica\Util;
use DOMDocument;
use DOMXPath;
use Friendica\Content\OEmbed;
-use Friendica\Core\Addon;
+use Friendica\Core\Hook;
use Friendica\Core\Logger;
use Friendica\Database\DBA;
-use Friendica\Object\Image;
-use Friendica\Util\Strings;
-
-require_once 'include/dba.php';
/**
- * @brief Class with methods for extracting certain content from an url
+ * Get information about a given URL
+ *
+ * Class with methods for extracting certain content from an url
*/
class ParseUrl
{
/**
- * @brief Search for chached embeddable data of an url otherwise fetch it
+ * Maximum number of characters for the description
+ */
+ const MAX_DESC_COUNT = 250;
+
+ /**
+ * Minimum number of characters for the description
+ */
+ const MIN_DESC_COUNT = 100;
+
+ /**
+ * Search for chached embeddable data of an url otherwise fetch it
*
* @param string $url The url of the page which should be scraped
- * @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 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
*
* @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
+ * 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 $no_geuessing = false)
+ * array 'images' => (optional) Array of preview pictures
+ * string 'keywords' => (optional) The tags which belong to the content
*
- * @see ParseUrl::getSiteinfo() for more information about scraping
+ * @throws \Friendica\Network\HTTPException\InternalServerErrorException
+ * @see ParseUrl::getSiteinfo() for more information about scraping
* embeddable content
*/
public static function getSiteinfoCached($url, $no_guessing = false, $do_oembed = true)
DBA::insert(
'parsed_url',
[
- 'url' => Strings::normaliseLink($url), 'guessing' => !$no_guessing,
+ 'url' => substr(Strings::normaliseLink($url), 0, 255), 'guessing' => !$no_guessing,
'oembed' => $do_oembed, 'content' => serialize($data),
'created' => DateTimeFormat::utcNow()
],
return $data;
}
+
/**
- * @brief Parse a page for embeddable content information
+ * Parse a page for embeddable content information
*
* This method parses to url for meta data which can be used to embed
* the content. If available it prioritizes Open Graph meta tags.
* \<meta name="description" content="An awesome description"\>
*
* @param string $url The url of the page which should be scraped
- * @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
+ * @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
*
* @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
+ * 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 $no_guessing = false)
+ * array 'images' => (optional) Array of preview pictures
+ * string 'keywords' => (optional) The tags which belong to the content
*
- * @todo https://developers.google.com/+/plugins/snippet/
+ * @throws \Friendica\Network\HTTPException\InternalServerErrorException
+ * @todo https://developers.google.com/+/plugins/snippet/
* @verbatim
* <meta itemprop="name" content="Awesome title">
* <meta itemprop="description" content="An awesome description">
*/
public static function getSiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1)
{
- $a = get_app();
-
- $siteinfo = [];
-
// Check if the URL does contain a scheme
$scheme = parse_url($url, PHP_URL_SCHEME);
if ($scheme == '') {
- $url = 'http://' . trim($url, '/');
+ $url = 'http://' . ltrim($url, '/');
}
+ $url = trim($url, "'\"");
+
+ $url = Network::stripTrackingQueryParams($url);
+
+ $siteinfo = [
+ 'url' => $url,
+ 'type' => 'link',
+ ];
+
if ($count > 10) {
Logger::log('Endless loop detected for ' . $url, Logger::DEBUG);
return $siteinfo;
}
- $url = trim($url, "'");
- $url = trim($url, '"');
-
- $url = Network::stripTrackingQueryParams($url);
-
- $siteinfo['url'] = $url;
- $siteinfo['type'] = 'link';
-
$curlResult = Network::curl($url);
if (!$curlResult->isSuccess()) {
return $siteinfo;
}
// If the file is too large then exit
- if (defaults($curlResult->getInfo(), 'download_content_length', 0) > 1000000) {
+ if (($curlResult->getInfo()['download_content_length'] ?? 0) > 1000000) {
return $siteinfo;
}
$charset = trim(trim(trim(array_pop($matches)), ';,'));
}
- if ($charset == '') {
- $charset = 'utf-8';
- }
+ if ($charset && strtoupper($charset) != 'UTF-8') {
+ // See https://github.com/friendica/friendica/issues/5470#issuecomment-418351211
+ $charset = str_ireplace('latin-1', 'latin1', $charset);
- if (($charset != '') && (strtoupper($charset) != 'UTF-8')) {
Logger::log('detected charset ' . $charset, Logger::DEBUG);
$body = iconv($charset, 'UTF-8//TRANSLIT', $body);
}
$siteinfo['type'] = 'link';
}
- if (empty($siteinfo['image']) && !$no_guessing) {
- $list = $xpath->query('//img[@src]');
- foreach ($list as $node) {
- $img_tag = [];
- if ($node->attributes->length) {
- foreach ($node->attributes as $attribute) {
- $img_tag[$attribute->name] = $attribute->value;
- }
- }
-
- $src = self::completeUrl($img_tag['src'], $url);
- $photodata = Image::getInfoFromURL($src);
-
- 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;
- }
- $siteinfo['images'][] = [
- 'src' => $src,
- 'width' => $photodata[0],
- 'height' => $photodata[1]
- ];
- }
- }
- } elseif (!empty($siteinfo['image'])) {
+ if (!empty($siteinfo['image'])) {
$src = self::completeUrl($siteinfo['image'], $url);
unset($siteinfo['image']);
- $photodata = Image::getInfoFromURL($src);
+ $photodata = Images::getInfoFromURLCached($src);
if (($photodata) && ($photodata[0] > 10) && ($photodata[1] > 10)) {
$siteinfo['images'][] = ['src' => $src,
}
}
- if ((@$siteinfo['text'] == '') && (@$siteinfo['title'] != '') && !$no_guessing) {
- $text = '';
-
- $list = $xpath->query('//div[@class="article"]');
- foreach ($list as $node) {
- if (strlen($node->nodeValue) > 40) {
- $text .= ' ' . trim($node->nodeValue);
- }
- }
-
- if ($text == '') {
- $list = $xpath->query('//div[@class="content"]');
- foreach ($list as $node) {
- if (strlen($node->nodeValue) > 40) {
- $text .= ' ' . trim($node->nodeValue);
- }
- }
- }
-
- // If none text was found then take the paragraph content
- if ($text == '') {
- $list = $xpath->query('//p');
- foreach ($list as $node) {
- if (strlen($node->nodeValue) > 40) {
- $text .= ' ' . trim($node->nodeValue);
- }
- }
- }
-
- if ($text != '') {
- $text = trim(str_replace(["\n", "\r"], [' ', ' '], $text));
-
- while (strpos($text, ' ')) {
- $text = trim(str_replace(' ', ' ', $text));
- }
-
- $siteinfo['text'] = trim(html_entity_decode(substr($text, 0, 350), ENT_QUOTES, 'UTF-8') . '...');
+ 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::log('Siteinfo for ' . $url . ' ' . print_r($siteinfo, true), Logger::DEBUG);
+ Logger::info('Siteinfo fetched', ['url' => $url, 'siteinfo' => $siteinfo]);
- Addon::callHooks('getsiteinfo', $siteinfo);
+ Hook::callAll('getsiteinfo', $siteinfo);
return $siteinfo;
}
/**
- * @brief Convert tags from CSV to an array
+ * Convert tags from CSV to an array
*
* @param string $string Tags
* @return array with formatted Hashtags
}
/**
- * @brief Add a hasht sign to a string
+ * Add a hasht sign to a string
*
- * This method is used as callback function
+ * This method is used as callback function
*
* @param string $tag The pure tag name
* @param int $k Counter for internal use
}
/**
- * @brief Add a scheme to an url
+ * Add a scheme to an url
*
* The src attribute of some html elements (e.g. images)
* can miss the scheme so we need to add the correct
projects / friendica.git / blobdiff