namespace Friendica\Content;
use DOMDocument;
+use DOMNode;
+use DOMText;
use DOMXPath;
use Exception;
use Friendica\Core\Cache\Enum\Duration;
use Friendica\Database\Database;
use Friendica\Database\DBA;
use Friendica\DI;
-use Friendica\Model\Post;
use Friendica\Network\HTTPClient\Client\HttpClientAccept;
use Friendica\Util\DateTimeFormat;
use Friendica\Util\Network;
+use Friendica\Util\ParseUrl;
use Friendica\Util\Proxy;
use Friendica\Util\Strings;
*/
class OEmbed
{
+ /**
+ * Callback for fetching URL, checking allowance and returning formatted HTML
+ *
+ * @param array $matches
+ * @return string Formatted HTML
+ */
+ public static function replaceCallback(array $matches): string
+ {
+ $embedurl = $matches[1];
+ $j = self::fetchURL($embedurl, !self::isAllowedURL($embedurl));
+ $s = self::formatObject($j);
+
+ return $s;
+ }
+
/**
* Get data from an URL to embed its content.
*
* @param string $embedurl The URL from which the data should be fetched.
+ * @param bool $no_rich_type If set to true rich type content won't be fetched.
+ * @param bool $use_parseurl Use the "ParseUrl" functionality to add additional data
*
* @return \Friendica\Object\OEmbed
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
- public static function fetchURL(string $embedurl): \Friendica\Object\OEmbed
+ public static function fetchURL(string $embedurl, bool $no_rich_type = false, bool $use_parseurl = true): \Friendica\Object\OEmbed
{
$embedurl = trim($embedurl, '\'"');
$href = str_replace(['http://www.youtube.com/', 'http://player.vimeo.com/'],
['https://www.youtube.com/', 'https://player.vimeo.com/'], $href);
$result = DI::httpClient()->fetchFull($href . '&maxwidth=' . $a->getThemeInfoValue('videowidth'));
- if ($result->isSuccess()) {
+ if ($result->getReturnCode() === 200) {
$json_string = $result->getBodyString();
break;
}
$oembed->html = str_replace(['http://www.youtube.com/', 'http://player.vimeo.com/'], ['https://www.youtube.com/', 'https://player.vimeo.com/'], $oembed->html);
}
+ // Improve the OEmbed data with data from OpenGraph, Twitter cards and other sources
+ if ($use_parseurl) {
+ $data = ParseUrl::getSiteinfoCached($embedurl, false);
+
+ if (($oembed->type == 'error') && empty($data['title']) && empty($data['text'])) {
+ return $oembed;
+ }
+
+ if ($no_rich_type || ($oembed->type == 'error')) {
+ $oembed->html = '';
+ $oembed->type = $data['type'];
+
+ if ($oembed->type == 'photo') {
+ if (!empty($data['images'])) {
+ $oembed->url = $data['images'][0]['src'];
+ $oembed->width = $data['images'][0]['width'];
+ $oembed->height = $data['images'][0]['height'];
+ } else {
+ $oembed->type = 'link';
+ }
+ }
+ }
+
+ if (!empty($data['title'])) {
+ $oembed->title = $data['title'];
+ }
+
+ if (!empty($data['text'])) {
+ $oembed->description = $data['text'];
+ }
+
+ if (!empty($data['publisher_name'])) {
+ $oembed->provider_name = $data['publisher_name'];
+ }
+
+ if (!empty($data['publisher_url'])) {
+ $oembed->provider_url = $data['publisher_url'];
+ }
+
+ if (!empty($data['author_name'])) {
+ $oembed->author_name = $data['author_name'];
+ }
+
+ if (!empty($data['author_url'])) {
+ $oembed->author_url = $data['author_url'];
+ }
+
+ if (!empty($data['images']) && ($oembed->type != 'photo')) {
+ $oembed->thumbnail_url = $data['images'][0]['src'];
+ $oembed->thumbnail_width = $data['images'][0]['width'];
+ $oembed->thumbnail_height = $data['images'][0]['height'];
+ }
+ }
+
Hook::callAll('oembed_fetch_url', $embedurl, $oembed);
return $oembed;
* Returns a formatted string from OEmbed object
*
* @param \Friendica\Object\OEmbed $oembed
- * @param int $uriid
* @return string
*/
- private static function formatObject(\Friendica\Object\OEmbed $oembed, int $uriid): string
+ private static function formatObject(\Friendica\Object\OEmbed $oembed): string
{
$ret = '<div class="oembed ' . $oembed->type . '">';
- if (!self::isAllowedURL($oembed->url)) {
- $oembed->html = '';
- if ($oembed->type == 'rich') {
- $oembed->type = 'link';
- }
- }
-
switch ($oembed->type) {
case 'video':
if ($oembed->thumbnail_url) {
'$escapedhtml' => base64_encode($oembed->html),
'$tw' => $tw,
'$th' => $th,
- '$turl' => Post\Link::getByLink($uriid, $oembed->thumbnail_url, Proxy::SIZE_MEDIUM),
+ '$turl' => $oembed->thumbnail_url,
]);
} else {
- $ret .= Proxy::proxifyHtml($oembed->html);
+ $ret = $oembed->html;
}
break;
case 'photo':
- $ret .= '<img width="' . $oembed->width . '" src="' . Post\Link::getByLink($uriid, $oembed->url, Proxy::SIZE_MEDIUM) . '">';
+ $ret .= '<img width="' . $oembed->width . '" src="' . Proxy::proxifyUrl($oembed->url) . '">';
break;
case 'link':
$ret .= '<a href="' . $oembed->embed_url . '" rel="oembed">' . $oembed->embed_url . '</a>';
}
$ret .= "</h4>";
- if ($oembed->type == 'link') {
- if (!empty($oembed->thumbnail_url)) {
- $ret .= '<img width="' . $oembed->width . '" src="' . Post\Link::getByLink($uriid, $oembed->thumbnail_url, Proxy::SIZE_MEDIUM) . '">';
- }
- if (!empty($oembed->description)) {
- $ret .= '<p>' . $oembed->description . '</p>';
- }
- }
} elseif (!strpos($oembed->html, $oembed->embed_url)) {
// add <a> for html2bbcode conversion
$ret .= '<a href="' . $oembed->embed_url . '" rel="oembed">' . $oembed->title . '</a>';
* Converts BBCode to HTML code
*
* @param string $text
- * @param int $uriid
* @return string
*/
- public static function BBCode2HTML(string $text, int $uriid): string
+ public static function BBCode2HTML(string $text): string
{
- if (!preg_match_all("/\[embed\](.+?)\[\/embed\]/is", $text, $matches, PREG_SET_ORDER)) {
- return $text;
+ if (DI::config()->get('system', 'no_oembed')) {
+ return preg_replace("/\[embed\](.+?)\[\/embed\]/is", "<!-- oembed $1 --><i>" . DI::l10n()->t('Embedding disabled') . " : $1</i><!-- /oembed $1 -->", $text);
}
+ return preg_replace_callback("/\[embed\](.+?)\[\/embed\]/is", [self::class, 'replaceCallback'], $text);
+ }
- foreach ($matches as $match) {
- $data = self::fetchURL($match[1]);
- $text = str_replace($match[0], self::formatObject($data, $uriid), $text);
+ /**
+ * Find <span class='oembed'>..<a href='url' rel='oembed'>..</a></span>
+ * and replace it with [embed]url[/embed]
+ *
+ * @param string $text
+ * @return string
+ */
+ public static function HTML2BBCode(string $text): string
+ {
+ // start parser only if 'oembed' is in text
+ if (strpos($text, 'oembed')) {
+ // convert non ascii chars to html entities
+ $html_text = mb_convert_encoding($text, 'HTML-ENTITIES', mb_detect_encoding($text));
+
+ // If it doesn't parse at all, just return the text.
+ $dom = new DOMDocument();
+ if (!@$dom->loadHTML($html_text)) {
+ return $text;
+ }
+ $xpath = new DOMXPath($dom);
+
+ $xattr = self::buildXPath('class', 'oembed');
+ $entries = $xpath->query("//div[$xattr]");
+
+ $xattr = "@rel='oembed'"; //oe_build_xpath("rel","oembed");
+ foreach ($entries as $e) {
+ $href = $xpath->evaluate("a[$xattr]/@href", $e)->item(0)->nodeValue;
+ if (!is_null($href)) {
+ $e->parentNode->replaceChild(new DOMText('[embed]' . $href . '[/embed]'), $e);
+ }
+ }
+ return self::getInnerHTML($dom->getElementsByTagName('body')->item(0));
+ } else {
+ return $text;
}
- return $text;
}
/**
* @return boolean
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
- private static function isAllowedURL(string $url): bool
+ public static function isAllowedURL(string $url): bool
{
if (!DI::config()->get('system', 'no_oembed_rich_content')) {
return true;
* Returns a formatted HTML code from given URL and sets optional title
*
* @param string $url URL to fetch
- * @param string $title title (default: what comes from OEmbed object)
- * @param int $uriid
+ * @param string $title Optional title (default: what comes from OEmbed object)
* @return string Formatted HTML
*/
- public static function getHTML(string $url, string $title, int $uriid): string
+ public static function getHTML(string $url, string $title = ''): string
{
- $o = self::fetchURL($url);
+ $o = self::fetchURL($url, !self::isAllowedURL($url));
if (!is_object($o) || property_exists($o, 'type') && $o->type == 'error') {
throw new Exception('OEmbed failed for URL: ' . $url);
$o->title = $title;
}
- $html = self::formatObject($o, $uriid);
+ $html = self::formatObject($o);
return $html;
}
+
+ /**
+ * Generates the iframe HTML for an oembed attachment.
+ *
+ * Width and height are given by the remote, and are regularly too small for
+ * the generated iframe.
+ *
+ * The width is entirely discarded for the actual width of the post, while fixed
+ * height is used as a starting point before the inevitable resizing.
+ *
+ * Since the iframe is automatically resized on load, there are no need for ugly
+ * and impractical scrollbars.
+ *
+ * @todo This function is currently unused until someone™ adds support for a separate OEmbed domain
+ *
+ * @param string $src Original remote URL to embed
+ * @param string $width
+ * @param string $height
+ * @return string Formatted HTML
+ *
+ * @throws \Friendica\Network\HTTPException\InternalServerErrorException
+ * @see oembed_format_object()
+ */
+ private static function iframe(string $src, string $width, string $height): string
+ {
+ if (!$height || strstr($height, '%')) {
+ $height = '200';
+ }
+ $width = '100%';
+
+ $src = DI::baseUrl() . '/oembed/' . Strings::base64UrlEncode($src);
+ return '<iframe onload="resizeIframe(this);" class="embed_rich" height="' . $height . '" width="' . $width . '" src="' . $src . '" allowfullscreen scrolling="no" frameborder="no">' . DI::l10n()->t('Embedded content') . '</iframe>';
+ }
+
+ /**
+ * Generates attribute search XPath string
+ *
+ * Generates an XPath query to select elements whose provided attribute contains
+ * the provided value in a space-separated list.
+ *
+ * @param string $attr Name of the attribute to search
+ * @param string $value Value to search in a space-separated list
+ * @return string
+ */
+ private static function buildXPath(string $attr, $value): string
+ {
+ // https://www.westhoffswelt.de/blog/2009/6/9/select-html-elements-with-more-than-one-css-class-using-xpath
+ return "contains(normalize-space(@$attr), ' $value ') or substring(normalize-space(@$attr), 1, string-length('$value') + 1) = '$value ' or substring(normalize-space(@$attr), string-length(@$attr) - string-length('$value')) = ' $value' or @$attr = '$value'";
+ }
+
+ /**
+ * Returns the inner XML string of a provided DOMNode
+ *
+ * @param DOMNode $node
+ * @return string
+ */
+ private static function getInnerHTML(DOMNode $node): string
+ {
+ $innerHTML = '';
+ $children = $node->childNodes;
+ foreach ($children as $child) {
+ $innerHTML .= $child->ownerDocument->saveXML($child);
+ }
+ return $innerHTML;
+ }
+
}
--- /dev/null
+<?php
+/**
+ * @copyright Copyright (C) 2010-2024, 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/>.
+ *
+ */
+
+namespace Friendica\Module;
+
+use Friendica\BaseModule;
+use Friendica\Content;
+use Friendica\Core\System;
+use Friendica\DI;
+use Friendica\Util\Strings;
+
+/**
+ * Oembed module
+ *
+ * Displays stored embed content based on a base64 hash of a remote URL
+ *
+ * Example: /oembed/aHR0cHM6Ly9...
+ *
+ * @author Hypolite Petovan <hypolite@mrpetovan.com>
+ */
+class Oembed extends BaseModule
+{
+ protected function content(array $request = []): string
+ {
+ // Unused form: /oembed/b2h?url=...
+ if (DI::args()->getArgv()[1] == 'b2h') {
+ $url = ["", trim(hex2bin($_REQUEST['url']))];
+ echo Content\OEmbed::replaceCallback($url);
+ System::exit();
+ }
+
+ // Unused form: /oembed/h2b?text=...
+ if (DI::args()->getArgv()[1] == 'h2b') {
+ $text = trim(hex2bin($_REQUEST['text']));
+ echo Content\OEmbed::HTML2BBCode($text);
+ System::exit();
+ }
+
+ // @TODO: Replace with parameter from router
+ if (DI::args()->getArgc() == 2) {
+ echo '<html><body>';
+ $url = Strings::base64UrlDecode(DI::args()->getArgv()[1]);
+ $j = Content\OEmbed::fetchURL($url);
+
+ // workaround for media.ccc.de (and any other endpoint that return size 0)
+ if (substr($j->html, 0, 7) == "<iframe" && strstr($j->html, 'width="0"')) {
+ $j->html = '<style>html,body{margin:0;padding:0;} iframe{width:100%;height:100%;}</style>' . $j->html;
+ $j->html = str_replace('width="0"', '', $j->html);
+ $j->html = str_replace('height="0"', '', $j->html);
+ }
+ echo $j->html;
+ echo '</body></html>';
+ }
+ System::exit();
+ }
+}
use DOMDocument;
use DOMXPath;
+use Friendica\Content\OEmbed;
use Friendica\Content\Text\HTML;
use Friendica\Protocol\HTTP\MediaType;
use Friendica\Core\Hook;
* Search for cached embeddable data of an url otherwise fetch it
*
* @param string $url The url of the page which should be scraped
+ * @param bool $do_oembed The false option is used by the function fetch_oembed()
+ * to avoid endless loops
*
* @return array which contains needed data for embedding
* string 'url' => The url of the parsed page
* @see ParseUrl::getSiteinfo() for more information about scraping
* embeddable content
*/
- public static function getSiteinfoCached(string $url): array
+ public static function getSiteinfoCached(string $url, bool $do_oembed = true): array
{
if (empty($url)) {
return [
$urlHash = hash('sha256', $url);
- $parsed_url = DBA::selectFirst('parsed_url', ['content'], ['url_hash' => $urlHash]);
+ $parsed_url = DBA::selectFirst('parsed_url', ['content'],
+ ['url_hash' => $urlHash, 'oembed' => $do_oembed]
+ );
if (!empty($parsed_url['content'])) {
$data = unserialize($parsed_url['content']);
return $data;
}
- $data = self::getSiteinfo($url);
+ $data = self::getSiteinfo($url, $do_oembed);
$expires = $data['expires'];
'parsed_url',
[
'url_hash' => $urlHash,
- 'oembed' => false,
+ 'oembed' => $do_oembed,
'url' => $url,
'content' => serialize($data),
'created' => DateTimeFormat::utcNow(),
* \<meta name="description" content="An awesome description"\>
*
* @param string $url The url of the page which should be scraped
+ * @param bool $do_oembed The false option is used by the function fetch_oembed()
+ * to avoid endless loops
* @param int $count Internal counter to avoid endless loops
*
* @return array which contains needed data for embedding
* </body>
* @endverbatim
*/
- public static function getSiteinfo(string $url, int $count = 1): array
+ public static function getSiteinfo(string $url, bool $do_oembed = true, int $count = 1): array
{
if (empty($url)) {
return [
$body = $curlResult->getBodyString();
+ if ($do_oembed) {
+ $oembed_data = OEmbed::fetchURL($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') {
+ if (!empty($oembed_data->title)) {
+ $siteinfo['title'] = trim($oembed_data->title);
+ }
+ if (!empty($oembed_data->description)) {
+ $siteinfo['text'] = trim($oembed_data->description);
+ }
+ if (!empty($oembed_data->author_name)) {
+ $siteinfo['author_name'] = trim($oembed_data->author_name);
+ }
+ if (!empty($oembed_data->author_url)) {
+ $siteinfo['author_url'] = Network::sanitizeUrl($oembed_data->author_url);
+ }
+ if (!empty($oembed_data->provider_name)) {
+ $siteinfo['publisher_name'] = trim($oembed_data->provider_name);
+ }
+ if (!empty($oembed_data->provider_url)) {
+ $siteinfo['publisher_url'] = Network::sanitizeUrl($oembed_data->provider_url);
+ }
+ if (!empty($oembed_data->thumbnail_url)) {
+ $siteinfo['image'] = $oembed_data->thumbnail_url;
+ }
+ }
+ }
+ }
+
$charset = '';
try {
// Look for a charset, first in headers
}
}
if ($content != '') {
- $siteinfo = self::getSiteinfo($content, ++$count);
+ $siteinfo = self::getSiteinfo($content, $do_oembed, ++$count);
return $siteinfo;
}
}
projects / friendica.git / commitdiff